Ahad hadith - hadith narrated by only one narrator
some scholars claim there are around 5000 ahad hadiths out of 7200+ hadiths.
How many ahad hadiths there in Sahih al-Bukhari?
If no exact number available can I get a rough number.
I found an upper bound for the amount of ahad hadiths of both sahihs in a statement of ibn Hajjar al-'Asqalani (in his book called al-Madkhal fi osol al-Hadith المدخل في أصول الحديث) saying it is around 200 hadiths, commenting a "False" statement of al-Hakim:
وقد رد عليه الحافظ حجر بقوله : ... "وأما قوله : إن الغرائب الأفراد ليس في الصحيحين منها شيء ، فليس كذلك ، بل فيها قدر مائتي حديث قد جمعها الحافظ ضياء الدين المقدسي في جزء مفرد" (from islamport)
--- (My own translation take it carefully)--- ibn Hajjar has replied this view saying: "and about his statement that "there are not any gharib (the pl. gharaaib غرائب is used in the original text) of the single hadiths in both sahih, this is not correct, but there are about 200 hundered which are compiled by al-Hafidh Diya' ad-Dyn al-Maqdisi (a hanbali scholar) in one volume".
Note that this count is including what you meant by ahad as defined at the time of ibn Hajar: single narrator as well as more than one narrator.
Ahad with single narrator in any level of the chain is called gharib غريب and sahih al-Bukhari opens and closes with one hadith of this kind. A third hadith of this kind I could find is this one. So we have a lower bound of at least three ahadith of this kind.
It seems to me that the definition of ahad didn't find a consensus in the early centuries so the only thing they were in consensus about is that ahad means that a hadith didn't reach tawatur, as they had different opinions about tawatur (10 chains seems to be accepted later) this consensus was still a bit waggly. So now we have three kinds of ahad:
They also classify gharib so there are:
At the beginning of the development of the hadith sciences different terms were in use so if you find Imam at-Tirmidhi qualifying a hadith as gharib (one of the Arabic synonyms of gharib is strange) than he might point at a "strangeness" either in the content (that a hadith is widely known with an other wording, but in a particular chain it has the "actual" wording) or in the narrator chain (a weakness or that the chain is unique). So at-Tirmidhi may point at an ahad by adding the the statement "لا نعرفه الا من هذا الوجه" in his qualification which indicate that he know the particular hadith only with this chain (sanad) or/and content (matn).
Please note that even if you find in sahih al-Bukahri a count of 7300-7400 hadiths that doesn't mean that they are totally different. Al-Bukhari and (especially) Muslim and many other scholars used to repeat a hadith quoting different chains (Muslim often make a link to another chain if there is a common part) also in cases a repetition was necessary because a hadith also fitted into an other chapter. So according to this scholar site the sum of sahih ahadith in both sahih books is around 2980 hadiths only. This scholar assumes that there are around 4400 sahih hadith based on similar statments of many hadith scholars like Imam Ahmad, Sho'aba, at-Thawri, Yahya ibn Sa'id al-Qatan, ibn al-Mahdi and others. It would be interesting to know whether ibn Hajar's quote includes hadith repetitions also!
See also this fatwa in Arabic explaining ahad.
And the Arabic wikipedia site about ahad hadith.
I'm still trying to find out if the book of ad-Diya' al-Maqdisi still exists or to find a book which helps me finding the ahad or more exactly gharib ahadith from sahih al-Bukhari only to reduce the bounds. Any help on that subject is welcome!