Between Implementation and Legislation: The Shiʿi Imam Muḥammad al-Jawād’s Khums Demand Letter of 220 ah/835 ce


The Imami Shiʿa are usually treated as a community defined by belief. By analysing a letter attributed to the ninth Imami Imam, Muḥammad al-Jawād dated to the year of his death in 220/835, I show that the Imami Shiʿa were defined also by institutional structures that tied them to their Imam in his capacity as community leader. Details of transmission, form and content suggest that the letter may well be authentic, giving us a unique window onto the Imamic administration. The letter is a tax demand, encouraging payment of the khums levy upon the spoils of war and other items. My analysis suggests that the understanding of khums and ghanīma among Imamis at this time continued to be fluid, subject to the Imam’s adjustment, and that implementation influenced the elaboration of the law. Subsequently, hadith scholars and jurists were thus forced to interpret how such ad hoc, pragmatic acts fit into Islamic law, which is conceived as eternal and divine.

legitimate object of study, a key aspect of an Imam's role was that of community leader, part of which centred on interpreting and implementing God's law. But how did this work in practice? The Imami Imams were not state actors with the state's apparatus of courts and coercive power at their disposal. To what extent, then, can we understand the Imams to have presided over a functional system of law, in addition to being the articulators of an Imami legal framework on an abstract level? The key to answering these questions is to see that they were embedded in a set of social institutions that defined their role as Imams. It is, then, with these social institutions that we must start if we want to understand the Imams' historical efficacy as implementers of law. In this essay, I will introduce a uniquely useful, yet hitherto neglected set of sources for understanding the practical authority of the Imams and their relations with their followers: the letters attributed to the Imams. These letters are preserved in later hadith compilations and biographical works.2 They contain much that is obscure and ad hoc, including certain elements that contradict later orthodoxy. Their obscurities make them difficult to interpret, but also give us some confidence that they may, indeed, represent authentic, datable historical communications between the Imams and their followers. 3 In this article I will introduce and discuss one important letter attributed to the ninth Imam of the Imami Shiʿa, Muḥammad b. ʿAlī, (d. 220/835) who was known as al-Jawād, "the Generous", or less frequently, "the Devout" (al-taqī), though he is most often referred to in Imami hadith as Abū Jaʿfar. This letter of Imam Jawād is a demand for the levy tolled on the spoils of war (ghanīma) following a battle between some of his followers and a group of Khurramiyya in the Jibāl. Although this letter has not gone totally unnoticed, it has received 2 Many of these letters have been collected in Makātīb al-aʾimma, 7 vols., edited by ʿAlī Aḥmadī Miyānjī (Qumm: Dār al-Ḥadīth, 1387/1429), but the work does not provide contextual analysis. The only other sustained study of Imamic letters I am aware of is Edmund Hayes, "The Epistolary Imamate: Circular Letters in the Administration of the Shiʿi Community," in The Ties that Bind: Mechanisms of Social Dependency in the Early Islamic Empire, edited by Petra Sijpesteijn and Edmund Hayes, forthcoming. This article includes analysis of a letter demanding the payment of ṣadaqa by the Imam al-Ḥasan al-ʿAskarī that is comparable to the khums letter discussed here. 3 The principle of valuing difficult texts as likely to be early is well established in textual criticism as "lectio difficilior potior", that is, the most difficult reading is the strongest. If used cautiously this principle can be applied to establishing the probable historicity of a text embedded in a textual canon that may not value its specific content. In Islamic history, this approach has most famously been applied to the so-called "Constitution of scant attention and its importance has not been explored.4 In this article I will argue that this letter has a high likelihood of preserving a historical correspondence between the Imam and his followers. My optimism about its authenticity is based on three considerations: first, its complex structure makes it unlikely that it was fabricated to fill a particular ideological need; second, its content is peculiar and represents a divergence from other Imami statements on khums,5 before and after; and third, the historical positioning of its purported transmitter is plausible. My optimism is not extended to the Imami hadith corpus generally. Imamic hadith may, in many cases, represent the positions of the Imams, but we cannot assume that they do so without investigation. Beyond 4 Jassim Hussain mentions the letter, coming to the conclusion that "the Imam's followers in Khurasan allowed themselves to be recruited into the ʿAbbasid army and participated in subduing the rebellion of the Khurramiyya". The Occultation of the Twelfth Imam: a Historical Background (London: Muhammadi Trust, 1982), 47. The idea that the Imam's followers fought in the army, is not, however supported by the letter, but presumably derives from a presentist assumption that the ʿAbbasids had a monopoly on violence throughout their domains.
Modarressi mentions the letter, saying "One of the last acts of the ninth imam, Muhammad ibn ʿAlī al-Jawad, in the year of his premature death in 835, was to order the regular collection of khums." Hossein Modarressi, Crisis and Consolidation in the Formative Period of Shi'ite Islam (Princeton: Darwin Press, 1993), 12. However, Modarressi does not explain why he sees this as an order to begin regular collection. As we will see, the contents of the letter do not provide evidence for the initiation of regular khums collection. However, Modarressi's comment was sufficient to lead Arjomand to assert, "The systematic collection of the levy as a mandatory tax seems to have started in 220/ 835 when Imām Muhammad al-Jawād ordered his financial representatives to collect the khums on certain kinds of income." Said Arjomand, "Crisis of the Imamate and the Institution of Occultation in Twelver Shi'ism: A Sociohistorical Perspective," International Journal of Middle East Studies 28: 4 (1996): 491-515, at 498. In this article I attempt to move beyond impressionistic assertions to place the study of Imamic institutions on more stable foundations. Qalamdārān discusses the letter over several pages, but his analysis is based on the assumption that Imamic statements should reflect the unchanging normative force of the Twelver law, and he therefore rejects this letter as deviating from the norm. Ḥaydar ʿAlī Qalamdārān, Baḥth ʿamīq fī masʾalat al-khums fī al-kitāb wa-l-sunna wa fatwā ʿulamāʾ al-shīʿa, translated from the Persian by Saʿd Rustam (Riyāḍ: Sharikat al-ʿAbikān li-l-taʿlīm, 1435), 216-30. 5 In this article I do not provide an overview of early Shiʿi attitudes to khums, although I hope that my contribution here will help to establish a clear historical foothold that might facilitate such a study in the future. Currently, as with almost all areas of pre-Occultation Shiʿi legal thinking, our understanding of khums is inadequate. The major historical study of khums in Twelver Shiʿism is Norman Calder, "Khums in Imāmī Shīʿī Jurisprudence, from the Tenth to the Sixteenth Century A.D.," Bulletin of the School of Oriental and African Studies 45:1 (1982): 39-47. However, Calder's article focuses exclusively on the post-Occultation era, leaving developments during the lives of the Imams untouched. Modarressi's Crisis and Consolidation condenses great insights into a few pages, but often leaves assertions unexplained (see comments throughout this article). Qalamdārān's Baḥth ʿamīq contains some interesting insights, though it is highly polemical.
hayes Islamic Law and Society 28 (2021) 382-414 the traditional Twelver hadith sciences, there have been very few systematic efforts to date and authenticate Imamic hadith in the academy.6 However, implicitly, this question is at the heart of some the great controversies raging between scholars of early Shiʿism, who are divided over exactly what the Imams stood for.7 In the absence of convincing evidence, it is generally safer to assume that hadiths attributed to the Imams should be understood as representing the ideas and institutions of the broader community rather than the Imams themselves. A corollary of this is that we usually cannot be sure of the dating of a report. Therefore, if this letter can be shown to be authentic, it is of great significance for the history of the Imamate, establishing a beachhead of historicity for understanding the kinds of interactions that occurred between the Imams and their followers. In addition to its significance for the understanding of the Shiʿi Imamate, the letter is of great interest for our understanding of how non-governmental institutions operating in the interstices of empire regulated themselves. The letter makes possible a number of different comparisons that would otherwise be difficult. It bears witness to non-governmental administrative practices that are, however, comparable to governmental bureaucracy visible from studies of contemporary administrative letters, both those preserved in literary sources and physical letters discovered on papyrus or parchment.8 I propose that we study this text not only as a "hadith" but also as a letter, composed according to the rules and conventions of epistolary form and the institutions of correspondence of this era. This will allow us to see how such letters served as tools 6 On the state of the field, see Kohlberg, "Introduction: Shiʿi Ḥadīth" in Scholarship on administrative procedure in this period is usually restricted to governmental procedure (whether based on the study of physical documents or literary texts), but I would argue that the letters of the Imam open a window onto another domain of document-making in Islamic society: these letters issue from an Imamic chancery that parallels many governmental forms and institutions.11 Given that the Imam is writing not just as a community leader, but also as a supreme legal specialist,12 Jawād's letter illuminates the dynamics of interaction between the formation and implementation of Islamic law. Although this particular letter was written to respond to an ad hoc situation, the statements of all of the Imams were liable to be taken as the basis for future legal rulings.13 In this case, the potential legislative force of the Imam's ad hoc pronouncements proved uncomfortable for later Imamis who did not recognize certain legal ramifications embodied in this letter. Thus, this letter provides a window onto the very formation of Islamic positive law through practice. The implications of these processes are significant, not just for Shiʿi law, but for Islamic positive law more generally, which did not appear ex nihilo, but developed through social processes channelled by concrete social institutions. These processes are often invisible, and this letter gives us an important opportunity to study them. 9 See

Transmission history the in context of letters and works on khums
Before I proceed with a commentary on this letter, an analysis of its transmission history will help us contextualize what we are reading. I maintain that the clear and plausible context for the preservation of this letter, along with other indications that I will deal with below, should give us confidence that this letter may well preserve something close to the authentic instructions of the Imam to his followers.
This letter first appears in the Tahdhīb al-aḥkām14 by Shaykh al-Ṭāʾifa, Muḥammad b. al-Ḥasan al-Ṭūsī (d. 460/1067). The Qummī isnād is unremarkable, but attention should be drawn to the first transmitter of this letter: the well-known agent, ʿAlī b. Mahziyār al-Ahwāzī (death date unknown). Both his bibliographical record, and his historical positioning as an agent active in western Persia place him as a plausible figure for the transmission of this letter. As expressed in its preamble, the letter was written by the Imam to ʿAlī b. Mahziyār, and then read by his companion on the road to Mecca, perhaps in the context of Ḥajj, which would have been an important moment for coordinating with other members of the Imami community, as well as an important ritual observance. In carrying the Imam's letter, ʿAlī b. Mahiziyār was fulfilling one of the central aspects of the office of agent, which is primarily concerned with collecting money on behalf of the Imam; transmitting petitions and questions to the Imam; and carrying the Imams' letters to deliver or to read aloud to targeted individuals and groups in the community.15 ʿAlī b. Mahziyār is an interesting figure. Either he himself or his father was a Christian convert to Shiʿi Islam,16 a fact which evokes a hitherto unrecovered history of the Imami Imams as transitional figures for the conversion of the non-Muslim inhabitants of the empire.17 He was from Dawraq near Ahwāz,18 and transmitted hadith reports from Imam Riḍā, but thereafter became the agent responsible for Imamic communications with "several regions" under Imams Jawād and Hādī.19 Both his regional origin and his activities as an agent would have led him to establish connections with the inhabitants of the Jibāl to whom this letter was addressed. ʿAlī b. Mahiziyār's identity as an agent places his involvement in the transmission of this letter into perspective. He was not acting as a scholar, as a mere conduit for textual knowledge, but rather he was deeply involved in the institutions of Imamate, and his transmission of this text was part of his duty within those institutions. This involvement in the Imamic institutions is visible in the textual output of ʿAlī b. Mahiziyār. In general, it is difficult to discern particular interests in acts of hadith transmission, given the usually miscellaneous nature of hadith content. However, if we broaden our field of view to look at who else is preserving material similar to our letter, we can see something interesting. To contextualise ʿAlī b. Mahiziyār's transmission of a letter related to khums, let us look into the Shiʿi bio-bibliographical literature for precisely these two elements: Imamic letters, and fiscal-financial topics. Turning first to indications of interest in financial-fiscal information in the rijāl literature, the following is a list of authors mentioned by Najāshī as having compiled a Kitāb al-khums:20  When we compare these two lists, which represent separate spheres of transmission, we find only a single name present in both: ʿAlī b. Mahziyār.

Authors credited with a Kitāb al-khums in Najashī's Rijāl
If we reverse this experiment, we would expect that ʿAlī b. Mahziyār should be represented in the isnāds of hadiths relating to khums and letters and this is indeed the case,36 and it is notable that questions about khums are relatively commonly transmitted to the Imam via letter which helps us understand how a geographically widespread revenue-collection network was maintained and justified. This khums letter, then, appears to be one of a dossier of khums hadiths in which ʿAlī b. Mahziyār plays a notable role, which reflects his historical occupation as a fiscal agent for the Imam, rather than merely as a muḥaddith interested generally in the preservation of Imamic knowledge.
The fact that ʿAlī b. Mahziyār was both an agent and a muḥaddith is crucial. It reminds us that the collection of the Imami hadith corpus was implicated in practical administrative activities such as revenue collection and the drafting and dissemination of official letters. Legal hadith produced during the lives of Imams were not produced with post-Occultation legislation in mind, though that is how they are usually treated, but with the day-to-day implementation of legal precepts in life under an active Imamate. All of the contextual information we can find about the letter analysed here suggests that it did, indeed, emerge in the context of practical communications between the Imam and his followers, mediated by an agent, and responding to conditions on the ground. Whether the Imam himself put pen to paper is another question, but there is no good reason to doubt that, at the very least, the letter emerged from an Imamic chancery or secretariat and expresses the aims of the Imamate as an institution.

Historical context
Let us now examine the historical context suggested by the letter itself. While explicit details are scanty, the letter is a richly suggestive document that 36 ʿAlī b. Mahziyār is recorded as transmitting a letter on khums in the following cases: Ibn Bābawayh, Man lā yaḥḍuruhu al-faqīh, 2: 44; Kulaynī, Kāfī, 1: 547. In order to study the question of Imami khums I have collected as many hadith on this topic as I have been able to find, focusing initially on the early "Four Books" of Twelver hadith: Abū Jaʿfar Muḥammad b. provides valuable social context for the Imam's interaction with his community. The precise details of the historical context of this text are not clear, but we are fortunate to have a date for the letter, probably added parenthetically into the text of the letter at some point during its transmission. The date given is 220/835, the last year of the Imam's life. The historical event that the letter responds to is a battle in which followers of the Imam based in the Jibāl gained booty from a group of defeated Khurramiyya. The Khurramiyya were largely rural Iranians who objected to Arab-Muslim rule. They had Iranian dualist religious beliefs and cultural practices, which meant they would have been considered equally heretical by both ʿAbbasids and Imamis. It is not clear which engagement with the Khurramiyya the letter refers to. There were several battles with Khurrami rebels in and around western Iran at this time. The most famous rebel was Bābak (d. 223/838), who was based in Adharbayjān, rather than the Jibāl, though it is possible that the term "Jibāl" refers loosely to a broad mountainous zone that extends into Adharbayjān. In 220/835, there were, indeed, battles against Bābak, after the caliph Muʿtaṣim (r. 218-27/833-42) sent the general al-Afshīn (d. 226/841) to address the Khurrami threat. However, the sources do not report that large amounts of booty were taken in this year. Bābak was finally defeated in 222/837. An earlier battle may fit more closely with geographical hints and the mention of large amounts of booty. If we assume that the letter's date was added by later transmitters, it is possible that the booty mentioned was taken at an engagement against another Khurrami, ʿAlī b. Mazdak, who was defeated at a large battle in 218/833 near his stronghold in the village of Shahristāna, in the region of Farīvār, one of the five districts of Hamadān in the Jibāl. In this action, the caliph Muʿtaṣim sent 40,000 men with his general Isḥāq, and 10,000 Khurramis surrendered. Those who escaped marched on Isfahan, commanded by ʿAlī b. Mazdak, and started pillaging. The governor of Isfahan, ʿAlī b. ʿĪsā, was not in the city, but the city's Qāḍī and notables marched out and defeated the Khurramiyya, killing many and enslaving their women and children.37 Both the first defeat by the ʿAbbasid army and the latter defeat by the people of Isfahan appear to have resulted in booty being taken from the Khurramiyya. Hamadān is fairly regularly encountered in Imami sources as a place where members of the Imami community lived in this period. The fact that the letter's recipient, ʿAlī b. Mahziyār, was from Ahwāz in western Iran is suggestive of the way that regional communications were often handled by agents who hailed from the same region. It is, of course, possible that the engagement referred to in the letter was one that is not mentioned in the historical record. In any of these cases, it is clear that the military engagement was not led by the Imam himself, and could well have reflected Imamis fighting within the armies of the caliphate. The idea of Shiʿa fighting for the caliph seems unusual from the perspective of later doctrine, but it would probably have been unremarkable for those engaged in it, a case perhaps comparable to that of bureaucrats serving in caliphal administrations.38 However, we may nonetheless speculate that fighting for the caliph may have represented an infraction that required such expiation as referred to in the letter,39 or, more likely, that profits made in such an action were assessed as not fully ḥalāl, and therefore required purification through an expiatory payment.

The Arabic letter text
The letter is first recorded in Ṭūsī's Tahdhīb    for English translations of Qurʾānic passages, with alterations, which I deem suitable to the context of this letter. 60 Arberry translates ṣadaqa as "freewill offering." I prefer to leave it untranslated, as Jawād provides his own interpretation of Qurʾānic terms in his letter. 61 Arberry has "He turns". 62 Given its position following the Qurʾānic proof text, "that" must refer to a ṣadaqa taken on the properties (amwāl) of the believers as a purification in response to an act of repentance. Jawād uses this proof text to justify the khums. See commentary, below. 63 I.e. God imposed the zakāt explicitly through Qurʾānic statements. The Imam seems thereby to be treating his own legal injunctions as a separate layer of legislation alongside God's Qurʾānic stipulations, but without equal force. Twelver Shiʿi scholars tend to view the statements of Imams as representing God's law, but the Imam here clearly disinguishes between his own actions, reported in the first person ("I have imposed", "what I have done") and what God has imposed in the Qurʾān. 64 The reference to "the tax year (ḥawl)" is regarded as strange by later jurists (see below), for it is usually applied to the obligatory zakāt, but not to the khums. 65 Kennedy notes that, "By the eleventh century the Arabic world sulṭān is used as the title of individual rulers… In the ninth and tenth centuries, however, it was used to denote "the administration" or, in a very real sense, "the state" in Arabic historical sources. The question of whether a Shiʿi should pay twice for something that the government misappropriates is a common issue in Shiʿi law, discussed with regards to kharāj and zakātṣadaqa in different places. See Hayes, "Alms and the Man." 67 Arberry has "whatever booty you take". I have translated the phrase more literally as "whatever you gain." See the Imam's interpretation of "gains" (ghanāʾim) below. 68 I have placed "on" in parentheses to signal alternate possible interpretations of iḥtasaba, which can signify "to consider", "to account for"; "to reckon on", "to count on"; or "to prepare oneself for a reward." See Lane, Lexicon, 1: 565. 69 It is strange that an inheritance taken from someone who is neither father nor son should be classified among the "gains". Later jurists puzzled over this classification. See below, on the later reception of this letter. after some time has elapsed, for the intentions of a believer are better than his deeds. (7) As for what is obligatory upon the agricultural estates (ḍiyāʿ) and harvest revenues (ghallāt) each year, well, it is half a sixth [i.e. a twelfth] from someone whose estate is sufficient for his upkeep, and if his estate does not supply enough for his upkeep, then he need not pay half a sixth or anything else."

Structure
The absence of an opening formula or final blessings suggests that what is quoted here is not the entire letter. It may, however represent the majority of the letter, with opening and closing omitted in the interests of concision. This letter may have been responding to an earlier letter, for the Imam refers to the shortcomings of his followers, perhaps suggesting that earlier demands had not yet been fulfilled. Extant letters from this period on papyrus often begin by alluding to earlier correspondence. In other ways, the stylistic features of the letter fit generally with letters from the period.73 The categories referred to in this letter are not entirely familiar from the perspective of classical Imami fiqh. But before we discuss the later reception of this letter, let us attempt to understand it on its own terms. In the letter we can distinguish the following interrelated categories: (1) A khums levy understood as an expiatory and purificatory ṣadaqa.
(2) Khums imposed "this year" on gold and silver that has passed a tax year.
(3) Khums that is exempted "this year" but that could be imposed upon other assets (household goods, kitchenware and tableware, riding beasts, slaves) and revenues (the profits of trade and agriculture) (4) The obligatory zakāt imposed by God. (5) "Gains and benefits" (al-ghanāʾim wa-l-fawāʾid), which include the spoils of war, valuable gifts, certain categories of inheritance, and property gained whose owner is not known.
"let him make an intention to deliver it". This translation would fit the Imam's discussion of intentionality further on in the letter. See (6) A levy of a twelfth imposed on agricultural estates. To anyone familiar with classical fiqh, these categories will seem peculiar. While the zakāt is clearly separated from the other categories,74 the exact relationship between the expiatory khums and the "gains and benefits" is not clear. The letter is structured using various text markers, with new sections introduced by inna, innamā, fa-ammā, and fa-qad. These markers are important for understanding the relations between the categories. A further structural cue is provided by the two Qurʾānic proof texts that justify certain parts of the Imam's rulings. Three umbrella categories emerge from a reading of these proof texts: the expiatory khums justified with reference to Q 9:103-5; the obligatory zakāt, for which no text is provided here, but it is implied that God has made it explicit elsewhere; and the "gains and benefits" justified with reference to Q 8: 41.
Ghanīma is usually closely related to khums, and we would therefore expect the expiatory khums (category 1) to be an umbrella category including all the other levies except the obligatory zakāt, which is distinct. However, while categories 2 and 3 from the list above seem to belong to the expiatory khums, the syntax and structure of the letter sets category 5 clearly apart, through the structural marker "and as for" (fa-ammā), placing it in an ambivalent relation to the other categories. Another element that sets category 5 apart is the statement that the "gains and benefits" are said to be obligatory every year, in contrast to category 3, for which an exemption has been granted. Category 6, agricultural estates, has already been mentioned with categories 2 and 3. However, unlike categories 2 and 3, category 6 is not a levy of a "fifth" (khums), but rather a twelfth (niṣf al-suds). Thus, though we may be excused for attempting to impose the classical Twelver khums as an umbrella category that contains categories 1, 2, 3, 5 and 6, this letter provides fragile support to attest to such a master category. To resolve these difficulties, we must discuss the categories in detail.
Imam Jawād begins with a discussion of the expiatory khums. He states that he imposes the khums "this year" in particular, and that he intends thereby to purify his followers. To support this, Jawād quotes a Qurʾānic proof text, Q 9: 103-105, to show why his followers should pay the money demanded. This passage refers to the ṣadaqa in both singular and plural, as well as using the verb "purify" (tuzakkī), which is typically related through its root to the category of 74 The Imam leaves the nature of the zakāt without qualification or discussion in this letter: apparently this category is self-evident. We may assume, therefore, that the general outlines of zakāt were self-evident to the letter's intended recipients. The law of zakāt seems to have been relatively stable in its main features across communities in this period, and statements relating to zakāt attributed to earlier Imams do not diverge dramatically from Sunni and Zaydi conceptions. See Hayes, "Alms and the Man". zakāt.75 However, it is clear from what follows that this passage from the Qurʾān is invoked to support the Imam's demand for payment of khums, and not zakāt, for he follows the proof text with the statement, "But I did not impose that upon them every year; I do not impose anything upon them except for the zakāt, which God has made an obligation (faraḍa) upon them." The positioning of this Qurʾānic proof text in the letter makes it clear that it is being used to support khums. Because this proof text is being cited to justify the khums, we can infer that Imam Jawād considers the khums to be a kind of ṣadaqa: a pious offering that results in purification. The proof text also links ṣadaqa to contrition (tawba). This reference to the payment of a ṣadaqa as an act of contrition echoes the Imam's introductory comment about the shortcomings of his followers and his use of khums to purify them, suggesting that the payment of the khums serves as a kind of expiation for wrongs committed by them. One would be excused for assuming that this expiation was related to the failure of the Imam's followers to send the booty to him after their encounter with the Khurramiyya. However, this is not necessarily the case, for the direct link with expiation is only made in the case of the khums levied on gold and silver that has passed a tax year, and not on the spoils of war, which apparently fall into a different category.76 The Imam does not offer any explicit interpretation of the nature of the expiation he is offering by collecting khums on gold and silver. The opening statement that some of the Imam's followers had fallen short in their obligations and that he "wanted to purify them" using the khums suggests that the Imam was mandating the payment of khums as a one-time expiation for a one-time infraction. However, this seems to contradict the letter's suggestion that regular khums payments were expected unless a dispensation had been offered. It is not clear how to resolve this tension between the ad hoc and the expected. One interpretation might be that the previous infraction was precisely that the Imam's followers had not paid an earlier instalment of 75 EI2, s.v. Zakāt (Aron Zysow). 76 The case might be made that the gold and silver that have passed a tax year (ḥawl) are identical with the coin (ḍarb) taken from the Khurramiyya. Although this argument might seem plausible, it is incorrect, for the levy on the immediate windfall profit from military booty is distinct from monetary capital that has remained in one's possession for a full year. In addition, if they both belonged to the same category, then why would Imam Jawād deal with gold and silver in the section of his letter dedicated to the expiatory khums, and the coin from the Khurramiyya in a separate section dealing with ghanāʾim and fawāʾid? And why does he say that he is imposing the khums on gold and silver only "this year", while maintaining that the ghanāʾim (including that on the spoils taken from the Khurramiyya) is to be imposed every year? hayes Islamic Law and Society 28 (2021) 382-414 the khums, perhaps justified by complaints about the burden of governmental taxation, alluded to as "what the government misappropriates from their properties".77 By offering a dispensation, therefore, the Imam may have been indicating that the regular payment of the khums was still required, but that in practice he was willing to offer a discount in recognition of the difficult circumstances created by the illegitimate ruler's misappropriations.78 According to this interpretation, the khums would be understood as levy, at least some of which should be paid every year. In the case referred to in the letter, the expiatory function of the khums was invoked to meet specific circumstances, while also being understood as part of the regular, ongoing function of khums in general. Indeed, if we look at statements attributed to earlier Imams, the khums was already understood to have a purificatory function.79 Other interpretations are, of course also possible, and ultimately we must accept that we do not know what infractions motivated this letter and its imposition of the expiatory khums.
The status of Islamic canonical taxes levied by an illegitimate government is a frequent question in hadith and in later juristic discussions. Pious Muslims were understandably interested in whether wealth illegitimately seized as zakāt and kharāj by an oppressive government could be counted towards one's legitimate zakāt and khums.80 The reference to governmental misappropriations as the reason for the dispensation, then, is part of a larger discourse. In the absence of any further evidence, we may assume that the misappropriations refer either to regular governmental collections of zakāt (ʿushr) or to kharāj on agricultural lands. 77 Another possible explanation is that khums was a recognized category levied upon recognized items, but not at regular intervals. It is also possible that the mention of the dispensation is a rhetorical move, implying that the Imam's followers are lucky that he is giving them a discount on what they owe. While the Imam Jawād clearly states that the expiatory khums is separate from the obligatory zakāt, at least some of the items he mentions as subject to the khums (gold and silver that has passed a tax year, and agricultural profits) are precisely those that are subject to zakāt. This is a problem that was identified by later jurists, and I will return to it below.
The Imam indicates that the spoils from the Khurramiyya action fall under a different category from the levy on gold and silver, household items, and agricultural estates. While this categorization may have made sense to his audience, it appears strange from the perspective of classical Twelver fiqh. The spoils are termed as part of a category known as the "gains and benefits" (al-ghanāʾim wa-l-fawāʾid), which, in contrast to this expiatory khums, are dealt with in a paragraph that is introduced with a clear marker of contrast: "and as for…" (fa-ammā). In particular, the latter categories (2-3) appear to be a non-obligatory khums, while the Imam designates the "gains and benefits" as "obligatory… every year," and in order to demonstrate its obligatory character he cites a Qurʾānic proof text. This separation of a non-obligatory, or rather ad hoc, khums from a permanently obligatory ghanāʾim/fawāʾid does not appear to correspond to any other taxonomy of Islamic revenue categories of which I am aware. Ghanīma is usually considered one of between four and seven major categories of income upon which the khums is levied, along with mining, discovered treasure, pearl diving, salt production, and so on.81 In early Imamic hadith, terms are often left unexplained, but ghanīma, ghanāʾim and cognates like maghnam are usually understood as referring to the spoils of war. Jawād's list of items that are considered to be ghanāʾim and fawāʾid is unique, to my knowledge.
As for the term "benefit" (fāʾida, pl. fawāʾid), a few hadiths attributed to ear-lier82 Imams suggest that the meaning of fāʾida had already been raised as a question that required elucidation. Thus, for example, in one hadith, Imam Ṣādiq is asked about the meaning of ghanīma in Q 8:41, and he glosses ghanīma as referring to one's benefit or profit (ifāda) earned from day to day.83 The rest of the hadith is not without its difficulties, but Ṣādiq's statement suggests that fāʾida or ifāda was recognized as a distinct legal category, but one which was then still unclear and needed elucidation. The relationship between ifāda and zakāt or khums is left unexplained. Another hadith reported from an anonymous Imam by a transmitter who was a companion of al-Riḍā, al-Jawād, and al-Hādī84 equates fāʾida again with daily income, including the profits from trade and agriculture.85 But if, as these two hadiths suggest, fāʾida was daily income, then Jawād's intervention was doubly strange, for he lists the profits of trade and agriculture as part of the expiatory khums, clearly separated from the fawāʾid. If we assume that Jawād's letter does, indeed, postdate these hadiths, the apparent contrast between the letter and earlier tradition can be interpreted in a number of ways. It is possible that Jawād was unaware of these other Imamic statements, that he disagreed with them, or that he simply did not take them to be binding on his practice. None of these conclusions need disturb us, in spite of the orthodox Twelver expectation that the law of the Imams should be uniform, with any contradictions being merely apparent.
The category of the inheritance "not reckoned (on), from someone who is neither a father nor a son" is also obscure. The inclusion of this category as part of khums or ghanīma does not coincide with classical categories of inheritance law, whether Sunni or Shiʿi. Jawād's imposition of a levy on inheritance, is, it is true, restricted to a particular case. Later interpreters found the phrasing of the inheritance clause in this letter difficult.86 The phrase "not reckoned (on)" (lā yuḥtasab) admits at least two interpretations. (1) It could mean that the inheritance is unexpected and has not been reckoned on or counted on previously. It is a windfall profit, like other categories mentioned amongst the "gains and benefits". (2) Alternatively, it might mean that the inheritance is an uncanonical pay-out that is not to be considered as part of the inheritance proper. In this case, the phrase might be translated as follows: "The inheritance that is not to be accounted/considered." This interpretation implies that the kind of inheritance under consideration is not one that is recognized by some Shiʿi jurists, an interpretation taken up by later Shiʿi jurists, as we will see.
These interpretations of the inheritance clause lead us to an interesting set of debates surrounding the category of "neither father nor son", evoking an area of inheritance law in which the Imamis took a distinctive position which contrasted with that of both Sunnis and Zaydis. The problematic Qurʾānic category of kalāla has been debated energetically by scholars in recent years. In classical jurisprudence, kalāla came to be understood as the "collateral" heirs other than sons and fathers.87 In Imami jurisprudence, a similar understanding of kalāla exists. However, as Madelung notes: Sunni law, evidently following the actual practice, interpreted these Koranic rules [for the farāʾiḍ shares] as merely modifying the basic right of the ʿasaba [male relatives related to the deceased entirely through male links] who would inherit whatever was left after the distribution of the shares fixed by the Koran and, if there were no heirs entitled to a Koranic share, would take the total inheritance. The female relatives and dhawu 'l-arḥam [Sic. Male relatives related through female links] generally remained excluded except if there were no ʿasaba, however remotely related… In Imami law, the closest relatives, irrespective of sex or whether related through male or female links, inherit after satisfaction of the claims of any Koranic heirs.88 In the case raised by Jawād's letter, an inheritance that was from "neither a father nor a son" would be subject to divergent procedures in Imami law and Sunni law. If, for example, a Shiʿi male were to receive a share of inheritance that had been divided up by a Sunni executor, it would raise the possibility that he was excluding a female relative who was more closely related and more entitled to it than him, due to Sunni rules that favored the ʿaṣaba (taʿṣīb). It is possible, then, that Jawād's letter may refer to existing discussions within the Imami community about how disjunctions between Imami and Sunni practices were to be managed on a practical level. The challenge of managing the differences between Shiʿi concepts and Sunni concepts would have been constant in the lives of pious Shiʿa, especially in practical areas like inheritance law, and Shiʿi ideals must often have been challenged by prevailing practice in locations where the law of the land was administered by Sunnis.89 Khums was one way of managing the gap between ideals and reality. Statements ascribed to earlier Imams indicate that khums should be paid on property that includes a mixture of ḥalāl and ḥarām,90 in order to purify it. It may be that inheritance (mīrāth) 87  allocated according to Sunni categories was not considered (lā yuḥtasab) as legitimate, and it was for this reason that khums was levied on it to rehabilitate it from its partly ḥarām state. If this is what Jawād's letter is referring to, then the expanded meaning of the phrase would be: "Khums must be paid on an inheritance that is not considered legitimate due to the fact that it was inherited from someone other than one's father or son, and divided up according to Sunni rather than Imami rules for apportioning shares." As we shall see below, some later Imami jurists clearly imply that this is the correct interpretation.
The final category in the letter, agricultural estates, is also introduced by "and as for…" (fa-ammā), and has already been mentioned among the items exempted from khums this year. This earlier mention suggests that the agricultural estates are part of the expiatory khums but do not belong to the part that is levied in full (on gold and silver). A full dispensation on agricultural estates has not been granted, even though it was first mentioned alongside categories such as household goods and slaves for which a full dispensation was granted. As we will see, some later Imami scholars were perplexed that Jawād suggests a rate of a twelfth instead of a fifth (the literal meaning of khums), or the tenth (ʿushr) typically considered to be part of zakāt. This issue is not, however, problematic for Jawād, who wastes no time in justifying it Qurʾānically or rhetorically. However, it is worth noting again that this rate of a twelfth is unique, hence the consternation it raised later. The low rate of a twelfth that he imposes is perhaps part of the dispensation the Imam is offering, having been discounted from the higher rate of a fifth. If we follow this logic, then the one-twelfth levy on agricultural estates could be considered as a part of this expiatory fifth (khums), which includes the levy on gold and silver and household items. If this is the case, then we should note an interesting linguistic development whereby the khums has ceased to be merely a "fifth", a mathematical fraction, as its literal meaning suggests, and has instead become a proper noun, the name of a tax levied at various rates.

Implications
How should we understand the idiosyncratic relations between legal categories introduced by Imam Jawād in this letter? How do they add to our understanding of the historical development of the Imamate? This is where we must return to the concept of implementation, as opposed to legislation. While Occultation-era jurists tend to view Imamic hadith as a source of legislation, in its historical context this was a letter aimed at implementing certain administrative goals for revenue management.
In his brief mention of this letter, Modarressi asserts, "The systematic collection of the levy as a mandatory tax seems to have started in 220/835 when between implementation and legislation Islamic Law and Society 28 (2021) 382-414 Imam Muḥammad al-Jawād ordered his financial representatives to collect the khums on certain kinds of income."91 However, as we have seen, this letter does not provide any evidence for the introduction of a new systematic scheme for khums collection. What is introduced in this letter is neither new nor systematic. The letter clearly implies that the categories in operation in the year of writing had existed previously. The fact that these categories already existed does not mean that they were part of a regularized system. But if they had not been regularized in the past, they certainly were not regularized by means of this letter, for the Imam clearly states that he is imposing certain levies and excusing others for this year and not enforcing something regular and codified. The Imam offers a dispensation on various items, and one cannot have a dispensation without a prior expectation of regular payments. However, the mere existence of expectations does not imply a regularized system. While the tone of the letter implies that the institution of khums was, at this point, sufficiently established to make dispensations possible, it nonetheless shows that khums was subject to (and perhaps dependent on) ad hoc adjustment for its implementation. Instead of the systematization of regular collection, then, we should see this letter as an attempt to intervene in an ad hoc manner by drawing upon a complex web of expectations about how the Imam and his followers should interact.
This letter indicates that Islamic legal categories were in flux and open to interpretation. The obligatory zakāt was recognized as clearly mandated by God. Thus, the Imam's ability to innovate in the law is constrained by previous interpretations of the meaning of Qurʾānic writ and precedents and expectations about how this writ should be enforced. However, there is more room for interpretation in areas in which the Qurʾānic mandate is not so clear, and this is the case with the khums and ghanīma, which do not appear as stable, although they are justified with Qurʾānic proof texts. The understanding of khums in category (2) here has taken on the language of the tax year (ḥawl), which was an important aspect of the calculation of the classical zakāt. While ghanīma as spoils of war is a recognizable category, the other examples listed in this category are less clearly familiar from other texts in the Imami tradition. They appear to be linked by a common theme of unexpected or ownerless profit. Jawād's formulation, which links ghanāʾim and fawāʾid, is idiosyncratic, but he was conforming to a long-term trend in Imamic statements that interpret the Qurʾānic ghanīma verse as applying to more than just the spoils of war, thereby expanding the revenue that could be claimed as the prerogative of the Imam. Curiously, however, he does not do so by asserting that ghanīma as "gain" should be interpreted as the gains made in daily commerce and trade, that is, income. This would seem to be the simplest exegetical move, and one that ultimately became standard, but this interpretation of the Qurʾānic ghanīma appears not yet to have become dominant. Although categories like khums, ghanīma and fawāʾid must have been familiar to the Imam's followers and are clearly associated with each other, apparently they had not yet been systematically formulated into the classical Twelver category of khums, which perhaps required the collapse of the Imamate to fully crystallize. In contrast with the unequivocal nature of the zakāt, which is frequently mentioned in the Qurʾān as one of the central duties of being a Muslim, khums and ghanīma were sufficiently malleable categories to allow successive Imams to reinterpret and reapply them, albeit within the limitations of established exegetical and institutional norms.
The Imam's implementation of revenue-collection procedures are comparable to those used in governmental and non-governmental administrations. The fact that the Imam, at this moment, was interested in taxing only capital held in gold and silver, the booty taken from the Khurramiyya, and the profits of agricultural estates suggests that he was not merely clarifying the law and the Qurʾān, but acting like a pragmatic administrator in taxing the taxable: the practicality of levying revenue on money and agricultural produce rather than livestock and other kinds of property has been often noted with regard to governmental taxation, and it is likely that the Imam was motivated by similarly pragmatic considerations in deciding which categories to levy at that moment in time.

Reception of this letter during the lifetimes of the manifest Imams
The idiosyncratic nature of this letter was already noted during the Imamate of Jawād's heir and successor. Kulaynī reports the following hadith, which refers directly to this letter: tradition. Jawād's letter could not simply be ignored: it had been recorded and it represented legitimate Imamic writ, even if Jawād was merely implementing what he regarded as an ad hoc pragmatic solution to his immediate circumstances. In retrospect, his letter had the potential to be considered part of divine legislation. In his response, however, Imam Hādī shows no interest in discussing the normative force of Imamic precedents. He merely clarifies the term upkeep (maʾūna), apparently asserting a normative ruling without attempting to justify it in the context of earlier precedent.94 It is notable that Imam Hādī's response does not appear to address what seems to be a key problem: the fact that his father suggested that a twelfth should be levied on estates, rather than the canonical rates of a fifth or a tenth. It is possible, of course, that Hādī's response has been edited to omit details deemed unnecessary or inconvenient. Even so, it suggests that Hādī did not find it problematic to overrule recorded Imamic precedents to establish his ruling on what is to be considered as "upkeep".
This correspondence indicates that the legal precedents of earlier Imams limited the innovative legal capacity of living Imams, and that a ruling that contradicted an earlier precedent might need to be reconciled to the framework of earlier norms. It is notable that both Jawād and Hādī were capable of establishing new rulings on the finer points of khums law. However, their interventions were of a different nature. Jawād's letter was aimed at implementing his interpretation of the law. By contrast, Hādī's statement articulates a legal definition. In response to an abstract question about a generally applicable ruling, Hādī provided a generally-applicable rule. However, in spite of the ad hoc, practical nature of Jawād's letter, it was still an Imamic statement, which therefore created legal facts that had to be explained by later generations.

Later Twelver discussions of this letter
A brief discussion of the problems posed by this letter to later Imami scholars will help us place it in the perspective of longue durée historical changes in Imami legal and institutional structures. Ṭūsī includes this letter in his Istibṣār and in his Tahdhīb, but does not mention any of its aspects as problematic. Indeed, he discusses very little of the content that the letter is designed to convey. Instead, 94 The Imam explains that upkeep refers not only to the cost of the upkeep of the estate, including the governmental taxes to be paid upon it, but also to the upkeep of its owner and its dependents: the Imam did not want to impoverish his followers, after all. This position, however contrasts with the position of earlier Imams as reported by Hādī's petitioners. found this letter to be potentially problematic in several ways: "On the face of it, this hadith presents a number of problems that have perplexed those who have used it…" The problems he enumerates are much the same as we have noted when attempting to contextualize Jawād's letter in the framework of its immediate historical context: (1) The Imams are supposed to present a uniform and unchanging interpretation of divine law. So how could Imam Jawād impose one thing for one year, and not for others, according to his choice? (2) Jawād says that he imposes nothing except the zakāt that God has imposed, but then contradicts this when he says that the ghanāʾim and the fawāʾid are obligatory every year. (3) Jawād mentions the imposition of khums on gold and silver that has passed a full tax year (ḥawl), but the concept of the ḥawl applies to zakāt, not khums. (4) The reduction to the rate of a twelfth on agricultural estates is contradicted by other hadith which indicate that a fifth (khums) is to be paid on these estates, unless the revenue does not suffice for general upkeep.96 The twelfth to be levied on agricultural estates was also considered problematic by al-Ḥurr al-ʿĀmilī (d. 1104/1693) in his Wasāʾil. In his view, this anomalous twelfth was a temporary dispensation diverging from the basic category of a fifth.97 Jawād's letter, then, was something of a thorn in the side of Imami scholars who hoped to achieve a systematic harmonization of Imami jurisprudence, within which Imamic statements were supposed to represent God's eternal law, rather than the ad hoc deliberations of a pragmatic community-leader.
The difficulties of interpreting the inheritance clause in Jawād's letter also posed difficulties for later interpreters. In addition to the problems mentioned above, al-Muḥaqqiq al-Ardabīlī (d. 993/1585) cites the fact that the letter imposed khums on part of the inheritance as one of "the many rulings (aḥkām)" in this letter that "conflict with the madhhab" of Twelver Shiʿism.98 95 Unlike other jurists, Ṭūsī uses Jawād's letter to prove that the Imams had provided dispensations on khums regarding slave women (manākiḥ), but they stringently upheld the necessity of paying khums on gold and silver. Ṭusī, Tahdhīb, 4: 140. Although doubt was cast on the legal force of this letter, it has come to be seen in modern Uṣūlī jurisprudence as affirming the general principle that there is usually no khums on inheritance. Muḥammad Bāqir al-Ṣadr (d. 1980) resorts to a customary understanding (al-ʿurf al-ʿāmm) to interpret the letter, noting that it indicates that there is no khums on inheritance between fathers and sons.99 In the very restricted cases in which khums on inheritance is indicated by the letter, the principle of precaution (iḥtiyāṭ) has been invoked to imply that it might be better to uphold the ruling on the payment of khums on inheritance.100 However, the exact ruling of the letter remains open to question. Ayatollah ʿAlī Sīstānī, again invoking the principle of precaution, articulates the ruling for khums on inheritance as follows: If a Shiʿa Muslim inherits from a source which is not accepted in our fiqh, like inheriting from a distant relative despite his heirs being present (taʿṣīb), it will be considered a gain, and khums will have to be paid from it. Similarly, if a person inherits from an unexpected source, neither from his father nor from his son, then as an obligatory precaution, he will pay khums from that inheritance if it exceeds his annual expenses.101 The perplexity with which post-Mongol Imami jurists approach this letter indicates a change in the way the khums hadiths were viewed over time. Although the letter was a problem for later jurists, it was not for Ṭūsī. This is perhaps because it was Ṭūsī himself who was responsible for laying the foundations of the mature Twelver system for harmonizing the complex and contradictory evidence of Imamic hadith with the dictates of reason and accepted Imamic dogma. In doing so, he helped to lay the foundations for more stable positions within the Twelver madhhab.102 Thus it is not surprising that Ṭūsī should be comfortable with taking an unusual hadith like Jawād's letter and turning it to his ends without complaint, while those who worked from the crystallized positions of a post-Ṭūsī, and post-Ḥillī law system had difficulty in reconciling this hadith with the idea that khums should be a single, clear and unvarying category. Furthermore, for jurists of the Safavid era and later, khums was no longer just a legal abstraction, but an important pillar of the economic viability of Twelver institutions.103

Conclusions
Jawād's khums letter provides us with a glimpse into the workings of the Imamate as an institution and the role of the Imam as interpreter and arbitrator of the law. The details of the letter's preservation and transmission and the difficulty that it posed to later readers all suggest that it was not fabricated ex post facto. The letter shows that during the mid-eighth century ce, the category of khums and the related category of ghanīma were subject to ad hoc Imamic adjustment. Soon after, the letter was flagged as problematic in an epistolary exchange between Imam Hādī and one of his followers, giving us a sense of how the precedents of individual Imams were discussed before the Occultation removed the option of consulting an imam altogether. Modern scholarship often implies that as long as the Imams were alive, the community could simply ask them for a legal opinion. However, this letter and its reception indicate that Imamic statements, which were foundational for law, at least in potentio, were subject to complex assessments in order to determine how different statements related to each other. The Imams themselves had no interest in formulating abstract theories to define the normative force of their rulings. Thus, during the late Imamate, Shiʿi scholars had to assess earlier Imamic statements alongside those of the present Imam, and, if in doubt, petition the living Imam for a new opinion.
In the formative period of Islamic law (Shiʿi and Sunni alike) fiscal categories were fluid. The boundaries were blurry between categories like khums and ghanīma, jizya and kharāj, waqf and ṣadaqa, and zakāt and ṣadaqa. The development of different local usages over time complicated matters so that it is impossible to put forward a single stable definition for many of these terms. From at least the mid-second century ah onwards, however, the emerging class of jurists, the Shiʿi Imams among them,104 made efforts to clarify legal terms and to shape Islamic law into a coherent and effective system. This effort at systematization complicates our understanding of terms, for the systematizing 103 Calder jurists often attempted to assert a universal, final meaning for these terms, which could obscure or distort the original context. Scholars should seek to understand terms like khums within the semantic and institutional dynamics of specific historical cases. Attempts to find universally applicable definitions for many such terms is likely to fail. While laws promulgated by early caliphs may have had been ad hoc interventions that introduced new categories that were later canonized, Jawād's khums letter shows that the Imami Shiʿi Imams continued to redefine terms well into the ʿAbbasid period, at a time when Sunni schools had begun the project of systematizing legal systems that incorporated precedent and abstract legal taxonomies into a coherent whole. While it is true that the Occultation was a great impetus for the systematization of Twelver law, the reception of Jawād's khums letter indicates that, well before the Occultation, the Imams' ability to act as charismatic innovators of the law was restricted by precedent, preserved in the form of Imamic hadith, which were diffused among members of the community. The restriction of Imamic agency by hadith does not amount to systematization, but it does show that the Occultation era was not the moment of genesis of an Imami law based on hadith. On the contrary, the Imams themselves were subject to the precedent of earlier Imamic hadiths.
I am optimistic about the prospects for refining our understanding of the evolution of the Imami hadith corpus through the close study of key textual artefacts like Jawād's letter. However, we need to be precise in documenting our understanding of the institutional framework in which hadith circulated. In this case, the epistolary context of Jawād's intervention is crucial, for it highlights the complex relationship between legislation and implementation. Many Imamic statements are responses to abstract questions posed in the form: "what do you say about a man who does x?" Here, instead, we see the Imam responding to specific (albeit tantalizingly unspecified) historical circumstances. A key aspect of the practical functioning of the Imamate is the capacity of agents to arbitrate the Imam's writ in a distant community. The Imam was not present in the Jibāl to ensure that the khums was collected. Instead, his letter was carried by agents and disseminated from person to person.105 the Imam's legal rulings had become limited to a defined, bounded and separated Imami Shiʿi community, and communication was organized in a web of influence centred on the There is a prevalent model of engagement with hadiths that relies upon the idea of a "common pool" of reports that can be drawn upon at will to cement certain positions. But hadith should not be treated as isolated narratives free-floating in the sea of knowledge. Hadiths were produced in specific circumstances (epistolary exchanges, study-circles, lectures, petitions, private audiences) and thereafter circulated in mnemonic clusters: notebooks, selections, treatises and larger compilations. We should not treat all hadiths alike. Theological, fiqhī, and narrative hadith all deserve to be treated in different ways. Ultimately, we should neither assume that the hadith corpus of the Imams is wholly fabricated, nor wholly reliable. We need to work to build tools that enable us to more precisely identify the building blocks from which today's Imami hadith corpus was formed. Likewise, we should not treat the Imami Shiʿi community in isolation. This community shared sociological and intellectual commonalities not only with other Muslim sub-communities, but also with non-Muslims who shared the same spaces, institutional conventions and practices, and modes of living. Jawād's khums letter may be unusual within the Twelver hadith corpus, but it is comparable to letters sent by caliphs106 and their administrators, and preserved in historical works and the documentary record.