|
Shaykh al-Islam Abu Zakariyya al-Nawawi (raḍia-LLahu ʿanhu) said in the
'Adhkar':
"The ulama among the experts in ḥadīth and the
experts in law and others have said: it is permissible and
(also) recommended that the religious practice (al-ʿamal)
concerning good deeds and good character (al-fadā'il),
encouragement to good and discouragement from evil (al-targhib
wa al-tarhib) be based (even) on weak ḥadīth (bi al-ḥadīth al-
daʿīf) as long as it is not forged (mawduʿ).
As for legal rulings (ahkām) such as what is permitted and what is forbidden, or the modalities of trade, marriage, divorce and other than that: one's practice is not based upon anything other than sound (sahih) or fair (hasan) ḥadīth, except as a precaution in some matter related to one of the above, for example, if a weak ḥadīth was cited about the reprehensibility (karahat) of certain kinds of sales or marriages. In such cases what is recommended (al-mustahabb) is to avoid such sales and marriages, but it is not obligatory (la yajib)."
Disagreeing with this Abu al-ʿArabi al-Maliki said:
"Absolutely
no practice is based on weak ḥadīth." [Also, Ibn Taymiyya was of the opinion
that no ruling of mustahabb can ever be based on a weak ḥadīth.]
I have heard my Shaykh (Ibn Hajar al-ʿAsqalani) insist on the following, and he put it to me in writing himself:
"The conditions for religious practice based on weak ḥadīth are three:
This is unanimously agreed upon (muttafaqun ʿalayh):
1 - That the weakness must not be very strong (ghayr shadid). This excludes
those ahadith singly recorded by liars or those accused of
lying, and those who make gross mistakes.
2 - That there be a general legal basis for it. This excludes
what is invented and has no legal basis to start with.
3 - That one not think, while practicing on the basis of it, that
it has been established as true (an la yaʿtaqida thubutahu).
This is in order that no words which the Prophet ﷺ did not
(verifiably) say be attributed to him."
He continued:
"The last two conditions are from Ibn ʿAbd al- Salam and his companion Ibn Daqiq al-ʿId; Abu Saʿeed al-ʿAla'i (specialist in forgeries) reported unanimity over the first one."
I say: It has been reported from Imam Ahmad that one may practice on the basis of the weak ḥadīth if there is no other ḥadīth to that effect and also if there is no ḥadīth that contradicts it. In one narration he is reported to say: "I like weak ḥadīth better than men's opinions."
Ibn Hazm has similarly mentioned that all Hanafi scholars unanimously agree that the school of Abu Hanifah (raḍia-LLahu ʿanhu) holds that weak ḥadīth is preferable to opinion (ra'y) and analogy (qiyās). Ahmad was asked about someone finding himself in a country with, on the one hand, a memorizer of ḥadīth (sāhib hadīth) who does not know the sound from the unsound, and, on the other, an authority in opinion (sāhib ra'y): who should he consult? He replied: "Let him consult the memorizer of ḥadīth sāhib hadīth and not the authority in opinion (sāhib ra'y)."
[Note: Some question the authenticity of the above opinion of Imam Ahmad in the light of Ibn Taymiyya's assertion:
"The one who relates from Ahmad that he used to rely [in shariʿa] upon the weak ḥadīth, which is not sahih or hasan, has erred." Qaʿida jalila p. 82.
But this does not contradict the opinions
of Imam Ahmad quoted by Sakhawi above. Even so, and even
in case the above opinions were not recognized by Ibn Taymiyya
as genuinely representative of Imam Ahmad's position, it is
clear that Sakhawi did not question their authenticity.
The truth of the matter is that Ibn Taymiyya in the "Qaʿida"
gives two mutually contradictory views concerning Imam Ahmad's
position:
see [2] 
]
Abu ʿAbd Allah Ibn Mandah reported from Abu Dawud, the author of the 'Sunan' and a student of Imam Ahmad, that Abu Dawud used to cite the chain of transmission of a weak ḥadīth if he did not find other than it under that particular heading (bāb), and that he considered it stronger evidence than authorized opinion (ra'y al-rijāl).
What emerges from this is that there are three diverging views:
- No practice is based on weak ḥadīth whatsoever (mutlaqan);
- Practice is categorically (mutlaqan) based upon it if no other
evidence is found under the same heading;
- The majority of the scholars (al-jumhur) hold that it can be
used as basis for practicing good deeds and achieving good
character (yuʿmalu bihi fi al-fadā'il) but not for legal rulings
(dūna al-ahkām). And God is the Granter of success.
b) Translated from Muhammad Zaki Ibrahim in "Usul al-wusul"
(Cairo: Azhar, 1984):
If not proven to be forged, in which case there is absolutely no truth in it, the ḥadīth [is] daʿīf (weak), although the pillars of veracity in it are not complete, nevertheless retains a part of truth.
Imam Nawawi said:
"The ulama among the muhaddithun..." [as
quoted by Sakhawi above].
I say: This is the principle adopted by the ḥadīth master (hafiz) Ibn al-Salah, as well as what we know of the imams of ḥadīth science among the early generations (salaf) such as Sufyan al-Thawri, Ahmad Ibn Hanbal, Ibn ʿUyaynah, Ibn al- Mubarak, Ibn Mahdi, and Ibn Maʿīn... Al-Khatib al-Baghdadi devoted a chapter to that topic in his 'Kifayah'.
End of translated excerpts.
Note:
There is also a detailed discussion on the topic in Nuh
Keller's translation 'Reliance of the Traveller' p. 954-957.
I recapitulate the list of ḥadīth masters who accept the use of hadīth daʿīf at the very least for religious practice related to ethics (fada'il al-aʿmal) and in some cases even for legal rulings (Ahmad, Abu Dawud, and the entire Hanafi school), according to the above three sources (Sakhawi, Ibrahim, Keller):
1- Nawawi
2- Ibn al-Salah
3- Sufyan al-Thawri
4- Ahmad Ibn Hanbal
5- Ibn ʿUyaynah
6- Ibn al-Mubarak
7- Ibn Mahdi
8- Ibn Maʿīn (forgery specialist)
9- al-Khatib al-Baghdadi in 'al-Kifayah', chapter entitled:
"strictness with regard to aḥadīth pertaining to rulings
and leniency with regard to those pertaining to virtuous
actions"
10- Bukhari as proven by his use of them in 'al-Adab al-
mufrad'
11- Ali al-Qari (forgery specialist)
12- Ibn Hajar al-Asqalani.
13- Ibn Abd al-Barr in 'al-Isaba'.
14- Ibn al-Qayyim in 'Iʿlam al-muwaqqiʿīn'.
15- Sakhawi
16- Abu Saʿīd al-ʿAla'i (forgery specialist).
17- Abu Dawud.
18- Hanafi school.
It is the Consensus of the Ulema that weak ḥadīths can be narrated and put into practice in Islam according to according to al-Bayhaqī, Ibn ʿAbd al-Barr, al-Nawawī, Ibn Taymiyya, al-Qārī, and ʿAlawī ibn ʿAbbās al-Mālikī in his manual al-Manhal al-Laṭīf fī Maʿrifat al-Ḥadīth, provided certain conditions are met.[4]
Ibn al-Ṣalah, al-Nawawī and al-`Irāqī's sole conditions werre that
(1) the ḥadīth be related to good deeds (faḍā'il al-aʿmāl)
without bearing on legal rulings and doctrine and
(2) the ḥadīth not be forged.
After them, Ibn Daqīq al-ʿīd, al-Zarkashī, and Ibn Ḥajar added three furhter conditions: that the ḥadīth not be very weak;[5] that it be subsumed under a principle already established in the Law; and that one not positively believe that the Prophet ﷺ said or did it.[6]
Ibn al-Mubārak said: "One may narrate from [a weak narrator] to a certain extent or those ḥadīths pertaining to good conduct (adab), admonition (mawʿiẓa), and simple living (zuhd)."[1]
This conditional rule for narrating - and practicing - weak ḥadīths is in conformity with the unanimous view of the Salaf who permitted their use in faḍā'il al-aʿmāl as opposed to ʿaqīda or the rulings pertaining to ḥalāl and ḥarām. This is stated or practiced by Sufyān al-Thawrī, Ibn ʿUyayna, ʿAlī ibn al-Madīnī, Yaḥyā ibn Maʿīn, Aḥmad, ʿAbd al-Raḥmān ibn Mahdī, Ibn Abī Ḥātim, al-Bukhārī in al-Adab al-Mufrad, al-Tirmidhī, and many others.[2]
Ibn al-Ṣalah said in his ʿUlūm al-Hadīth:
"Know that the forgery is the very worst of the weak ḥadīths and that it is not licit for anyone who knows a ḥadīth is forged to narrate it in any sense whatsoever except by showing, at the time, that it is forged, contrary to other types of weak ḥadīths, which are possibly true in an unapparent way. It is permitted to narrate the latter in [matters of] encouragement [to good deeds] and deterrence [from evil ones.]…
Among the experts of ḥadīth and others than them, it is allowed to lower the standards in the transmission chains and to narrate all kinds of weak ḥadīth other than the forgeries without attention to showing that they are weak except with regard to the Divine Attributes and the rulings of the Law in the licit and the illicit and other [rulings] besides these two. This is the case, for example, in exhortations and [didactic] storytelling, meritorious deeds, all the varieties of encouragement and deterrence, and all that is unconnected with legal rulings and doctrinal beliefs. Among those from whom we narrate such a stipulation are ʿAbd al-Rahmān ibn Mahdī and Ahmad ibn Hanbal - Allah be well-pleased with both of them!"
This [above-cited] rule was mentioned by Ibn al-Ṣalāḥ and others in Maʿrifat ʿUlūm al-Ḥadīth and its commentaries.[3]
..........
The dissents reported from Yaḥyā ibn Maʿīn, al-Bukhārī, Muslim, Ibn Ḥazm, and Ibn al-ʿArabī al-Mālikī and al-Shawkānī are inaccurate. The correct position of Imām Muslim in the introduction to his Ṣaḥīḥ is that he forbade the use of forgers and other abandoned narrators, not of truthful weak ones, in conformity with the position of Aḥmad and the rest of the Salaf.[7]
Muslim also says: "The sound reports from the trustworthy (thiqāt) narrators and those whose reliability is convincing are more than that we should be forced to transmit reports from those who are not trustworthy and whose reliability is not convincing."
The difference is clear between saying we are not forced to use weak narrators and saying that one cannot transmit anything from them. A proof of this is his use of the weak narration from ʿA'isha: "Treat people according to their ranks" and the fact that his strictness in narrators drops a notch or two in the ḥadīths of raqā'iq or faḍā'il al-aʿmāl in the Ṣaḥīḥ, as in the case of Shaddād ibn Saʿīd Abū Ṭalhā al-Rāsibī or al-Walīd ibn Abī Walīd, as does al-Bukhārī's in his also.[8],,pAl-Bukhārī uses an even lesser criterion for narrations pertaining to Faḍā`il in his Adab al-Mufrad.
The claim about Ibn Ma`īn began with Ibn Sayyid al-Nās in his introduction to `Uyūn al-Athar and was imitated by others since, although it is contradicted by the early sources we cited.
The similar claim that Ibn al-ʿArabī was opposed to the use of weak ḥadīths in absolute terms is put to rest by his own statement about a certain weak ḥadīth: "Its chain is unknown, but it is preferable to put it into practice…"[9]
As for Ibn Ḥazm's statement against the use of weak narrations in absolute terms:[10] he elsewhere states preferring the use of weak ḥadīth over the use of juridical opinion (ra'ī), as does Ibn al-ʿArabī himself.[11] Finally, al-Shawkānī in Nayl al-Awṭār recommends putting into practice the ḥadīth on the preferable timings of cupping despite its severe weakness.
And Allah knows best.
[1] Narrated by Ibn Abī Ḥātim in Muqaddimat al-Jarḥ wal-Taʿdīl (2:30) and cited by Ibn Rajab in Sharḥ ʿIlal al-Tirmidhī (1:73). 
[2] Cf. al-Khaṭīb, al-Kifāya (p. 162-163=133-134), Ibn Abī Ḥātim, Muqaddimat al-Jarḥ wal-Taʿdīl (2:30-38), Ibn Rajab, Sharḥ ʿIlal al-Tirmidhī (1:73), Ibn Ḥajar, end of al-Nukat ʿalā Ibn al-Ṣalāḥ (2:887-888), al-Suyūṭī, Tadrīb al-Rāwī, al-Lacknawī, al-Ajwiba al-Fāḍila, etc.
[3] Ibn al-Ṣalāḥ, ʿUlūm al-Ḥadīth (p. 93=1984 ed. p. 103).
[4] Al-Bayhaqī, Dalā'il al-Nubuwwa (1:33-34); Ibn ʿAbd al-Barr, al-Tamhīd (1:127); al-Nawawī, al-Majmūʿ (5:63), Irshād Ṭullāb al-Ḥaqā'iq (p. 107-108), Sharḥ Ṣaḥīḥ Muslim (introduction), and al-Adhkār (introduction p. 5) cf. Ibn ʿAllān, al-Futūḥāt al-Rabbāniyya (1:84); Ibn Taymiyya, Sharḥ al-ʿUmda (1:171), Majmūʿ al-Fatāwā (18:26, 18:65-66), and Miswaddat āl Taymiyya (p. 233, 246, 461); al-Qārī, Sharḥ al-Shifā' (2:91) and Mirqāt al-Mafātīḥ (2:381); ʿItr, Manhaj al-Naqd (p. 291-296) and Us.ūl al-Jarḥ wal-Taʿdīl (p. 140-143).
[5] Even so, al-Sakhāwī said in al-Qawl al-Badīʿ (p. 432) of a certain ḥadīth: "In sum, it is a very weak ḥadīth (ḍaʿīf jiddan) that is written in meritorious deeds (yuktabu fī faḍā'il al-aʿmāl), but as for its being forged, no, it is not [forged] ."
[6] Cf. l-Sakhāwī, al-Qawl al-Badīʿ and al-Suyūṭī, Tadrīb (p. 196).
[7] Cf. al-Nawawī, Sharḥ Ṣaḥīḥ Muslim (introduction), Ibn al-Qayyim, Iʿlām al-Muwaqqiʿīn (1:31), al-Sakhāwī, al-Qawl al-Badīʿ (p. 474), and ʿItr, notes on Ibn Rajab's Sharḥ ʿIlal al-Tirmidhī (1:75-76).
[8] The claim of a handful of authors such as al-Qāsimī in Qawāʿid al-Taḥdīth (p. 94) or ʿAjāj al-Khaṭīb in Uṣūl al-Ḥadīth (p. 231) that Ibn al-ʿArabī and Ibn Maʿīn were opposed to the use of weak ḥadīths in absolute terms, stems from good faith in Ibn Sayyid al-Nās, al-ʿIrāqī, al-Sakhāwī, and al-Suyūṭī's claims to that effect.
[9] Ibn al-ʿArabī, ʿāriḍat al-Aḥwadhī (10:205) cf. Fatḥ al-Bārī (10:606) as cited by Muḥammad ʿAwwāma in his marginalia on al-Qawl al-Badīʿ (p. 472).
[10] Ibn Ḥazm, al-Fiṣal fīl-Milal wal-Niḥal (2:83=2:69).
[11] Cf. Ibn Ḥazm, al-Iḥkām (6:225-226) and Ibn al-ʿArabī, al-Maḥṣūl (p. 98) and Marāqī al-Zulaf as cited in Ibn ʿArrāq, Tanzīh al-Sharīʿa (2:209-210).
[12] This chapter has been printed 2005 in Sunnah Notes, vol1., p.100,
see: Texts, Books, Videos By Shaykh Gibril F. Haddad
The chapter has been updated to the text in Sunnah Notes as correct as possible, but without a review of the footnotes.
Also see below, the video on the science of ḥadīth or ḥadīth nomenclature, ”Bayquniyya Poem on Ḥadīth Terminology.”
Bayquniyya Poem on Ḥadīth Terminology by Sh GF Haddad, 2 parts
vs.3.2