This is reproduction of APPENDIX II from Shaykh Muhammad Mustafa Azmi’s work “Studies in Early Hadith Literature” Suhail Academy, Lahore 2001 pp. 301-305
The Problem of Enormous Numbers of Hadith
بسم الله الرحمن الرحيم الحمد لله وحده و الصلاة و السلام على من لا نبي بعده و على آله و أصحابه أجمعين
There are references to hundreds of teachers from whom al-Thauri, Ibn al-Mubarak, al-Zuhri, etc. had written ahadith. In the works of biographers we find a long list of teachers and students of eminent scholars. There are at least fifty students of al-Zuhri who made their written collections from him. If, on average, every one of them had written only five hundred traditions from him, then this number would have been 25,000. If we go a step further and assume for example that every student of al-Zuhri had only two or three students, then this numbers of traditions might have increased at the end of the second century to some 75,000, and in the time of Bukhari and his contemporaries they have been in hundreds of thousands.
Thus, the number of a few thousand ahadith reached about three quarters of a million in the mid third century.
According to Ibn Hanbal’s statement, over 7,000,000 traditions were sound, of which 6,000,000 were memorized by Abu Zur’ah. 
Al-Bukhari claimed that he made his collection of traditions out of six hundred thousand. His book contains only 7,397 Hadith with repetition, and only 2,602 Hadith without repetition.
The actual number of traditions preserved in the Sihah and other collections is only a small fraction of the body of the traditions described above. This is a puzzling problem. Many scholars have been perplexed, and so have reached very strange conclusions. Guillaume says, “Bukhari’s biographer says that he selected his material from no less than 600,000 hadith. If we allow for repetitions which occur under different heads, he reduced this vast number of forgeries or dubious reports to less than 3,000 Hadith. In other words, less than one in every 200 traditions which circulated in his day could pass his test.”
The problem consists of (a) Hadith and (b) enormous numbers and their implications.
(a) Hadith in the terms of some traditionalists, means utterance, deeds and tacit approval of the Prophet, while in definition of other scholars it covers utterances, deeds, legal decisions and tacit approval of the Prophet as well as those of Companions and the Successors.
(b) As for the problem of enormous numbers, every channel of transmission is counted as a separate hadith. ‘Abd al-Rahman b. Mahdi (d. 198) says, “I have thirteen traditions from Mughriah transmitting from the Prophet, concerning al-mash ‘ala Khuffain.”
It is quite obvious that al-Mughirah is reporting single action or habit of the Prophet. It does not matter how many times this action was repeated. It would be reported as a single action. As this single action is reported to ‘Abd al-Rahman b. Mahdi from thirteen channels, he counts them as thirteen traditions.
The first four centuries of the Hijrah were the golden age of the science of tradition, and the number of transmitters grew tremendously. Ibn Khuzaimah (d. 311) gives some thirty isnad for one hadith is one chapter, concerning the single act of ‘Aishah for cleansing the cloth. Meanwhile it is obvious that there might have been many other channels of transmission which were unknown to him. Muslim bin al-Hajjaj (d. 261) cites the names of a great number of transmitters, when he argues about certain points, especially when there is a mistake committed by some transmitters. For example, he gives thirteen traditions concerning the single incident of Ibn ‘Abbas and his tahajjud prayer. In the prayer, he stood on the left of the Prophet and then Prophet pulled him to his right side. Yazid b. Abu Ziyad related on the authority of Kuraib that Ibn ‘Abbas stood on the right side of the Prophet, but he was placed on the left. On this occasion Muslim gives thirteen isnads – making thirteen Hadith – contradicting Yazid’s statement. Furthermore, he does not give the complete isnad and their full growth until his time. He mostly gives the details of channels until about 130 A.H. had he given the complete comprehensive isnad flourishing in his own time, they might have grown to fifty traditions at least.
Growth and development of isnad in the third century
There have been some traditionalists who claim that they had every Hadith from one hundred channels, and many others who have written every Hadith from twenty or thirty channels. So, we may now infer what the real numbers of the traditions were which were described as 600,000. Another point is that they were not purely traditions of the Prophet, but the sayings of the Companions and the Successors and their legal decisions as well; the word ‘Hadith’ covers all these subjects and matters in some scholars’ terms.
The True Numbers of the Traditions
What is the real number of authentic traditions? The exact number is unknown, but according to Sufyan al-Thauri, Shu’bah, Yahya al-Qattan, ‘Abd al-Rahman b. Mahdi and Ibn Hanbal 4,000 Hadith only. The statement is incomprehensible. Gilani is inclined to a number less than 10,000 Hadith, based in his statement, on the quotation from Tahrir al-Jaza’iri, who in turn was quoting al-Hakim al-Nishaburi. This is apparently a misinterpretation of al-Hakim’s attitude. He gives an estimate of less than 10,000 Hadith for the first-class authentic traditions which are transmitted according to al-Bukhari and Muslim b. al-Hajjaj’s stipulation. Furthermore he himself objects to this number, saying, “How can it be said that this [Prophet] traditions do not reach 10,000 traditions when 4,000 Companions… have transmitted traditions from him, who associated with him for more than twenty years…”
It is said that the Musnad of Ibn Hanbal consists of some 40,000 Traditions, and without repetitions it would be about 30,000, but there has not been any research so far.
Al-Bukhari, his Sahih and other Traditions
Al-Bukhari did not claim that what he left out were the spurious, nor that there were no authentic traditions outside his collection. On the contrary he said, “I only included in my book al-Jami’ those that were authentic, and I left out many more authentic traditions than this to avoid unnecessary lengths.” He had no attention of collecting all the authentic traditions. He only wanted to compile a manual of Hadith, according to the wishes of his Shaikh Ishaq b. Rahwaih, and his function is quite clear from the title of his book “Al-Jami’, al-Musnad, al-Sahih, al-Mukhtasar, min umur Rasul al-Allah wa Sunanihi, wa ayyamih.” The word al-Mukhtasar, ‘epitome’, itself explains that al-Bukhari did not make any attempt at a comprehensive collection.
Now it is clear that when traditionalists give enormous numbers for the traditions, they mean channels and sources of their transmission, and do not mean real numbers of Hadith. But when they give small figures, saying: “Al-Zuhri has 1,000 Hadith, or al-Qasim has 200 Hadith” they most probably mean Hadith as a subject matter not counted according to its isnad.
Does ‘Unauthentic’ Mean a False Statement?
Traditionalists, at first, look into the isnad and if it is defective, they call the Hadith defective, without scrutinizing the subject matter; because a Hadith, according to their criteria, cannot be authentic unless both its parts are perfect.
Authentic matter with false isnad is a false statement. This will be clear from – [Nuskhah of Zubair b. ‘Adi] -- which is a collection of traditions, transmitted by Bishr b. Malik on the authority of Zakariya b. ‘Adi from Anas b. Malik from the Prophet – is called spurious, though about one quarter of the traditions of this collection are found in Bukhari and Muslim’s Sahih collections, and are called authentic. The only reason for discarding them is that it is maintained that Zakariya did not hear all these traditions from Anas, and they are falsely attributed to him.
Therefore, if the scholars say that 200,000 Hadith were not authentic, it does not mean that they spurious. It only means that there isnads are questioned while the subject itself may or may not be false.
(Azmi, Muhammad Mustafa, Studies in Early Hadith Literature, Suhail Academy, Lahore 2001 pp. 301-305)
Orientalist Dr. James Robson also mentioned this in his introduction to translation of Mishkat al-Masabih.
"Bukhari, in compiling his work, said he had collected 600,000 traditions, and yet he included only 7,275 altogether in his Sahih, a total which is said to be reduced to 4,000, or even 2762 when repetitions are eliminated. Abu Dawud out of 500,000 traditions included only 4,800 in his Sunan, and of these he said that some were not sound. It should be understood, however, that when one speaks of 600,000 traditions, for example, this does not mean that number of separate items of information. Each tradition has two parts, the isnad (chain of authorities through it is transmitted) and the matn (text). If therefore we found the same text with, say, three different isnads, that would be considered to represent three traditions."
(James Robson, Dr., Mishkat al-Masabih- English Translation with Explanatory Notes, Sh. Muhammad Ashraf Publishers, Lahore, 1991, Vol.1, vii)
 Al-Hakim, Madkhal, 13.
 Al-Kathib, Tarikh Baghdad (Bagh.), ii, 8,14. Other traditionalists also gave an enormous number which they memorized or wrote down. I discuss only one case of al-Bukhari to clarify the problem.
 Ibn Hajar, Hady al-Sari, as quoted by al-Siba’i in Sunnah, 501.
 Guillaume, Islam, 91; a similar idea is maintained by Ahmad Amin, Fajr al-Islam, 211-12; Muir, Mahomet, xxxvii; see also J. Robson, Traditions in Islam, M.W., vol.xli, pp.101-2; Nicholson, A Literary History of the Arabs, p.146; Gibb, Mohammedanism 79, Haikal, Hayat Muhammad, p.49
 Tahanwi, Kashshaf, 279; Qasimi, al-Tahdith 61; Suyuti, Alfiyah 3; Subhi, Mustalahat 3; Sakhawi, Mughith 4.
 Jurjani, Risalah, 1; see Tahanwi, Kashshaf, 279; Suyuti, Tadrib, quoting al-Tibi 6; Sakhawi, Mughith, 12 “predecessors called them Hadith”; Gilani, Tadwin 62; for early usage of this word for the sayings other than the Prophet see Hasan b. ‘Imarah’s discussion with al-Zuhri, Islam, v, 147
 Razi, Introd. 261
 Muslim, Tamyiz, fol. 6b-7a; for more examples see fol. 10a; 11 channels; fol. 11b; 17 channels; these isnads are shown in the mid-second century as they flourished , not at the time of Muslim, who was a century later.
 Siba’i, Sunnah, 224, quoting sayings of Ibrahim b. Sa’id al-Jauhari from Tanib al-Khatib.
 Madkhal, 9; and it is quite possible, at least 50 students transmitted al-Zuhri’s book, so within 25 years’ time his traditions might have grown 30 or 40 times, see also Majruhin, 10a; Jami’ 165a; Mizan, I, 35
 Rashid, Ibn Majah, 164, quoting al-Amir al-San’ani
 Gilani, Tadwin, 66-67
 Madkhal, 11-12
 Shakir, Commentary on Suyuti’s alfiyah, pp.218-222, Shakir gives the estimate of the early scholars of 30,000 to 40,000; perhaps the first number indicates traditions without repetition.
 (Lately Musnad Ahmad has been published with proper continuous Hadith numbering. The total number of hadith according to al-Resala ed. is 27647. However if we count each channel separately the number will be quite large. See Hadith No. 535, 2008 etc – Waqar A. Cheema)
 Baji, 9b; also Ibn Hajar, Hady al-Sari, I, 18; also Bagh., ii, 8-9
 Ibn Hajr, op. cit., 18; Bagh. Ii, 18
 Ibn al-Salah, ‘Ulum al-Hadith, 24-5
 See for the priority of isnad criticism, Robson, Materials of Tradition, M.W., vol. xli, p.166; Guillaume, 55