See W. A. Graham, Divine Word and Prophetic Word in Early Islam (1977); G. H. A. Juynboll, Muslim Tradition (1981).
In Islamic terminology, the term hadith refers to reports about the statements or actions of the Islamic prophet Muhammad, or about his tacit approval of something said or done in his presence. Classical hadith specialist Ibn Hajar says that the intended meaning of "hadith" in religious tradition is something attributed to Muhammad, as opposed to the Qur'an. Other associated words possess similar meanings: "khabar" (news, information) often refers to reports about Muhammad, but sometimes refers to traditions about his companions (sahāba) and their successors from the following generation (tābi'īn); conversely, "athar" (trace, vestige) usually refers to traditions about the companions and successors, though sometimes connotes traditions about Muhammad. The word sunnah (custom) is also used in reference to a normative custom of Muhammad or the early Muslim community.
Hadith were eventually written down, evaluated and gathered into large collections mostly during the reign of Umar II (bin Abdul Aziz, grandson of Umar bin Khattab(RAA)2nd Caliph) during 8th century, and also in the 9th century. These works are referred to in matters of Islamic law and History to this day.
Muslim historians say that khalifa Osman (the third khalifa, or successor of Muhammad, who had formerly been Muhammad's secretary), was the first to urge Muslims to write the Qur'an in a fixed form, and to record the hadith.Osman's labours were cut short by his assassination, at the hands of aggrieved soldiers, in 656.
The Muslim community (ummah) then fell into a prolonged civil war, which Muslim historians call the Fitna. After the fourth khalifa Ali ibn Abi Talib was assassinated in 661, the Umayyad dynasty seized control of the Islamic empire. Ummayad rule was interrupted by a second civil war (the Second Fitna), re-established, and ended in 758 when the Abbasid dynasty seized the khilafat, and held it, at least in name, until 1258.
Muslim historians say that hadith collection and evaluation continued during the first Fitna and the Umayyad period. However, much of this activity was presumably oral transmission from early Muslims to later collectors, or from teachers to students. If any of these early scholars committed any of these collections to writing, they have not survived. The histories and hadith collections we have today were written down at the start of the Abbasid period, more than a hundred years after Muhammad's death.
Scholars of the Abbasid period were faced with a huge corpus of miscellaneous traditions, some of them flatly contradicting each other. Many of these traditions supported differing views on a variety of controversial matters. Scholars had to decide which hadith were to be trusted as authentic and which had been invented for political or theological purposes. To do this, they used a number of techniques which Muslims now call the science of hadith.
At the beginning of the 7th century, those receiving the hadith started to question the sources of the saying. The hadith were eventually recorded in written form, had their Isnad evaluated, and were gathered into large collections during the 8th century.
Non-Muslim scholars note that there is a great overlap between the records of early Islamic traditions. Accounts of early Islam are also to be found in:
Hadith are generally categorized as sahīh (sound, authentic), da`īf (weak), or mawdū` (fabricated). Other classifications used also include: hasan (good), which refers to an otherwise sahīh report suffering from minor deficiency, or a weak report strengthened due to numerous other corroborating reports; and munkar (ignored) which is a report that is rejected due to the presence of a solitary and generally unreliable transmitter. Both sahīh and hasan reports are considered acceptable for usage in Islamic legal discourse. Classifications of hadith may also be based upon the scale of transmission. Reports that pass through many reliable transmitters at each point in the isnad up until their collection and transcription are known as mutawātir. These reports are considered the most authoritative as they pass through so many different routes that collusion between all of the transmitters becomes an impossibility. Reports not meeting this standard are known as ahad, and are of several different types.
Another area of focus in the study of hadith is biographical analysis (`ilm al-rijāl, lit. "science of people"), in which details about the transmitter are scrutinized. This includes analyzing their date and place of birth; familial connections; teachers and students; religiosity; moral behaviour; literary output; their travels; as well as their date of death. Based upon these criteria, the reliability (thiqāt) of the transmitter is assessed. Also determined is whether the individual was actually able to transmit the report, which is deduced from their contemporaneity and geographical proximity with the other transmitters in the chain. Examples of biographical dictionaries include Ibn Hajar al-Asqalani's "Tahdhīb al-Tahdhīb" or al-Dhahabi's "Tadhkirat al-huffāz.
However, some Muslim scholars have undergone Western academic training and attempted to mediate between the traditional Muslim and the secular Western view. Notable among these was Fazlur Rahman Malik (1919-1988) who argued that while the chain of transmission of the hadith may often be spurious, the matn can still be used to understand how Islam can be lived in the modern world. Liberal movements within Islam tend to agree with Rahman's views to varying degrees.
While both hadith and Qur'an have been translated, most Muslims believe that translations of the Qur'an are inherently deficient, amounting to little more than a commentary upon the text. There is no such belief regarding hadith. Practicing Muslims cleanse themselves (wudu) before reading or reciting the Qur'an; there is no such requirement for reading or reciting the hadith. Even for Muslims who accept the hadith, they are lower in rank when compared to the Qur'an.
Muslims also use the Ahadith to interpret parts of the Qur'an when verses are not clear or even when verses are clear to achieve an in-depth understanding. This process is called Tafsir.
| Name | Collector | Size |
| Sahih Bukhari | Imam Bukhari (d. 870) | 7275 hadiths |
| Sahih Muslim | Muslim Ibn al-Hajjaj (d. 875) | included 9200 |
| Sunan Abi Da'ud | Abu Da'ud (d. 888) | |
| Sunan al-Tirmidhi | al-Tirmidhi (d. 892) | |
| Sunan al-Sughra | al-Nasa'i (d. 915) | |
| Sunan Ibn Maja | Ibn Maja (d. 886) |
Sahih Bukhari and Sahih Muslim are usually considered the most reliable of these collections. There is some debate over whether the sixth member of this canon should be Ibn Maja or the Muwatta of Imam Malik, which is the earliest hadith canon but predates much of the methodology developed by the classic hadith scholars.
While there are still many traditional Muslims who rely on the ulema and its long tradition of hadith collection and criticism, other contemporary Sunni Muslims are willing to reconsider tradition. Liberal Muslims are most apt to trust the individual conscience, but there are also Salafis who demand the same freedom. The Salafis claim that the ordinary believer can trust his or her own judgment (even if he or she is not trained in Islamic scholarship) if he or she relies on Bukhari and Muslim, the commentators deemed to be sahih, and ignores the weak hadith.
Although Twelver Shi'ism is by far the largest branch of Shi'i Islam, there are various branches within Shi'ism and within each branch, various traditions of scholarship. Each branch and scholar may differ as to the hadith to be accepted as reliable and those to be rejected.
Four prominent Twelver Shi'a hadith collections are written by three authors who are known as the `Three Muhammads`. They are:
| Name | Collector | Size |
| Usul al-Kafi | Muhammad ibn Ya'qub al-Kulayni al-Razi(329 AH) | 15,176 hadith |
| Man la yahduruhu al-Faqih | Muhammad ibn Babuya | 9,044 |
| Al-Tahdhib | Shaykh Muhammad Tusi | 13,590 |
| Al-Istibsar | Shaykh Muhammad Tusi | 5,511 |
Unlike Akhbari Twelver Shi'a, Usuli Twelver Shi'a scholars do not believe that everything in the four major books are sahih. Every hadith must be individually examined through the process of ilm-ul-hadith. Any hadith that conflicts with the Quran or logic is excluded. Nizari Ismaili have a book of speeches of Ali called Qalam-e-Mowla. For Mustaali Ismaili, a book of hadith called Daim al-Islam narrates events of the Imams of the Fatimid Empire.
The principal hadith collection accepted by Ibadis is al-Jami'i al-Sahih, also called Musnad al-Rabi ibn Habib, as rearranged by Abu Ya'qub Yusuf b. Ibrahim al-Warijlani. A large proportion of its narrations are via Jabir ibn Zaid or Abu Yaqub; most are reported by Sunnis, while several are not. The total number of hadith it contains is 1005, and an Ibadi tradition recounted by al-Rabi has it that there are only 4000 authentic prophetic hadith. The rules used for determining the reliability of a hadith are given by Abu Ya'qub al-Warijlani, and are largely similar to those used by Sunnis; they criticize some of the companions, believing that some were corrupted after the reign of the first two caliphs. The Ibadi jurists accept hadith narrating the words of Muhammad's companions as a third basis for legal rulings, alongside the Qur'an and hadith relating Muhammad's words.
... it is not surprising that, among the hotly debated controversial issues of Islam, whether political or doctrinal, there is not one in which the champions of the various views are unable to cite a number of traditions, all equipped with imposing isnads
The next generations of Western scholars were also sceptics, on the whole: Joseph Schacht, in his Origins of Muhammadan Jurisprudence (1959), argued that isnads going back to Muhammad were in fact more likely to be spurious than isnads going back to the companions. John Wansbrough, in the 1970s, and his students Patricia Crone and Michael Cook were even more sweeping in their dismissal of Muslim tradition, arguing that even the Qur'an was likely to have been collected later than claimed.
Contemporary Western scholars of hadith include:
Madelung has immersed himself in the hadith literature and has made his own selection and evaluation of tradition. Having done this, he is much more willing to trust hadith than many of his contemporaries.
Some quotes:
work with the narrative sources, both those that have been available to historians for a long time and others which have been published recently, made it plain that their wholesale rejection as late fiction is unjustified and that with [not without] a judicious use of them, a much more reliable and accurate portrait of the period can be drawn than has been realized so far.
the mere fact that ahadith and asanid were forged must not lead us to conclude that all of them are fictitious or that the genuine and the spurious cannot be distinguished with some degree of certainty.
The current research on the life of Muhammad is characterized by the fact that two groups of researchers stand directly opposed to one another: The one group advocates, somewhat aggressively, the conviction that all transmitted traditions, in part because of great inner contradictions, legendary forms, and so forth, are to be rejected. The other group is opposed to that view. According to these researchers, the Islamic transmission, despite all these defects, has at least a genuine core, which can be recognized using the appropriate source-critical methods. The difficulty certainly consists of finding criteria by which the genuine is to be differentiated from the spurious.
Ignaz Goldziher was of the opinion that most hadiths had been invented by the transmitters to justify certain opinions of their own. According to him hadiths should not be seen as authentic historical accounts. Goldzihers suggestion has been refuted to a certain level by Fuat Sezgin. According to Fuat Sezgin most Hadiths are authentic.