It is a widespread assumption in Islamic studies that the fields of sīra or maghāzī on the one hand and ḥadīth on the other are closely related and should be studied together.Footnote ¹ While it is obvious that both fields have a great deal in common with regard to content, form and transmission, the nature of their relationship remains a matter of debate. This article aims to contribute to the discussion of the early development of both fields, their distinctive features and their mutual influence.

For the sake of simplicity, the field of the biography of the Prophet will consistently be referred to as maghāzī in what follows, although the sources use different terms, such as sīra, siyar and maghāzī, which may or may not be used interchangeably.Footnote ² For our purposes it is of only minor importance whether the material relating to the biography of the Prophet was referred to as maghāzī or sīra material.

In the twentieth century two radically different views of the relationship between the fields of maghāzī and ḥadīth were proposed. According to one view, maghāzī-material is simply exegetical and juridical ḥadīth chronologically arranged. This view was put forward by Henri LammensFootnote ³ and was followed at least partly by C. H. Becker.Footnote ⁴ Becker summarizes Lammens’ view as follows: “In its detailed accounts, which are often diffuse, the Sīra is not an independent historical source. It is merely ḥadīth-material arranged in biographical order. The individual ḥadīths, however, are either exegetical elaborations of Qur'anic allusions or later inventions of dogmatic-juristic tendency. […] The actual historical material is extremely scanty. So the allusions of the Qur’ān are taken and expanded; and, first and foremost, the already existing dogmatic und juristic ḥadīths are collected and chronologically arranged. The result is the Sīra.”Footnote ⁵

Thus, according to this view, the exegetical and juridical aḥādīth existed before they were used in the maghāzī tradition, and maghāzī material is derived from exegetical and juridical ḥadīth. As far as I can see, this view is popular today almost exclusively with regard to exegetical traditions and much less with juridical ones.Footnote ⁶

According to the other view, the development was the other way round: the maghāzī material is older and maghāzī traditions were deprived of their historical setting and context and reduced to the juridical or theological aspects they contained to be then used as normative aḥādīth. This development, from maghāzī to ḥadīth, was proposed by John Wansbrough, who observed a “development from loosely structured narrative to concise exemplum”Footnote ⁷ and concluded: “The movement from narratio to exemplum illustrates perfectly the stylistic difference between Sīra and Sunna, between the mythic and normative preoccupations (Geistesbeschäftigungen) of early Muslim literature”.Footnote ⁸ He further remarked that it is possible to trace a theme “from the sīra-maghāzī literature, where it was historically articulated, to the sunna-ḥadīth literature, where it was idealized and hence shorn of its historical dimension”.Footnote ⁹ Wansbrough's main argument, that the narrative biographical interest in Muḥammad preceded the interest in him as an authority for legal matters, is not our concern here. However, Wansbrough also argued that this development can be observed in the study of single ḥadīths.Footnote ¹⁰ A similar view was held by Tilman Nagel, who observed in the ḥadīth literature an aim to eliminate historical contexts and instead create timeless, universally valid statements.Footnote ¹¹ Martin Hinds seems in general to subscribe to Wansbrough's view, but sees the development as being rather “from maghāzī to sunna via siyar and then sīra”.Footnote ¹²

Both views imply that traditions originate in one field and are then transferred to another, being reshaped on the way. While this seems likely, it is difficult to prove. It has long been known that traditions were not stable and underwent considerable changes in the course of transmission and the process can easily be demonstrated by comparing several versions of the same tradition.Footnote ¹³ The main difficulty lies in establishing where a tradition, which can now be found in different genres of literature, originated. If it can be shown that a tradition originated in one field and was only later used in others, we may gain valuable insights into the mechanisms that govern this kind of transfer.

A glimpse at the material in question shows us that all observations regarding the relationship between maghāzī and ḥadīth only apply to a part of the material. There are a vast number of traditions in the ḥadīth collections which have no parallel in the works on maghāzī, for example a large part of the material with ritual content. On the other hand, the books on maghāzī comprise much more than just ḥadīth: apart from traditions given with an isnād, some of which may also be included in collections of ḥadīth, there are quotations from the Quran, poems, lists of persons who took part in different events, a few documents (whose historical value shall not be discussed here) and comments and introductory sentences to other material.Footnote ¹⁴ Nothing of these latter types of material is included in any of the ḥadīth collections.

Our view of the range of material included in works on maghāzī is of course heavily dependent on the extant sources, but there are indications that different types of material played a role in maghāzī traditions in earlier periods and are typical of the maghāzī literature. For instance, many early maghāzī scholars are said to have included poems in their works.Footnote ¹⁵Maghāzī scholars who are said to have used poems or be lovers of poetry include: Abān b. ʿUthmān,Footnote ¹⁶ ʿUrwa b. al-Zubayr,Footnote ¹⁷ Wahb b. Munabbih,Footnote ¹⁸ ʿAbdallāh b. Abī Bakr,Footnote ¹⁹ al-Zuhrī,Footnote ²⁰ and of course Ibn Isḥāq, al-Wāqidī and, to a lesser extent, Ibn Saʿd.Footnote ²¹ The same applies to lists of participants, which some of the early authorities of maghāzī are said to have kept,Footnote ²² and to documents, such as letters of the Prophet, which were included by some.Footnote ²³ Thus it is likely that earlier authorities on maghāzī collected different types of material of about the same scope as are included in the later works of Ibn Isḥāq, al-Wāqidī, and others.

It may be worth establishing the true scope of the intersection between the material included in books on maghāzī and that included in collections of ḥadīth, but this is not our concern here. For our purposes we will concentrate on the material included both in books on maghāzī and in collections of ḥadīth.

Considering the differing aims of muḥaddithūn and maghāzī scholars, it is to be expected that they dealt differently with the material. The muḥaddithūn were primarily interested in the transmission and preservation of the material according to certain standards, and in its legal or ritual relevance. They were considered to be authorities on the reliability of certain transmitters and certain lines of transmission, whether a certain tradition was more or less reliable or legally binding, the exact wording of certain traditions, and on who was the original narrator of a story.

Maghāzī scholars on the other hand were interested in creating a continuous and coherent narrative of the life of Muḥammad. To this end they had to draw connections between different traditions and establish causalities between them. As we have seen, they also drew on different kinds of material. They were considered experts not in the question of the authenticity of the material they used, but in questions of context. The maghāzī scholars knew – or at least were supposed to know – when a certain event took place, its causes, and whether or not it preceded another event. They were also experts on which people were involved: they knew who was present at a certain event, how many people took part in a battle, if a specific individual took part in a battle, if someone belonged to the muhājirūn or to the anṣār, if someone died before or after a certain event – the kind of information that could not usually be derived from a single ḥadīth and could not be passed on by relying solely on aḥādīth.

In order to form coherent narratives from their material, it was almost inevitable that authors would have to abstain from naming informants, and among the oldest authorities on maghāzī, such as ʿUrwa b. al-Zubayr (d. 93 or 94), Saʿīd b. al-Musayyab (d. 94) and Ibn Shihāb al-Zuhrī (d. 125) we frequently find traditions given without any indication of their sources. It is precisely this handling of the material that makes it possible to trace its origins back to the field of maghāzī and not to ḥadīth. This applies mainly to long traditions that contain complete accounts of what happened at a certain event, mostly major events in the life of Muḥammad such as the beginning of the revelation, the Hijra, the battles of Badr and Uḥud, the treaty of al-Ḥudaybiya and the conquest of Mecca, to name but a few. The long traditions regarding these events are usually made up of several elements which are combined into a coherent narrative. That they are made up of different and independent units can be seen since, in different versions of the narratives, these units often appear in a different order; sometimes they even appear in different contexts.Footnote ²⁴ These stories often presuppose an omniscient narrator, a narrator who knows and reports what is happening and what is said both in the camp of the Muslims and in that of the Meccans. In ʿUrwa b. al-Zubayr's account of al-Ḥudaybiya, for instance, discussions among the Muslims and those among the Quraysh in Mecca (when their delegates return from Muḥammad), are reported in direct speech. Later in this account, even the conversation between Abū Baṣīr and his two counterparts is given in direct speech.Footnote ²⁵ Such an omniscient narrator should not occur in the field of ḥadīth, where the rules say that the original narrator has to be an eyewitness,Footnote ²⁶ particularly when direct speech is reported.Footnote ²⁷ Whether these were really eyewitness reports or if this was just a literary convention is irrelevant here. In any case, it would be against the rules for an original narrator to present dialogues of scenes at which he cannot possibly have been present. Thus, while these stories do not conform to the standards of ḥadīth, they would still be considered to be akhbār.Footnote ²⁸ In general, the literary conventions of akhbār are not very different from those of ḥadīth, but they seem to have been less strictly observed in fields outside ḥadīth proper. One notable distinction is that for akhbār it is characteristic that the narrator is absent from the narration.Footnote ²⁹ This is usually the case in these long narratives, but not necessarily in ḥadīth.

Many of these complex narratives were based on earlier accounts from different informants and constitute combined reports. Combined reports are a common feature in Islamic historiography at least from the time of Ibn Isḥāq and al-Wāqidī. They result from the merging of different accounts into a single narrative. Usually Ibn Shihāb al-Zuhrī (d. 124/742) is credited with introducing the technique of combined reports.Footnote ³⁰ This attribution to al-Zuhrī seems to be based on his use of collective asānīd.Footnote ³¹ However, there is evidence that earlier authorities of maghāzī already combined different reports to a coherent narrative, although they mostly did not provide them with collective asānīd. A comparison of different versions of maghāzī traditions can provide evidence that the practice of combining reports was common among early maghāzī authorities.Footnote ³² A similar practice can be observed in the field of the ayyām al-ʿarab,Footnote ³³ which may have served as a model for the maghāzī scholars.

At the beginning these reports seem to have been given without any mention of isnād, and only later, possibly under the influence of the muḥaddithūn, with a kind of collective isnād. We can find narratives without an isnād going back to an eyewitness, for instance in the traditions of Shuraḥbīl b. Saʿd, ʿUrwa b. al-Zubayr, al-Zuhrī, Wahb b. Munabbih, ʿAbdallāh b. Abī Bakr, ʿĀṣim b. ʿUmar and Mūsā b. ʿUqba. The use of the collective isnād reached its highest development with al-Wāqidī, but it can also be found in Ibn Isḥāq's book. Occurrences of collective asānīd can also be detected in traditions from Mūsā b. ʿUqba, al-Zuhrī and even ʿUrwa b. al-Zubayr, but they seem to be the exception here rather than the rule.

With or without collective asānīd, these long and complex narratives originated in all likelihood in the field of maghāzī, and not in the field of ḥadīth. Nevertheless, a number of them can also be found in ḥadīth collections, sometimes in chapters on maghāzī, sometimes in other chapters, according to their legal implications.

These traditions thus enable us to observe how the muḥaddithūn were dealing with material which was originally part of the maghāzī tradition. Comparing different versions of these long accounts as quoted in the works on maghāzī on the one hand and in ḥadīth collections on the other, it can be observed that these long traditions are sometimes quoted in full in ḥadīth collections, while sometimes only the legally relevant parts are adduced, often in different chapters of the same collection: quoting only part of a tradition was obviously considered to be an acceptable practice.

Wansbrough, in a study of the story of the slander about ʿĀ’isha, the ḥadīṯ al-ifk, argued on the basis of a critical analysis of three versions of the tradition that the version included in al-Bukhārī's ḥadīth collection is a late reworking of the basic narrative recorded by Ibn Isḥāq. The purpose of the reduction of the story to its “parabolic nucleus”, according to Wansbrough, was exclusively paradigmatic.Footnote ³⁴ However, Schoeler, on the basis of an analysis of the texts and asānīd of numerous versions of the tradition, showed that the version included in al-Bukhārī's collection is very close to the oldest recension of the story, while Ibn Isḥāq combined different traditions to build his narrative.Footnote ³⁵ Thus in this case there is no indication of a deliberate reshaping of the tradition to make it conform to the needs of the muḥaddithūn.

The ḥadīth al ifk is one of a number of long and complex narratives on the life of Muḥammad going back to ʿUrwa b. al-Zubayr, versions of which are included both in maghāzī works and in ḥadīth collections. Some versions of these traditions are more elaborate and embellished than others, and some versions contain elements lacking in others. But the versions adduced in ḥadīth collections do not differ in any systematic way from those quoted elsewhere; for instance they do not as a rule contain fewer names or omit place names. What can be observed, however, is that they are usually equipped with complete asānīd, while in the works of maghāzī this may or may not be the case.

We may therefore infer that when muḥaddithūn included material based on long narratives originating in the field of maghāzī in their collections, they were trying to get hold of the versions that best conformed to their standards. They also felt free to quote only part of a tradition, but they did not as a rule reshape the tradition to produce normative texts.

Where shorter traditions are involved things become more complicated; unlike the long narratives or combined reports, these would also conform to the standards of ḥadīth. In this case it is more difficult to establish where a tradition originated and how it was employed in other fields.

In what follows two case studies shall be discussed. The first consists of several versions of a tradition relevant to the discussion of whether it is permissible to eat game while being in the ritual state of iḥrām. They served Nagel as the basis for his argument that the maghāzī materials were deprived of their historical setting when they were transferred to the field of hadīth.Footnote ³⁶ First I will present a common version of the tradition and then Nagel's view of its development. The tradition in one version in al-Bukhārī's Ṣaḥīḥ is given with the following wording: ʿAbdallāh b. Abī Qatāda al-Aslamī narrates from his father, Abū Qatāda, who says:

There are several versions of this tradition and they differ in many details. Sometimes the Prophet simply allows the pilgrims to eat from the meat without doing so himself, sometimes place names are given, in some versions the Prophet asks if anyone has encouraged or ordered Abū Qatāda to hunt the wild ass, and only after this is denied does he allow the pilgrims to eat from the meat.Footnote ³⁸

These details are of course important for the juridical aspect of this story: while it is forbidden to hunt in the state of iḥrām, this tradition can be adduced as an argument that it is permitted to eat game so long as the muḥrim is not involved in the hunting in any way. In fact, most elements of the tradition seem to serve a legal argumentation and show that Abū Qatāda's companions did not help him in any way. They did not call his attention to the wild ass and they did not hand him the whip or spear, although he asked them to do so.

Our main point of interest is the place name, as in some versions this story is said to have taken place in the year of al-Ḥudaybiya on the way to Mecca.Footnote ³⁹ This is also the context in which al-Wāqidī places the story: it took place near al-Abwā’, and although he does not mention all of the details given above, the outline of the story is the same.Footnote ⁴⁰

Nagel argued that this context, the story of the attempted pilgrimage of Muḥammad which resulted in the treaty of al-Ḥudaybiya, is indeed the origin of this tradition. In his view, in the ḥadīth-version the connection to al-Ḥudaybiya was eliminated and instead the explicit permission of Muḥammad to eat the meat, which can be found in several versions of the ḥadīth, was added. The aim of this was to eliminate the historical context and instead create a timeless, universally valid statement.Footnote ⁴¹

However, this view is questionable: the tradition in question cannot be found in the context of al-Ḥudaybiya in the maghāzī literature before the time of al-Wāqidī. The ḥadīth was apparently not included by Ibn Isḥāq and as I see it was not transmitted by Mūsā b. ʿUqba, ʿUrwa b. al-Zubayr, al-Zuhrī or any other famous authority of maghāzī. On the other hand, as we have seen, it is present in numerous ḥadīth collections and plays an important part in the question of whether it is permitted for a muḥrim to eat meat from hunted animals. In Muslim's Ṣaḥīḥ, nine versions of this ḥadīth are presented next to six aḥādīth giving the opposite view, saying that Muḥammad was in the state of iḥrām when he was offered meat from someone who hunted it and that he refused to eat from it. Other hadīth collections offer a similar view: the Abū Qatāda ḥadīth is one of several relevant to the question of whether a muḥrim is allowed to eat game.

In most of these aḥādīth no historical context is mentioned while in a few versions of the Abū Qatāda ḥadīth it is mentioned that the incident happened on the way to al-Ḥudaybiya or in the year of al-Ḥudaybiya. It is instructive to look at the asānīd: all versions containing the reference to al-Ḥudaybiya share the common transmitter Yaḥyā b. Abī Kathīr, while Yaḥyā is not present in any version of the ḥadīth without reference to al-Ḥudaybiya. The only exception is the version adduced by al-Wāqidī, in which reference to al-Ḥudaybiya is made, but in the isnād of which Yaḥyā b. Abī Kathīr is not mentioned. However, al-Wāqidī regularly uses fictitious asānīd in order not to reveal his true sources, as can be shown in several cases.Footnote ⁴² We should not therefore pay too much attention to his indication of the tradition's origin. Yaḥyā b. Abī Kathīr would thus be a partial common link according to the terminology of Gautier Juynboll.Footnote ⁴³ Juynboll was among the first to apply systematically the common link theory, first formulated by Joseph Schacht,Footnote ⁴⁴ and to develop it further, introducing a number of helpful technical terms. While the significance of the common link has been subject to debate,Footnote ⁴⁵ some conclusions can usually be drawn from a study of the asānīd and the respective variants in the transmitted texts. For instance, given that there are a large number of variants of one tradition, it seems safe to assume that a specific element in the text of a ḥadīth was introduced by a certain transmitter if this element occurs only in those variants that were passed on by this transmitter and does not occur in any other variant.

Judging from the asānīd we thus have to assume that it was Yaḥyā b. Abī Kathīr who first made a connection between the Abū Qatāda ḥadīth and al-Ḥudaybiya. Yaḥyā b. Abī Kathīr died in 129 or 132,Footnote ⁴⁶ and although he is usually considered to be trustworthy, al-Ṭabarī accuses him of tampering with asānīd.Footnote ⁴⁷

Taking into account the markedly legal character and the prominence of the Abū Qatāda ḥadīth in the ḥadīth collections – it is found in all canonical and several other collections – and its relative absence from the maghāzī tradition, it is highly unlikely that this ḥadīth originated in the maghāzī tradition on al-Ḥudaybiya. On the contrary, we must assume that this ḥadīth was circulating among the muḥaddithūn, that it was part of a purely legal discussion and that it was only later included by al-Wāqidī in his material on the maghāzī. The context apparently was provided within the legal discussion.

If this assumption is right, how can we explain that the ḥadīth was furnished with a historical context at a secondary stage within the legal discussion? Might this simply reflect a desire on the part of the transmitters to supply information originally left vague? This is, of course, possible. However, in a legal discussion, the historical context was not of primary importance. On the other hand, the context may become important when it comes to the possible abrogation of aḥādīth. Just as later verses from the Quran could abrogate contradicting earlier verses, the same was true for aḥādīth. A legal decision taken by the Prophet in the last years of his life would invalidate earlier rulings to the contrary, as is explained for instance in Ibn al-Ṣalāḥ's ʿUlūm al-ḥadīth.Footnote ⁴⁸ Books on abrogating and the abrogated in the ḥadīth form a literary genre of their own. The editor of Ibn Shāhīn's Kitāb Nāsikh al-ḥadīth wa-mansūkhihi, Karīma bt. ʿAlī, lists fourteen works on the topic.Footnote ⁴⁹ None of the works she mentions date from before the third century ah; however, it is most likely that the discussions about abrogation in ḥadīth predate the first works dedicated to the topic considerably. Providing the Abū Qatāda ḥadīth with a context in the year of al-Ḥudaybiya, only four years before Muḥammad's death, could strengthen the position of those scholars who argued that the muḥrim is allowed to eat game, since the ḥadīth might be abrogating earlier aḥādīth to the contrary. Al-Wāqidī was himself an expert in fiqh,Footnote ⁵⁰ and included far more legal aḥādīth in his work than did for instance Ibn Isḥāq.Footnote ⁵¹ It is conceivable that al-Wāqidī included the Abū Qatāda ḥadīth in his account on al-Ḥudaybiya not only because he felt that this was its correct historical context, but also because the legal view expressed in this ḥadīth corresponded to his own opinion. He does quote a tradition to the contrary – Muḥammad refuses to take a piece of wild ass offered to him because he is in the state of iḥrām – immediately following the Abū Qatāda ḥadīth.Footnote ⁵² But in the context of the farewell pilgrimage, yet another similar story is told: someone on the way offers Muḥammad a wild ass he hunted. Muḥammad and his companions are in the state of iḥrām and Muḥammad offers the meat to them, saying that it is allowed for them, so long as they did not hunt themselves nor order someone else to do so.Footnote ⁵³ Whether or not al-Wāqidī included these traditions to support a legal view, it should have become clear that this tradition first circulated among legal and ḥadīth scholars and was only transferred to the maghāzī-tradition at a secondary stage.

The second case consists of several versions of a tradition dealing with a woman who committed theft and who is punished by having her hand cut off. The tradition again is found in numerous versions that differ in several details.Footnote ⁵⁴ A typical version adduced by al-Bukhārī reads as follows:

While in most versions of this ḥadīth it is not indicated where or when the incident happened, in some versions the story is said to have taken place during the conquest of Mecca.Footnote ⁵⁶ Following the arguments of Wansbrough or Nagel we should assume that the story is taken from the maghāzī material and that the form given above – which does not mention this historical context – is a later reworking. But the story is not found in this context in most of the books on maghāzī. It is mentioned by Ibn Saʿd and Ibn Kathīr, but not by al-Wāqidī or Ibn Isḥāq. It is often quoted on the authority of al-Zuhrī and ʿUrwa b. al-Zubayr, but it is not included in their long accounts of the conquest of Mecca.Footnote ⁵⁷

It is helpful to look at the variants in more detail. The tradition is included in ʿAbd al-Razzāq's Muṣannaf, where the person intervening in favour of the woman is once given as Usāma b. Zayd (which is the most popular version), and once as ʿUmar b. Abī Salama.Footnote ⁵⁸ Muslim once names Umm Salama as the one who intercedes on behalf of the woman.Footnote ⁵⁹ Ibn Saʿd once gives the name of the woman – who is not usually mentioned by name – as Fāṭima bt. al-Aswad b. ʿAbd al-Asad and once as Umm ʿAmr bt. Sufyān b. ʿAbd al-Asad. In the latter case the affair is said to have happened at the farewell pilgrimage.Footnote ⁶⁰ It is interesting to note that Ibn Saʿd mentions the incident only in the biographical entries of Usāma b. ZaydFootnote ⁶¹ and Fāṭima bt. al-Aswad,Footnote ⁶² but not in the passages on the conquest of Mecca or the farewell pilgrimage.

It is very probable that this alleged saying of Muḥammad, too, was provided with a context only at a secondary stage: there are different versions regarding both the persons involved and the historical context, which is given either as the conquest of Mecca or the farewell pilgrimage. As in the case of the Abū Qatāda tradition, a study of the asānīd indicates who may be responsible for establishing a connection between the story and the conquest of Mecca: all versions placing the story in this context of the conquest share the transmitters ʿAbdallāh b. Wahb (125–197)Footnote ⁶³ and Yūnus (d. 152 or 159),Footnote ⁶⁴ while all versions transmitted by other individuals do not establish this connection. As both ʿAbdallāh and Yūnus occur in all versions mentioning Mecca, we cannot be sure who ultimately made the connection, but it seems probable that the connection was not made before the first half of the second century.

The variants regarding historical context and persons involved as well as the fact that the tradition cannot be found in the works on maghāzī prior to Ibn Saʿd make it probable that it is a legal tradition only later included in works on maghāzī. As in the case of the Abū Qatāda tradition, it is not unlikely that the association of the story with events in the last years of the Prophet – the conquest of Mecca and the farewell pilgrimage – was made in support of the legal implications of the ḥadīth and results from the discussion about abrogation of ḥadīth. Another possible explanation would again be the desire to supply additional information, which in this case might account for the conflicting identifications of the woman and those interceding on her behalf. In any case, the historical context seems to have been provided only at a secondary stage, but still within the legal discussion. This historical context, provided in some variants of the tradition, led to their eventual inclusion in works on maghāzī.

The case studies adduced above suggest that it is indeed possible in some cases to establish that the occurrence of a tradition in one field preceded its use in another. That some traditions were employed in one field before they were transferred to others is not per se surprising and can best be explained by envisaging different circles of scholars discussing different issues. Traditions with juristic content would circulate mainly among jurists and ḥadīth scholars, while traditions with historical content would circulate mainly among those occupied with the maghāzī or related fields.

We have seen that traditions in the course of transmission were subject to change; the circumstances of transmission and the different interests of the transmitters involved shaped them and led to the emergence of numerous variants of a tradition. The fact that ḥadīth and maghāzī scholars had different aims and priorities when passing on traditions this left its mark on those traditions.Footnote ⁶⁵

We can therefore show that some traditions – the long coherent accounts of the main events in the life of Muḥammad, but possibly other traditions too – were included in the field of ḥadīth only after they obtained their basic form in maghāzī circles. On the other hand we have juristic aḥādīth that were only introduced into the field of maghāzī after they had circulated among ḥadīth scholars and were shaped by them. The transfer of traditions from one field to the other did not necessarily involve any deliberate changes to the text.

It is less obvious than it first seems that maghāzī material found its way into some collections of ḥadīth. Of course, it can be argued that everything the Prophet did is sunna and therefore everything from the maghāzī tradition might become part of the ḥadīth as long as it met the formal standards. But as we know the ahl al-maghāzī were often said to be transmitting traditions according to standards that were not acceptable to the muḥaddithūn, and so the muḥaddithūn sometimes adduced maghāzī material although it did not really conform to their standards. We can find such traditions, for instance combined reports going back only to ʿUrwa b. al-Zubayr or al-Zuhrī and not to an eyewitness in ʿAbd al-Razzāq's or Ibn Abī Shayba's Muṣannafs.

Apparently, the traditions of some of the authorities on maghāzī were deemed by the muḥaddithūn to be good enough to be included in their works. It seems that this happened only with traditions of those early maghāzī scholars who were also known for their expertise in ḥadīth, as Saʿīd b. al-Musayyab, ʿUrwa b. al-Zubayr, ʿAbdallāh b. Abī Bakr, or al-Zuhrī.Footnote ⁶⁶ The combined reports of later authorities in both maghāzī and ḥadīth, such as Mūsā b. ʿUqba were not adopted by the muḥaddithūn.

The versions included in the canonical ḥadīth collections are usually the ones with the best isnād. It is not clear at this point whether the isnād was deliberately improved when the materials came to be used and transmitted by the muḥaddithūn, or whether the muḥaddithūn simply selected the versions with the best asānīd from all the different versions that existed. In some cases, the isnād of one of the informants, of one part of the combined report, seems to have been regarded as the isnād for the whole tradition.

Maghāzī scholars, on the other hand, did not of course have a problem including juristic aḥādīth in their materials. Apparently, as the corpus of ḥadīth was growing and the first collections of ḥadīth were emerging, maghāzī scholars were starting to use and exploit these sources as they were using poems, verses from the Quran, stories from quṣṣāṣ and other material. This is a trend that seems to have continued for a long time: Ibn Isḥāq does not quote many explicitly legal aḥādīth, yet al-Wāqidī has a large number of these aḥādīth and Ibn Kathīr adduces even more. This is of course not necessarily a general development but might also be due to the personal preferences of the authors.

The influence of ḥadīth on maghāzī is not limited to the additional material it provided. Maghāzī scholars were also influenced by the muḥaddithūn regarding formal aspects of transmission. The use of the isnād – the backbone of the sciences of ḥadīth – came to be more important in the field of maghāzī. It was quite possibly the scholars versed in both ḥadīth and maghāzī who first advocated and advanced the use of the isnād in the field of maghāzī, and the importance of the isnād and its use in the field of maghāzī was continually growing.

It is noteworthy that the early expert in maghāzī and ḥadīth, ʿUrwa b. al-Zubayr, only occasionally gave indications of his sources in his long historical accounts. This is in contrast to the legal or exegetical traditions transmitted on his authority, which are usually provided with asānīd that include ʿUrwa's sources. However, it is impossible to say whether these asānīd grew backwards and were improved in the course of transmission. A generation later, with Ibn Shihāb al-Zuhrī, it had become customary to furnish the long historical traditions with some kind of isnād, but there are still traditions attributed to him and not traced back any further. A further generation later, with Ibn Isḥāq, it seems to have become the rule that most of the material should be introduced with an isnād; Ibn Isḥāq regularly employs collective asānīd when he introduces his combined reports, and so does al-Wāqidī.

We may conclude that although the fields of maghāzī and ḥadīth are closely related, they remained distinct.Footnote ⁶⁷ They influenced each other, and quite a number of traditions from one field could also be seen relevant to the other. But neither can the maghāzī be regarded as secondary to and derived from the ḥadīth, nor can the opposite view be upheld.