THE STRUCTURE OF AUTHORITY IN IMMZ

SHI CT JURISPRUDENCE

NORMAN CALDER

PhD

SCHOOL OF ORIENTAL AND AFRICAN STUDIES

RI..

LO.U1N.
UFI.

NORMAN CALDER: THE STRUCTURE OF AUTHORITY IN IMMI SHI

JURISPRUDENCE.
ABSTRACT
This work is concerned with the problem of
authority in Islamic jurisprudence. Its main aim is to
analyse and interpret the structure of authority implicit
in imbff Shii juridical works from the tenth to the
sixteenth centuries. It begins however with a consideration of early Sunni developments.
Chapter one (based on early Sunni juridical texts)
demonstrates the emergence of a coherent theory of clerical
authority based on custody of revelation, while chapter two
illustrates the (later) elaboration of a compromise between
clerical authority and princely power (Tabari to MwardI).
Chapter three analyses the implications of the theory of
the Ghayba as presented in ImmI polemical works of the
Buyid period. It is suggested that the theory of the Ghayba
may be interpreted as a re-assertion of clerical authority
and a re-alignment of clerical relations with political
powers.
Chapters four to six are concerned with the
conflict betwenclerical authority and de facto power, as
manifest in Imami works of' furt. The nature and development of clerical claims are discussed with reference to
qac', zak5t, khums, jihd and salt al-jum'a. In all
these fields the legitimate authority of the imm is seen
to be delegated to the fuqah'.
Chapters seven to nine (based on ImmT works of
ul al-figh) are concerned with the right of the clerical
class to define and articulate the law. Chapter seven, on
exegesis and consensus, considers the significance of the
scholarly apparatus for justifying the law. Chapter eight
looks at the role of reason as an independent source of
ethical values. Chapter nine is concerned with the
development of the Imml theory of ijtihd. That theory,
it is suggested, was intended to justify the authority and
the creative activity of the fugah'.
There is an appendix which provides the dates
and biographical details of the ImmI writers whose works
form the basis of this study.

A CKNOWLEDGEMEI'ITS

I should like to express my gratitude to all

those who have made it possible for me to write this thesis.
My debt to my supervisors, Professor A. K. S. Lambton and
Doctor John Wansbrough, for their advice, criticism and
encouragement, is immense. I regret that this work still
falls far short of the standards they represent. The
Central Research Fund of the Un!versity of London and the
Scholarships Committee of the School of Oriental and
African Studies are to be thanked for financial help towards
a period of study in Iran. I should like to thank those
who provided help and hospitality during that stay. I owe
too much to too many to define here the extent of my debt
but would mention in particular the insight and encouragement of A. A. Sharif and Dr. Stefan Sperl. For the support
and forbearance of my family, many thanks.

CONTENTS

Part One
Chap I
Chap II
Chap III

The Significance of the term Irnm.

The Imamate of the Abbasids.
The Absent Imrn.

1
24
J49

Part Two
Chap IV

Chap V

Chap VI

The Judicial Delegation.

Community Finances.

108

The General Delegation.

l4 7

Exegesis and Consensus.

171
206
223

66

Part Three
Chap VII

Chap VIII Reason and Revelation.

Chap IX
Doubt and Prerogative.
Appendix

214)4

Bibliography

2148

PART I

1
CHAPTER I
THE SIGNIFICANCE OF THE TERM IMM
Islamic juridical literature emerged towards the end
of the eighth century (AD) already characterised by acknowledgement of dispute and by appeal to the authority of precedent. Dispute existed within as well as between the local
schools and related both to the substantive content of the
law and to the identification of authoritative precedents.
Continued polemic (prior to as well as posterior to the
first appearance of a body of juridical literature) promoted
refinements in juridical techniques and terminology out of
which were moulded the classical expressions of Islamic law.
Central to all Islamic juridical discussions was the term
imm which though variously applied always denoted authority.1
The source or originator of an acknowledged norm
or sunna 2 was designated lmm. The close relationship between
imm and sunna is clearly manifest in this line from Labid's
mu 1allaga:

j4'
Belonging to a people whose forefathers have
laid down for them a sunna;
Every tribe has a sunna and its imm.3
That may be taken as a reference to imm as (presumed)
In religious literature
source of tribal practice (sunna).
the irnm was the source or precedent whereby to establish
religious law (also sunna, but more specifically sunnat al1.

The emergence of juridical literature is best described

in Schacht, Origins; for the domination of precedent, see

2.

also Wansbrough, Milieu, chap. II.

For this term see Schacht, Origins, 58-81; Goldzlher,
Studies, II, 11_27/2 14_37; Bravmann, Background, 123-198.

3. ZawznI, Sharla, 191.

. But cf. Wansbrough, Milieu, 32-39.

2
nab, the sunna of the prophet.) In Qur'n 2.l2 L God says
to Abraham: I shall make you an imni for the people,
ci, Tabar interpreted that as meaning:
(ji3
I shall make you for the people an imm who provides leadership and guidance; (
jL
c,
He continued:
God meant by the phrase, innT Jadiluka etc.:

ci)

I shall make you such that you lead those

who come after you who believe in me and my
prophets; you will go before them and they
will follow your guidance and they will find
their sunna in your sunna (yastannna bisunnatika) which you perform in accord with
my command and my revelation.5
An imm was a person whose actions became models of right
conduct for later generations: his practice, surina, was
to be the practice also of the people who followed him. In
the Qur'n the function of imm, model of correc,t conduct,
was peculiarly appropriate to prophets. In Q21.73 God says
with reference to the prophets Abraham, Lot, Isaac and Jacob:
We have made them imms who guide by our command and we have
inspired them to do good actions, to perform prayers and to
bring zakt. 6 Such quranic usage reflects the common JudaeoChristian assumption that a prophet is law incarnate; his
words and actions were manifestations of God's law, to be
obeyed and imitated.
That an imm was a person whose actions or commands
served as authoritative precedents even after his death is
further evident from a poem, cited in the Murj al-Dhahab of
Nasdi (d. 36/956). The poem, related apropos of the
events of Siffin, is ascribed to a follower of 'All.

LL J
A

We accept the Ixukin al1.h and nothIng else;

5. Tabarl, Tafslr, III, 18.
6. Cf. also Q32.24.

3
We accept God as Lord and the Prophet and
the Recitation
And we accept the brave
our imam;

one,

the guide, 'All,

We accept him [whether he is] dead or alive,

for he is
The irnm of guidance as to the status of
command and prohibition.7
The commands and prohibitions of an imm are permanent models
whose value and significance continue after his death. In a
similar way the author of the K. al-Irj'ascribed to kiasan
ibn Muhammad ibn al-Hanafiyya (d. c. 100/719), 'accepted' Abfl
Bakr and 'tJmar as im g ms and denied the authority of the
subsequent ahi al-furga (the people of division).

1z

Jb

L)15)J/

)' t >L Lf L+-

LLIL()

6j,

'S 1.)

j,4' i.)'
We accept Ab Bakr and 'Umar as imms; we
accept that they be obeyed and we deny that
they be disobeyed and we oppose on their
behalf those who oppose us. We suspend judgemerit on [the imms who were] the first
aM al-furqa.
Varied recognition and acceptance of diverse irnms was one
of the means whereby sectarian or partisan groups defined
themselves.
A similar usage may be detected in the juridical
works of Shfi'l Cd. 20 14/820), notably in the K. Ikhti1f
Mlik w'al-Shfi'l. 9 In that work the i1edinese justified
their substantive law by referring to the precedents
'7.

Mas'dl, Murj, II, 410.

8.

Van Ess, Das K. a1-'Ir', 23.

9.

This is one of the works contained in Shfi C I 7 s Urnm, VII;

it is designated by Schacht, Treatise III; see Schacht,
Origins, Appendix II.

4
established by various Companions and Successors, particularly
Ctjmar and Ibn CUmar but including also such late authorities
as the Umayyad caliph Curnar ibn t Abd al_cAzIz. The RshidIn
as sources of law were given no particular pre-eminence.
In spite of this, the work contains numerous generalisations
to the effect that Medinese law is based on the practice of
the imms, Abi Bakr, cUmar, 'Uthmn and, sometimes, 'AlT.
Schacht noticed the incongruity and concluded that ShfidI
had "forced the concept of the practice of Ab Bakr, tUmar and
on the Medinese) 0 It is clear that the term
a'imma would be perfectly appropriate if it were not glossed
(or sometimes replaced) with the names of the first three
This attempt to restrict
or sometimes all four Rshidln.
the denotation of the term is an interpolation, possibly by
Shui'T but more likely by a later redactor. In some contexts where the term has remained unglossed its wider
application is clear. For example the references to a'immat
al-ns and a'imrna on p. 188, lines 114 and 17 relate to a
long discussion in which the Medinese refer to the practice
or opinions of 'Umar ibncAbd al-'AzTz, 'Umar ibn al-Kha1b
and Ab Hurayra. All of these were amongst the imms of
the Medinese. Likewise in the discussion on pages 2 140 and
2141 the term a'imma was left unglossed and refers clearly
to all the authorities cited by the Mediriese. The original
application of the term was to all those who for the Medinese
constituted models of correct conduct.
Though the pre-eminence of the first four caliphs
had become an established belief amongst Muslims by the
first half of the ninth century 11 the term im gm was not even
then restricted in its application merely to those four. Ibn
Qutayba (d.276/889) dealing with the problem of how to define
and justify the law stated that Ab Bakr, 'Umar, 'tjthmn,
C All, Ibn f'ias'd and Zayd ibn Thbit were all imms and
constituted for Muslims a mafzaC: that is, in legal matters
not specIfically dealt with in prophetic Traditions (adlth)
or in the Qur'n, their reasoned opinion (ijtihd al-ra'y)
10. Ibid., 24.
11. See, e.g. Wensinck, Creed, 127, 151-152, 187.

5
was authoritative. 12 Thus even late in the ninth century
the term 1mm continued to designate one whose actions or
commands or opinions were precedents that might be used as
source or justification for the law.

An extended application of the term may be detected

al-'Ilm. There Shfii discussed
in Shf1'i's K. Jim
whether consensus (ijm t ) might be found amongst the leaders
of the provincial law chools. Referring to the leading
jurists of an earlier generation, Ibn Musayyib, 'A1', Shacbi
and Hasan al-.BarI 13 Shfiti asserted to his interlocutor:
These are the ones you have made a'imrna
f l-dTn.
(Din, here, as elsewhere in early juridical literature,
refers to religious prescription or law rather than religion
in the usual English sense of the word. 15 ) Pre.-eminent
jurists of the past were thus referred to as imms. The
same usage is found in the K. Ikhtilf Nlik w'a1-Shfir.
In that work ShfitT described the factionalism of the
ancient schools of law: each school had its own leader most
of whose opinions they followed (ittib'). Mecca found its
Iiujja ('proof') in CAt, Basra had its liujja in Hasan, Kufa,
in Shacbi etc.
All of these we have mentioned are ahl turn
wa-imma fi dahrihi. and are greater than
those who come after them.16
Pre-eminent jurists of the past were recognised as people of
imma.
That meant presumably that their opinions, no less
than the acts,orders or opinions of earlier imms, were
accepted (locally) as authoritative precedents. Mlik Cd.
179/795) is reported to have stated concerning a particular
12.
13.
1, .
15.
16.

Ibn Qutayba, Ta'wTl, 29-30; of., 367-368.

For whom, see Schacht, Origins, index.
Shfii, t.Jmxn, Vu, 258, line 13; likewise 259,9.
See also below, 14-15, and 171-172.
Shfi'i, Urnm, Vu, 2146.

6
legal problem:
j

ct)

r
-'j

We do not know that any of the imms in

olden times or in recent times has
expressed an opinion on [such and such].17
He meant that for a particular detail of law he knew of no
precedent, neither from the imms, f 1-gadim, that is,
presumably, the Companions and Successors of the prophet,
nor from the imms, fT l-liadlth, that is pre-eminent lawyers
of recent times.
The law that was thus preserved and defined by
local communities in Kufa, Basra, Medina, Mecca or Syria
was presented as sunnat al-nabi, the law or practice of the
prophet. That oncept, sunnat al-nabi, was both traditional
and widespread throughout the ancient Near East; Schacht's
supposition that it was a late or secondary development
within Islam has not found general acceptance) -8 Though the
concept of sunnat al-nabl was thus common to various schools
it was a vessel of varying content: what was in fact preserved
was a normative code, elaborated by a (local) clerical elite,19
product of a particular social milieu and subject to change
and development. Schacht's designation of that local customary law as 'living tradition' may be accepted on the understanding that it 'lived' not in the lives of the people but
primarily in the discussions and polemic of the clerical
elite 20
Judgement as to what is Iiall and what is
1arm is not based on such statements as
"the people have always done this". Most
of what the people have always done is not
Iiall.
Judgement is based only on the
17. Ibid., 2149.
18. See Schacht, Origins, 58-81; opposed to his view are:
Rahman, Methodology, chaps. 1 & 2; Bravmann, Background,

123-19 8 ; Wansbrough, Studies, 52, Milieu, 81.

19. For this term see Wansbrough, Milieu, 71-72, 78, 80-82.
20. See Schacht, Origins, 58; see also Wansbrough, Milieu, 81.

7
sunna from the prophet, from the salaf
of his aib and from the fuqah'.21
In effect a law was preserved, discussed, defined and no
doubt created by a clerical group who justified the law
by relating it to the practice or opinions of the prophet,
the Companions and the fugah', all of whom, as sources or
precedents for the law, were designated imm.
Closely allied to the ancient tradition of prophet
as source of law was the notion of a Book from God which in
turn confirmed or revealed the law. In the Qur'n where
the prophets are designated imms who guide (a'irnrna yandna)
the Book of Moses is also referred to as an imm and a mercy
(ralima) 22 or as a 'guidance' for the children of Israel
(hudan li-Bani Isrdll).23 The Qur'n itself was in Muslim
literature referred to as an imm.2
The ancient schools of law in their earliest
discernible phase related their laws to Traditions (liadlth),
whether from Companions or from the prophet, only in a
casual and Intermittent way. The stronger source of law
was found In the developing code preserved, discussed and
defined in local scholarly circles. The elaboration of
justificatory precedents acquired however considerable
Impetus from the facts of abrasive contact and polemic dispute, attested mostly between, but also within, the local
schools. The social and cultural imperatives of the time
compelled the disputants to justify the law by appeal to
precedent. The existence of known or approved (or desired)
sunna suggested that the precedent existed; the intensity
of dispute ensured it would be found. The end result was
a proliferation of precedents enshrined in iadith relating
to a variety of Imms (prophet, Companions, Successors)
and some considerable confusion. The local schools continued
to define the law in accord with local tradition (sunna)
and to select justificatory precedents (Iiadlth) on the basis
21.
22.
23.
24.

Ab Ysuf contra Awz$, in Shfi'i, Urnm, VII, 320.

Qll.17; 46.l2.
Q32.23.
Wansbrough, Studies, 45; Burton, Collection, 143.

8
of what was or was not accepted in local scholarly circles.
The increasing number of Traditions in general, and of
prophetic Traditions in particular, however, undermined
their position so that it was relatively easy for Shfi'i
to suggest that their system was incoherent and arbitrary.
They justified their claim that the law was X because a
Companion had so demonstrated. But there were also Traditions
indicating that Companion A had done something different,
or that Companions B or C or D or the prophet himself had
done something different. On what basis was choice to be
made? There was naturally a tendency for prophetic
Traditions to predominate, but not without considerable
opposition.
The local schools particularly objected to the
akhbr al-kha or akhbr Id (sparsely attested Traditions)
which tended to disrupt the locally established sunna. Ab
Ysuf (d. 182/798) in a lengthy reasoned argument against
the acceptance of such Traditions himself cited a Tradition
from the prophet via a reliable man, which included the
following:

l>L LL

4Jt

:)')s\

Make the Qurn and the knowii sunna a

guiding imgm.25
By "make the sunna an imm" he meant that the locally
established code, which had emerged in the course of scholarly
discussion, should be the major criterion for defining the
law. Indeed for the ancient schools of Islamic law in
general it was precisely the sunna which dictated their
acceptance of some and rejection of other Traditions. AbU
Ysuf joined the sunna with the Qur'n as imm.
Another
party, the ahi al-kalm or Muctazills, proposed that the
Qur'n alone should be the source of doctrine and iaw.26
That view, which was a response to juridical dispute and
therefore most appropriate to the latter half of the eighth
century, precipitated the formulation of Traditions in which
25. ShfidI, tjmrn, VII, 307-308; cf., Schacht, Origins, 28-29.
26. Ibid., 14O 11l, 258-259.

9
the Qur'n was portrayed as an imam instrumental in the
Still another group
resolution of dispute or fitna. 27
proposed that Traditions alone should be the source of
law, and that they should outweigh the established sunna
of local schools. A1mad ibn Hanbal who was heir to this
group is reported to have said concerning his collection
of Traditions (the Musnad):
nJ)J y

(TU j2J

)0

LL' C^' ')

I have made this book an imm to settle

dispute concerning any sunna of the prophet.28
The known sunna, the Traditions and the Qur'n were all
referred to as imms and their function was to resolve
dispute (fitna, ikhtilf). Other groups (the Shi c a) found
their imm in the charismatic leader of the community who
was also a descendent of the prophet. It may be that the
various clerical options which located authority in the past
(prophet and ab) or in other clerics (fugah') or in
clerical tradition (sunna) or in texts which were the
prerogative of the clerical class (Qur'n and Traditions)
were posterior to and a result of ShI

formulations. It
is in any case clear that one of the features of sectarian

or juridical dispute (the distinction is not always clear)

was definition and acceptance of one or more irnms.
Some parties responded to the probler of juridical
dispute by taking up an extreme position such as that the
Book alone or prophetic Traditions alone were authoritative.
Other groups adopted a more conciliatory approach characterised by the imposition of a determinate hierarchy upon a
number of otherwise conflicting sources. A number of
examples of such hierarchic solutions may be discovered in
the K. Ikhti1f Mlik w'al-Shfi'L The writer of that work,
having criticised the MlikI system, was asked to explain
his own view (fa-taglu anta mdh?). To which he replied:
El.] As long as kitb and sunna are present
there is no alternative for those who hear
them but to follow them. Otherwise we turn
27. See e.g. Burton, Collection, 113.
28. Cit., Patton, Alimad, 19.

10
to the opinions of the a1ib ras1 a11h
or one of them.
[2a.] The opinions of the imms, Abc. Bakr,
Cumar and 'Uthmn if we turn to them for
taglid are in my view preferable.
[3.] But this [is only the case] if we do
not find a dalla in ikhti1f which indicates
the closest [of the elements] of ikhtilf to
kitb and surina; in which case we follow
the opinion which agrees with the dalla.
[2b.] because concerning the opinions of an
imm It Is well-known that people cling to
them; and he whose opinion people cling to
is more well-known than people who give fatws
[for most fatws are given privately]
and the 'mrna do not care for the opinions of
muftis as much as they care for the opinions
of an imm .... If nothing is to be found
from the imms then we turn to the a1b, who
are of such status in din that we can accept
their opinions. To follow them is better than
to follow those who come after them.
[14 . ] Knowledge is in various degrees:
i. kitb and sunna ii. ijm where there is nothing in kitb
and sunna
iii. the opinion of one/some of the ashb
where we know of no-one who opposes
that opinion.
iv. ikhti1f of the aib
v.

giys

There should not be recourse to anything other

than kitb and sunna as long as these are
present; and knowledge should be sought from
29
the higher degree.
The text here displays not one but four hierarchic
systems. Section one suggests that after kitb arid sunna
29. Shfii, tJmm, vii, 26.

11
one should turn to the aib. Sections 2a and 2b propose
an intervening layer of three primary imms; this passage
may be related to the general glossing of a'lmma already
noted above. Section three is again an Intrusive gloss,
prescribing ikhtilf as a source immediately after kitb
and sunna and before the imms. Passage four offers a
coherent system stipulating ijm' after kitb and surina,
followed by
ikhtilf and giys. None of these four
systems corresponds precisely to the classical system
enunciated in the Risla. Whether these are revisions by
ShfIdI or the products of later redactions is a problem
probably not susceptible of solution; it is sufficient
to note that if the ascription to ShfitI has any value
whatsoever these systems cannot belong to a period earlier
than his lifetime, in fact not earlier than his fame.
Logically they might be thought prior to that of the Risla,
part of the process of getting to that point. It is
however certain that the dominance of the so-called
classical system did not prevent the continuance of other
systems, of which Ibn Qutayba's is only one example. He
rejected giys but accepted the Ijtihd al-ra'y of the
recognised imms (not to be confused with the caliphs).30
The common presence of kitb and sunna in all of the systems
might again suggest the influence of Shfi
were It not
for' the fact that these two terms were already paired in
the ancient schools (though their implications may have been
different); and indeed their pairing reflects an ancient
Judaic tradition. 31 Dating then is difficult; but it
seems not unreasonable to suggest that all such attempts
to Impose order and structure on diverse juridical sources
must be posterior to the emergence of juridical dispute.
As such, the second half of the eighth century may be
suggested as the terminus a quo.
The RIsla of Ibn al-Muqaffa' (d. 142/759)
represents an early response by a member of the secretarial
class to the problem of uncertain authority, and its
30. Ibn Qutayba, Ta'wIl, 29-30, 70.
31. See inter alia Neusner, Early Rabbinie Judaism, 3-33.

12
authenticity has been accepted by Schacht and Wansbrough.32
Concerning the Ris1a attributed to Iasan al-Basrr (d. 110/
728) Wansbrough has suggested that it is a work concerned
to promote the kitb allah as a source of communal authority
superior to (and presupposing the existence and acceptance
of) prophetic and other Traditions. As such he would
consider it posterior to the emergence of juridical dispute
and posterior to the challenge of the Traditionists and so
would date it towards the end of the eighth century.33
Wansbrough has further as a result of a literary analysis
of early quranic literature (slra and tafsr) suggested
that the Qurn itself is a composite document whose compilation and canonisation (but not necessarily its component
parts) may be dated likewise to the end of the eighth
century.
It was, he suggests, the existence of dispute
and the need for resolution which prompted appeal to the
Book and so made necessary the existence of a standard
canonic version. The function of the Book as 'proof' of
the existence of the prophet Mulammad made it acceptable
to the community at large (already committed to the notion
of sunnat al-nabi) and facilitated its emergence as itself
a source of authority. 35 As a result of that argument, and
of additional internal evidence of an
dispute, Wansbrough suggests at least some degree of interpolation in
the works attributed to CAbdallah ibn Ibad (d. c. 81/700)
and H. ibn M. al-Hariafiyya. 6 His views are persuasive
but must be compared with the opposed views of Van Ess.37
Concerning the K. ai-irja' of H. ibri M. ibn al-Hanafiyya
it may be noted that the author in defining his beliefs
proposed a distinct hierarchy of sources consisting of the
Qurn (imarn), Muhammad (prophet), Ab Bakr and 'timar (imams

ulI

32. Schacht, Origins, 58-59, 95, 101-102; Wansbrough, Studies,

158-160.
33. Ibid., 160-163.
314. Ibid., chaps. I & II, esp. 143-52.
35. Ibid., 77-78.
36. Id., Milieu, 119-120.
37. Van Ess, Beginnings.

13
to the exclusion of all other imms, the ahi al-furqa).
It seems to me that such a systematisation of imms presupposes the kind of juridical and sectarian dispute evident
in the works attributed to Shfi q . It should presumably
be dated to the same period as the other similar systematisations listed above from the K. Ikhti1f M1ik wtal_ShfiCl;
that is, in (or after) the second half of the eighth century.
The exploitation of historical event in Islam for
legal, doctrinal or sectarian ends has been frequently demonstrated (Goldzlher, Peterson, Schacht, Wansbrough). Literary
texts of the late eighth and ninth centuries were concerned
not with the objective assessment of historical truth but
rather, on the whole, with justification of the present;
history was formulated as symbol. What was symbolised was
a current preoccupation. The ascription to }I. ibn M. ibn
al-Hanafiyya or to Hasan al-Basri of particular

ulI

positions may be understood as the provision of a suitable

pedigree for ideas whose relevance was to the late eighth
century. Wansbrough has noted, with regard to the political
events of the Fitna (the war between cAll and Muwiyya),
that as preserved and transmitted the account [of that event]
might be read as dispute about sources (ul) of authority."8
The Fitna became a symbol of sectarian or juridical dispute
(fitna or ikhti1f). Hawting's detailed analysis of the
u1l disputes which underlay the slogans and polemic of the
Fitna reveals probably more about the concerns of the historians than about those of the participants in that struggle.39
The system of juridical sources which Shfi'l
elaborated in the Risla emerged as the dominant though not
the only system in later Islamic legal literature. Its
primary characteristic was insistence that the prophet's
sunna was to be justified by reference to prophetic Traditions
and that at least was thereafter universally accepted. The
argument of the Risla was organised in a fairly logical
and orderly manner such that the Book enjoined submission
and qiys was justified
to the surina, the sunna prov ed ijm
by reference to known and accepted aspects of the sunna.L
38. Wansbrough, Lii1ieu, 119.
39. I-iawting, The Significance of the Slogan.
L O. See Schacht, Origins, for a detailed study of Shfi'I's
system.

114
An important feature of Shfi d I T s assessment of the law
was his assumption that the Book contained guidance for
every eventuality.

r j

Jir
d' J' ' c- Li,' <j',
1

c>

No event befalls any one of the people of

God's din but that there is in the Book of
God an indicator as to the path of guidance
141
therein.
By "Book of God" here ShfiCi meant not simply the bare text
of the Qur'n but also Traditions, iJm and qiys in so far
as they were 'contained' in the Book. This was in effect an
assertion that there were no limits to the area of clerical
authority and Is to be contrasted with the more moderate
implications of the Mu'tazIlI theory, which limited both
the sources and the extent of divine law.142
The semantic spectrum of the term Imma Is further
clarified In the Risla:

1; t.. ji ,

> '3

j
He who grasps knowledge of God's commands
in His Book, whether.[deslgnated] explicitly
or by implication and to whom God grants
success in enunciating and acting on [that
knowledge] has attained excellence in his
din and his duny; he is free of doubts,
wisdom illuminates his heart and he merits
143
-
-
in din the status of imama.
Here it is explicitly stated that it Is knowledge of God's
41. Shfi'i, Risla, 48.
142.See further, below, 19 14_195, 206-214, 221-222.
143.Shfidi, Risla, 46.

15
law and corresponding action that produces in din (that is
religious law or prescription) the status of imma. This
may be related to the designation noted earlier of Iiasan,
Sha'bI and other local scholars as a'imrna fi 1-din. In
effect Shfidi was stating (and continuing thereby a
tradition already established in the older local schools)
It is in this sense that
that the fugah' were irnms.
Shfi d himself, Ab IanIfa, Ab Ysuf and Mlik were
recognised shortly after their deaths as irnms.
This understanding of what was Implied in clerical
literature by the term imm prompts further consideration
of the title of ShfI d I I s great compendium of law, the K.
al-Urnm.
While clearly a play on the quranic umm a1-kitb
this pun gains greatly in dignity and significance if it is
related to its cognates imm/imma. Discussing the application
of the term umm to the Ftitia, the first sara of the Qur'n,
Tabari said:
It is called umm [a1-qur'n] because the Arabs
so designate everything that gathers a matter
together or leads a matter [or] If it has
things that follow it such that it becomes an
Imm jmic [an imam that gathers things
together.]
That meaning was related to a secular usage illustrated by
the line from a poem attributed to Dh al-Rumma:

sJ

Lc>JU ti)ci'
At the head [of a lance] an umm whereby
we are guided. 4 The reference there was to the standard tied to the end of
a lance and leading the army. An umm then was an irnm JmiC
something which provided guidance and gathered together.
By implication it united divergent elements and resolved
dispute. That or something like It was the claim whIch lay
behind the title, K. al-tjmm, applied to Shfi d i ? s elaboratIon
of Islamic law. In another context Tabari implied that urnm
14L4 Tabarl, TafsTr, I, 108.

16
and imma were both verbal nouns of the verb amma.4
Shfi 1 I t s K. al-Umm was a Book of Guidance (imma)
by the Guide (imm) Shfidi. The implications of these
terminological usages may be summed up something as follows.
God was the ultimate authority whose command was to be
obeyed; (He too might be designated imm). '6 His will was
revealed to man through His Book (imm) and the actions of
His prophet (imm). Thereafter God's law was preserved and
expounded by the Companions, the Successors and the fuqaha'
who by their personal knowledge and implementation of the
law became themselves imms, to be imitated and obeyed by
the rest of the community. The prophet's role as mediator
between God and the community was thus re-enacted from
generation to generation by the fuqah'. Their status was
formalised in the warathat al-anbiy' Tradition which
recognised that the prophet left behind him not wealth but
only knowledge (cilm): the ulam' were heirs of the
prop.het .
4..

The clerical class in their capacity as imms and

heirs of the prophet defined and articulated God's law. That
law was presented as a normative code which, though product
no doubt of social pressures and historical events, was
thought to manifest an unchanging system of values quite
distinct from what actually happened. That which distinguished
the clerical class from the rest of the community and enabled
them to dictate what the law was was their knowledge (Cilm)
inherited from the prophet. Knowledge was understood by
Shfi C I to be a term exclusively applicable to knowledge
of God's command. In the Risla, in a lengthy and coherent
argument which stretches from the bb al-'ilm to the end
of that work, he discussed the nature of t ilm. Knowledge
was of two kinds. There was first dim mma, the knowledge
145.Ibid., III, 18.
146.E.g. in Ibn Qutayba, Ta'wil, 314.
147. Ibn kranbal, Musnad, V, 196.

17
belonging to a generality. This know1ede related to
general concepts such as that there were (as parts of worrn
ship) five prayers or zakt or ____ or that fornication,
killing, thieving and drinking wine were forbidden. All
servants (of God) were charged with knowledge of and action
in accord with these things.
This type of knowledge is found explicitly
(naan) in the Book of God or generally
('amman) amongst the people of Islam: the
generality of them ('awmnrnhum) relate it
from the general!ty, from the prophet of
God... This general knowledge is that
wherein there is no possibility of error,
nor of interpretation (ta T wll) nor of
dispute.
The other type of knowledge was a knowledge of the elite
(khsa).
It consists of the detailed duties and
prescriptions which fall to and concern
the servants of God concerning which
there is no explicit text in the Book,
nor in most of it is there an explicit
text of sunna, though there is sunna in
part of it: it is derived from akhbar al .
C amma; and it is
-khsanotfrmba1
that which permits of interpretation (ta'wIl)
- L!8
and is product of qiyas.
It is to be noted that the detailed duties and prescriptions
which were thus known to only a few were relevant to all.
This knowledge of the elite was a kind of 'ilm al-kifaya,
such that as long as a sufficient number of specialists
possessed it the rest of the community incurred rio fault in
not acquiring it.
Xonetheless In so far as they were
required to carry out the details of the law, the ordinary
people were expected to apply to the elite in order to
discover what to do. This theory in other words formulated
and justified the subordination of the generality (including
L 8. Shaficl, Risia, 96l-.. 967: cf. Schacht, Origins, 136.
9. Shafi'I, isaia, 971ff.

18
in this context rulers and princes) to the authority of
the clerical class.
The distinction between the two types of knowledge
was essentially one of sources, but it was also one of kind.
The knowledge of the elite permitted of interpretation and
difference. It was derived from the akhbr al-khsa and
from qiys and iJtihd, which were sources not permitting
of absolute certainty. 50 It is not immediately clear
from the Risia where ijmC fitted in this scheme of sources;
but a passage cited by Schacht from the Ikhtilf
iik w'alShfi C I makes it quite clear that ijm C - understood to be

of the C amma - is relevant only to the dim al-'mma.51

ShfidI was very sceptical about whether the consensus of
the elite could in fact be established. The implication of
his distinction between Cjlm mma and 'jim kha was
that the latter was precisely the area where there was no
ijm; and it was the area where if a scholar differed from
the majority one did not demand from him repentance.52
Difference of opinion thus was (while not
at least permissible. It may be noted in
ShfidI distinguished two types of ijm:
ijrn t which was the result of a Tradition

at all desirable)
passing that
there was the

and there was the

ijm which subsisted quite independent of any revealed
evidence. 53 Concerning the latter type of ijm
he thought

it was much weaker than al-sunnatal-mujtama alayh;5

(It is possible that he meant by it the ijmc of the clerics,
which would make it part of 'urn khsa.) With the development of exegetical technique after Shfidi and the increased
number of Traditions this type of ijm dropped out of
general consideration: ijm
in the developed tradition was
always thought to result from some detail of revelation.55
50. Schacht, Origins, 52, 127-128; Shfi'i, Risla, 1251,
1260-1.
51. Schacht, Origins, 92.
52. Shfi'I, Risla, 1261.
53. Ibid., 1309-1312.
SLI . Ibid., l8l-1817.
55. See further on ijm, below, chap. VII, 187-205.

19
ShfidIts distinction between the knowledge of
the generality and that of the elite was of immense
importance in so far as it justified the authority of the
clerical class and asserted the dependence of the rest of
the community upon them. The fugah/Cu1am? as heirs of
the prophet, custodians of revelation, with a unique access
to necessary knowledge, enunciated for the !'iuslim community
the law they should follow if they were to achieve an
orderly social life and, ultimately, salvation.

Alongside the theologically charged usage of the

term imm persisted a secular and more neutral usage. Imm
meant also military leader, governor or sultan. In the K.
Ikhtilf Mlik w'al-Shfi'i where imm is most frequently
exploited to mean precedent there exist also contexts where
it clearly means something different.
We say that the booty belongs to the killer
and the imm may not keep it back from him
under any conditions because the prophet's
giving [it to the killer] is an ordinance
(bukm ) from him.6
Here, imm means military leader, subject to not source of
the law. Likewise in the K. Siyar al-AwzT the term
imm/a'1mma is constantly in use to mean military leader.57
AWZ C I ? S usage is unambiguously clarified in Tabar's
Ikhtilf al-Fuqah'. Apropos of a problem related to the
usufruct of captured goods AwzI's interlocutor demanded:
Are you not then counting the amir al-sariyya as imrn? To
which Awz C I replied: Yes. 8 ,
1
Inim then meant the leader of a military band or an army.
It can be shown that Shfi'I used it extensively as a synonym
of wall (ruler, governor). The rules governing the collection
and distribution of zakat were articulated in the tJmm with
56. Shafi c i. Umm, VII-, 211.
57. K. Slyar al_Awza C l, in tJmm, VII, 303ff.
58. See Tabarl, Ikhtiiaf, 99-101. (Acknowledgement-s. to my
friend A.A. Sharif who pointed out this passage to me.)

20
reference to the

wlI

This would appear to be a general

term, including the caliph. Where distinctions were

necessary ShfiCI used also the terms wilT al-iglim alaTm
(provincial governor) and khalifa (caliph). 59 The rules
governing Jihd on the other hand were largely articulated
with reference to the imrn. The occasional use of
as
a variant 6 indicates that the walT and imam differed, If

wlI

they differed at all, in respect of function rather than

person. The governor (wall) was imam when he dealt with
military affairs. Shafl d l ? s assessment of the right of a
local ruler to nominate the imam (prayer leader) for Friday
prayer prompted reference to the ruler as variously wilT,
Imm and sultan, all apparently synonyms.61 Schacht's
translation of a'Imma as CalIphs62 is an unnecessary submission to one, not exclusive, late tradition; his translation of imm as government 6 is an impractical exclusion
of the field commander.
The varied uses of the term imarn may be Illustrated
also from Ibn Qutayba's Ta'wll. There he resolved the
apparent contradiction between two Traditions by pointing
out that imam meant in one case a'immat al-masajid (prayer
leaders) and In the other case It meant the sultan. Both
of these usages are to be distinguished from imam as mafzat
al-muslimln or source of law.6
The law declared the duties of the people and the
duties also of the rulers. For Shafi'l it was relevant to
all possible contingencies and so pre-empted for example
the discretion (ijtlhad) of a military leader. 6 Booty
under certain conditions he considered to belong to the
kIller, "whether the imm (military leader) says so or not;
59. Shafi c T, TJrnm, II, 61.
60. Ibid., IV, 63ff; imam, passim; wail, 80, line 19; 64,7.
61. Ibid., I, 138.
62. Schacht, Origins, 24, 70.
63. Ibid., 208.
64. Ibn Qutayba, Ta'wIl, 192-193; 29-30.
65. ShafI c i, tJmm, VII, 211.

21
(glahu a1-imm aw lam yagulhu)". 66 Concerning waste land
if the land had no owner and a Muslim 'revived' it then the
land was his and nobody else's:
2t
UzLJt oL
I do not care whether the sultan grants it
[formally] or not.6
Expressions such as these confirm the 'ideal' nature of the
Islamic code. No doubt generals continued to exercise
ijtihd, sultans to usurp 'legally' owned property and others
to perform acts not in accord with the law. Such acts were
discounted: legal 'facts' were not the same as historical
events. Islamic law thus had in this period a marked
ontological as well as soteriological function. It defined
a particular normative reality which subsisted quite
independently of de facto events. That characteristic of
the law (sunna) was already evident in Ab Ysuf's statement
that "most of what the people have always done is not Iiall."68
The fuqah' in other words conceived it their duty to state
what the law was. They formulated it as an ideal and denied
the impingement of actual practice on the expression of
that ideal. This is not to say that they were not concerned
with pragmatic aims: the law, systematically recorded and
elaborately portrayed as part of a theological structure,
was intended also to work, or at least to influence. What
is however strikingly absent from early definitions of the
law (Aba Ysuf, ShfiCI) is that peculiar spirit of compromise which led later Sunnl writers to concentrate on
demonstration that "the contemporary state could be regarded
as fulfilling by and large the demands of the sharica.h169
For Shfi

the law was a constant to which the ruler must

bend; for i'1T wardI and later writers there are signs that
the law might bend to accommodate the ruler.7
66. Ibid., 313, lines 23ff.
67. Ibid., 2l3-21'.
68. Above, 6-7.
69. Watt, Reflections, 1LI.
70. See below, chap. II,and IV, 87-88 and V, 117-120.

22
Consideration of the 'ideal' nature of the law
and the various usages of the term imm suggests some
conclusions as to

historical developments. In view of

the relatively late appearance of juridical and theological
literature (more than a century after the establishment of
Arab hegemony in the Near East) and the probable priority
of the secular less-charged usage of the term (imm =
military leader) it may be assumed that the clerical
elaboration of the law was in response to the arbitrary
usage of military power. That is, the clerical class,
which emerged initially in local communities, elaborated a
law whereby they hoped to impose order upon the community
and to limit the arbitrary and unpredictable exactions of
Arab military governors. In order to give their law force
and authority they related it to God and arrogated to themselves the title imm which was otherwise (and earlier?)
applicable to military powers. That development, though
hypothetical, is already implied in the works of Goldzhier
and Schacht and is the opposite of that suggested by Muslim
historians. They suggest that the Islamic community began,
a religious community and a model of perfection. It seems
now more likely that the model was fabricated by the clerical
class during the eighth century in order to enhance their
(religious) authority over and against the arbitrary (military)
authority of the ruling power. That this or something like
it was the pattern of development has been proposed by
Wansbrough. 71 He is Inclined to consider that Islamic
religious orthodoxy evolved as a result of extensive clerical
debate under Arabic political hegemony. He suggests that
the process might be the result of "the persistance of
Judaeo-Christlan sectarianism in the Fertile Crescent under
Arab political hegemony, the establishment of a modus vivendi
between the new authority and the indigenous communities, and
the distillation of a doctrinal precipitate (a common
denominator) acceptable initially to an academic elite,
eventually an emblem of submission to political authority."72
71. Wansbrough, Milieu, 120-127.
72. Ibid., 127; the date of the eventual submission remains
to be considered
see below, chap. II.

23
The acceptance by the community at large of the
values and ideology of a small group reflects partly the
forceful symbolism of their synthesis; but it reflects too
the ability of the group to express and resolve varying
social and economic tensions. The activities of the
clerical class in defining a code no doubt began under the
Umayyads and was encouraged in the early Abbasid period by
the ruling class (the caliphs) who sought support from them.
In order to bolster their own authority the Abbasid leaders
"submitted to the divine law even in respect of their own
person" and generally encouraged and patronised legal
studies.
If however such public stances were politically
expedient, they did not for all that preclude the promulgation of an alternative theory. Part of the movement that
brought the Abbaslds to power embraced extreme theories
of divine kingship 7 and indeed a strong theocratic element
continued to manifest itself In their propaganda. If it
was a logical conclusion of the clerical theory that they,
the scholars, were warathat al-anbiy' it was a no less
inevitable riposte that the caliph was:
c'
'_fJ

Heir of the [prophet's ] mantle, the

[prophet's] staff and the authority of God.75

73. The quotation

iS

from Goldziher, Studies, 57/63; 'cut

see all of 52-66/59-71.

VL Goitein, Turning Point, 156.

75. Cit., Sperl, Kingship, 21.

214

CHAPTER II
THE IMAMATE OF THE ABBASIDS
The 'classical' theory of the imamate as enunciated
by for example MwardI (d. 1450/1058) assumed a terminological
equivalence between imma and khilfa)
Most exponents
of the theory stated that there could be at a given time one
or possibly two imms. It is clear however that even late
in the ninth century these ideas had not crystallised. Ibn
Qutayba distinguished between imm as source of law (mafzac),
a term which included but was not restricted to the four
RshidIn, and imm as sultan, a term which presumably included
but was not restricted to the caliph. Early juridical writers
It is true, recognised the caliph as pre-eminent amongst his
peers. ShfidI for example acknowledged that the caliph, if
he entered a town otherwise ruled by a

wlI (governor),

had
more right to lead the Friday prayer than he. 2 Ab Ysuf
stressed that the caliph was appointed by God: "God has invested you with a great responsibility, He has made you
shepherd over the people (Istarkahum a1lh)" etc. But if
the caliph did not carry out the law as It was set out by
Ab Ysuf he was threatened with the possibility of divine
intervention to remove his power or eternal damnation.3
There were, further, Traditions which stressed the peculiar
authority of the caliph; (and others which limited the
duration of that Institution). There was however no generally
held belief that the caliph was the Irnm, not, that is,
amongst those who came to be recognised as Sunnl thinkers.
The caliph was at best one and admittedly the first amongst
several Imms. This Imamate of the caliph and other governors
was quite distinct from that of the fuqah', which depended
on and was a result of their ability to declare the law. The
early Abbasids in so far as they publicly submitted themselves to the law, or encouraged legal studies, or declined
1. For a general assessment of the classical theory, see
Madelung, Imma, in EI(ii).
2. Shafidl, Urnm, I, 139.
3. Ab Yisuf, Kharj, 3-5.

25

to implement the administrative projects of Ibn al-Muqaffa

At the same time,
recognised the authority of the fugah'.
at least in court circles, they revived and encouraged an
ancient tradition of divine kingship, which finally took
the form (in response to, perhaps, and as a result of,
clerical formulations) of a claim that it was the caliph
who was the heir of the prophet.5
Tension between the ruling and the scholarly class
became confrontation during the i1i1:ina, (2l8_234/833_8L8).
That was a period of dispute as to the nature of authoritr
so it is hardly surprising that in the polemic that was
generated (or in the reports of it later) the application of
the term imni was central. The caliph Ma'mn (d. 218/833)
both on his coins and in the letters attributed to him laid
claim to the title imm. 6 He conceived it as his duty to
conserve and maintain religion, 7 to make known by his
learning knowledge, to mark the way of salvation, to instruct
the people in the limits of their faith etc. 8 Thus learning,
knowledge and instruction, which were in scholarly literature
functions exclusively of the scholarly class, were claimed
as prerogatives of the ca1iph/imm. The writings of Jhiz
Cd. 255/868-9) were also a part of Abbasid propaganda. 9 }e
too stressed that the imrn (Abbasid caliph) was a teacher of
religious truth and In that respect In some way a successor
of the prophet. Pellat summed up his position thus:
La science religieuse est obscure, et l'on
n'y parvient que grace
une nature suprieure
et un soin assidu avec l'aide de l'enseignernent
des imms; 51 les hommes pouvaient atteindre
par eux-mgmes a tout ce qui convierit a leur vie
materielle et spirituelle ii seralt de peu
Lt. See Schacht, Origins, 58-59, 95, 102-103.
5. See Sperl, Kingship, 21-22.
6. See Patton, Alimad, for a still valuable account of the
Milina.
7. Ibid., 57-58.
8. Ibid., 65-66.
9. Pellat,
23.

26
d'utilit et de mince intrt que Dieu
envoie des prophtes .........
[En effet] ... 1'incapacit des hommes .
connatre tous les dtai1s de la religion
fait de la presence d'un irnam gui leur
enseigne tout ce que leur convient
une
necessit. 10
In opposition to these claims that the caliph/imm could
teach the people about religion it was asserted (at least
in later reports) that Abmad ibn Hanbal was imm or that
his Musnad was 1mm. The caliphal claim was associated
with sectarian theology (both Mu'tazilr and to some extent
Shidi) and was finally abandoned by the caliphs and repudiated
by the Sunni tradition as it developed after the Milina. It
is clear that the equivalence of khi1fa and imma was not
accepted by the (Sunnr) scholarly class either before or
during the !4ihna. It is further unlikely that It could have
been adopted by them in the period immediately after it.
Nonetheless it is in the period after the Mihna
that one must assume a gradual process of reconciliation
between the scholarly and the ruling classes. Something of
the nature and the chronology of that reconciliation may be
deduced from Coulson's study of pious reluctance to accept
judicial posts. 11 He notes that many scholars were unwilling
to take up judicial posts and he suggests that the reason
for their reluctance was that they perceived the law as an
ideal which might be subverted by imposition. They distrusted
official posts as leading inevitably to compromise. 12 "For
them the Sharlca represented a religious Ideal, to be studied
for its own sake, rather than applied as a practical system
of law.' 43 This attitude however was dominant only for a
short time: "later scholars showed a genuine concern for
the practical implementation of the 1aw.t14 Coulson finds
that the idealistic attitude to the law was dominant from the
10. Ibid., 38-39; (my underlining).
11. Coulson, Doctrine and Practice.
12. Ibid., 211-217.
13. Ibid., 230.
l4. Ibid., 223.

27
late second/eighth to the early fourth/tenth centuries.
Towards the end of and after that period a compromise was
worked out whereby ' t on the one hand the Sovereign himself
is subject to the overall supremacy of the Sharl'a;
the other hand he Is afforded the right to restrict the
scope of application of the law."15
That such a compromise was not envisaged by ShfI'I
has already been shown. A particular example of this spirit
of compromise, which began to dominate apparently at about
the beginning of the tenth century, may however be given.
In Q.59 believers are urged to obey God, the prophet of
l-amr,
Identification of the
l-amr.
God and the
those in authority, was a matter of some controversy. Tabarl,
in his Tafslr, which is a synoptic work surveying and
sununarising earlier works of the same genre, noted that
two distinct exegetical traditions existed with regard to
1-amr the
this verse. One group recognised in the
1-amr
fuqah and the c ulam t . The other group saw In the
a reference to the umar' and sa1!n, princes and sultans,
and to the leaders of the prophet's raiding parties. These
two opposed views, ascribed to various exegetes of the late
eighth and ninth centuries, symbolise the tension between
the authority of the scholarly and the ruling classes,
which had subsisted throughout that period. Tabarl himself
adopted the view that l 1-amr was a reference to umar',
wult and a'imma, terms used apparently as synonyms. He
continued:
It is known ... that He has commanded us to
obey the imms and those they appoint over
Muslims (man wa11hu l'muslimTh) and no one
else. It Is however a duty to accept [orders]
from all who command the abandonment of ma'Iyya
or who call to obedience to God. Obedience is
not incumbent in whatever is commanded or
forbidden when there Is no Iujja as to its
incumbency, except [when the command comes] from
the imms whom God has required His servants
to obey in whatever they command their people
15. Ibid., 225-226.

28
in the way of mala1ia. Obedience is incumbent
on those who are thus commanded, and in all
matters which do not constitute maiyya.16
The central proposition here is that obedience is not
incumbent when there is no I:lujja, except when those who
command are imms ( = urnar', wult). T abarl (in contrast
to Shfi d l) thus indicates an area of action in which there
is not a bujja:
it is in that area that obedience is due
to umar', wult, a'imma etc. The rulers were thus conceded
an area of discretion limited only by the never adequately
defined concepts of malaha and rna'siyya. 17 This assessment by Tabarl reflects a compromise similar to that detected
by Coulson: the sovereignty of the shar'a was maintained
but its scope was limited. Shfi'I had conceived of the law
as having a ruling on all possible eventualities and he had
expressed the law as an ideal. No doubt actual rulers for
the most part failed or were thought to have failed to
implement the law; to that extent the government would be
thought of as illegitimate. Tabari's compromise eased the
problem of legitimacy: a ruler's acts were legitimate
(shar'I, or at least not opposed to the sharl'a) if they
could be shown to be in accord with mala1ia and if they did
not constitute maiyya. These were terms of considerable
elasticity. Exploitation of these concepts, together later
with that of carral8 became in time a significant means of
adjusting the ideal sharita to the exigencies of daily life.
The evolution of a compromise between rulers and
the scholarly class was a result of discussion and polemic
which took place amongst scholarly groups throughout the
second half of the ninth century and the early tenth century.
The groups involved were those that came to be designated
Sunni. It seems likely that one of the factors which promoted
the elaboration of compromise was that, especially after the
r4i1na, Sunnl scholars, favoured by the ruling caliph (or
other military power), were increasingly incorporated into
16. Tabari, Tafslr, VIII, 503.
17. Cf., Lewis on ma'iyya in Concepts of Revolution, 33.
18. See e.g. Gibb, Mawardi, l64; also below, 87-88.

29
the political structure. Their political, social and
material interests became more and more identified with
the preservation of the military power. Conversely, the
caliph, or other military power, found that he could rely
on the clerical class to bolster his authority as long as
he recognised and supported theirs. There was thus during
the early tenth century a tendency for the Sunni clerical
class and the ruling class to provide each other with
mutual support. All this is plausible but is not adequate
to explain why the caliph came to be recognised as the
unique imm: that proposition may best be understood as a
response to external sectarian polemic.
Throughout the tenth century the political
authorities that ruled Iraq and adjacent areas were subject
to an ever sharper threat from and more intense polemic
with sectarian groups. Qarmita, Fatimids and Zaydis
became political powers which threatened or actually undermined the Abbasid caliphate. The adoption and propagation
by these sectarian revolutionary groups of ideological superstructures which proposed a determinate imm to replace the
Abbasid caliph may be thought the cause not the result of
Sunni concern with that particular institution. The Sunnis
had thrust upon them the concept of one imm. Before the
tenth century the Sunnis had responded to sectarian (ShI'I)
claims by insisting on the imamate of the prophet, the
It was the prior involvement
Companions and the fuqah'.
of the Sunn clerical class with a political power, and
the need for political (military) support against sectarian
(military) enemies that prompted the Sunnis to reformulate
their theory of irnma.

The nature of the reformulation

may be elucidated by comparing the theories of Ashcari (d.

32 L /935), at the beginning of the tenth century, and

Bqi1lnI (d. 1403/1013), at the end.
AsharI's discussion of the imamate related only
to that of the four RshidIn, and primarily to that of Ab
Bakr) 9 In order to prove the imamate of Ab Bakr he

19. AsharI, Ibna, 92-96.

30
adduced as proof: a. the ijmac of those praised by God in
and c. iJm'.
the Qurn (the Companions), b. quranic na
The second ijmc partly repeated the first proof and was
directed explicitly against those groups who claimed that
the imamate.had passed to either CAbbs or 'AlL Both of
these were amongst those who had recognised and accepted
the irnamate of Ab Bakr. Ash'ar3 thereby refuted the claim
of ShPI groups and extreme legitimist Abbasid groups. After
Ab Bakr the imarnate passed by a combination of quranic
and contract ('aqd) to 'tJmar, Uthmn and
na, personal na
c All. A final element in Ash'arl's proof was the Tradition:
UJ r
the khilfa

will be for thirty years, then mulk. That implies that
khilfa and imma were in this context synonyms but would
appear to mean that the caliphate/imamate of the Rshidln
was a unique institution not later repeated. Ash'arl went
on to indicate that the wars which broke out amongst the
Companions were due to matters of ijtihd and that no blame
was to be apportioned: all of the Companions were a'imma
ma' munun.
It is clear that Ash c arl, in referring to the
imamate of the Rshidln, meant their political leadership
of the community. In a different sense all of the Companions
(including the Rsh1dln) were imms; presumably in the
sense that, as Companions of the prophet, they were thought
to have preserved or acted in accord with the law. The
imamate/caliphate finished with CAll and was replaced by
mulk.
In his statement of orthodox belief at the
beginning of the Ibna Ash'arl stated:
[1.] We cling to the Book of our Lord and
the sunna of our prophet, and to what is
related from the Companions and the Successors
and the a'immat al-Iiadlth ... [and we cling
to] the opinions of Ahmad ibn Hanbal because
he is al-imm al-fi1 and al-ra'ls al-kmil
.. so the mercy of God be upon him who was
an imm muqaddam and a glorious khalll.20
20. Ibid., 8.

31
[2.] We consider it right to pray for the
welfare of the a'immat ai-musiimln and to
recognise their imma and to reject those
who propose rebellion (khurj) if they
[the imms] deviate from righteousness.
And we reject rebellion with the sword, and
we refrain from fighting in the [period of]
21
fitna.
The first of those passages suggests that there was an
irnamate of knowledge which passed from the prophet to the
Companions and to the Successors, and thence to the a'irnmat
al-adTth and finally to Ahmad. Recognition of AI?mad as
an imm muqaddarn and a khalfl implied that his function and
his status were similar to those of a prophet. 22 That was
however simply a variant in the traditional assessment of
The second passage is clearly a
clerical or legal imma.
reference to political powers, the a'immat al-muslimin, who
may or may not act in accord with the law. It is not clear
that the imms in question were necessarily the caliphs, .nd
in view of the development of that term as so far discussed,
it seems unlikely (though not impossible). The implications
of AsharPs use of the term, imm, might be expressed as
follows. The political power of the prophet passed to the
RshidIn whose uniqueness lay in that they combined with
their political leadership knowledge of the law: they were
imms in both senses of the word. The period of their rule
became therefore a symbol of political and religious unity,
a model of what political rule should be. Thereafter
political power became fragmented (fitna) and became separated
from religious knowledge: that was preserved by the other
imms, the fuqah'. The political powers which emerged
after the RshidIn were liable to deviate from righteousness
but were

nonetheless not

to be rebelled against. True

21. Ibid., 12.

22. Cf., Abraham as khalIl In Q 14.125; for a similar
assessment of Al?mad as imgm see Ibn Abi Yadl,
Tabaqt, 3-il.

32
believers were to avoid participation in fighting during
the fitna; (fitna here would appear to mean the whole
period which followed the disappearance of the ideal
caliphate/imamate with the death of "All.) Ashtarlts
discussion provides no evidence that the caliph was the
unique imm; it suggests rather that the imamate of the
Rshidln was sul generis in so far as it represented the
combination of political power and legal knowledge. Thereafter there were two types of imamate, separate and distinct
from one another, the military, political imms and the
clerical imms, represented by A1?mad.
Eq1llnl's discussion of the immate falls into
three parts. The first is an assessment of the sources of
knowledge 23 ; the second lays down the juridical rules
relevant to the imamate as an ins t i t ution , 2 (derived from
the precedents set by the Rshidln); the third applies
these rules to the imamates of the Rshidln and thereby
'proves' their validity. 25 The circularity of that argument
is patent and disturbing. It might therefore be permissible
to wonder whether in fact the real aim of the exercise was
to prove not the validity of their imamates but rather the
validity of the rules. Extrapolation of juridical rules

pertaining

to the imamate as an institution was not previously

a feature of Sunnl thought. Wensinck noted that the

conditions of the imamate and the fact that there was only
one imm were notions that did not appear in any of the
early creeds which he studied, that is, in none of the creeds
up to and including the Fiqh Akbar II, which he dated to
the tenth century. The earliest elaboration of a theory
of the imamate which he knew was in the Ul of Baghdd
'Tow, Bqillnl's Tamhld being available,
(d. 14 29/l037_8) . 26
it may be suggested that Bqillnl was the first (or at least
the first at present known) writer to extrapolate from the
available precedents a series of juridical rules pertaining
to the imamate. Isolation of such rules and demonstration
23. Bq1l1nl, Tamhld, i6O-i64.
24. Ibid., 164-187.
25. Ibid., 187-239.
26. Wensinck, Creed, 269-270.

33
that they were valid (in so far as they could be applied
successfully to the already established and accepted
imamates of the RshidIn) rendered it possible to test the
contemporary situation against them with a view to ascertaining whether or not an imm existed. It seems probable that
to provide such a test and to provide also for a positive
answer was BqillnI's intention.
The imamate of the RshidIn was known to be the
period when the sharica was fully implemented. The assertion
that the Abbasid caliphs were, in a way similar to the
RshidIn, imms had as a corollary the implication at least
that the sharl'a was being put into effect. It has already
been suggested that, in so far as the shara was an ideal
which the political authority to a greater or lesser extent
failed to implement, those authorities failed to achieve
legitimacy. It was however a feature of Sunnl thought from
about the turn of the tenth century that they permitted to
the rulers a certain degree of discretion whereby their
rule might more easily be interpreted as sharti and legitimate.
BqillnPs elevation of the Abbasids to the status of irnm
(by implication, like the RshidIn) made still easier the
A further imlicrecognition of the actual as the shar'T.
ation of this tendency was that the sharc3 ruler deserved
the obedience and support of the Sunn population.
The first part of BqillnI's juridical argument
is as follows:
and ikhtiyr (designation and choice)
a. na
are, as methods of gaining an imm, mutually
exclusive alternatives.
b. ShII theories of na

are refuted.

Hence c. ikhtiyr is the only valid method

of gaining an imm.27
Bqil1n thereafter assumed rather than proved that
ikhtiyr belonged to the umma. That was the conclusion he
desired and for which he fabricated the inItial premise (a.).
had played a
In fact both quranic na and prophetic na
part in earlier Sunni defense of the Rashid n.

By

27. Bqi1lnT, Tamhld, l6 L_ l78; cf., Abel, BgillnI, 57-63.

28. For quranic nass, see Ash c arl above; and for prophetic
na, Bqilln!, Tamhld, 195-196.

3L

disregarding the prophetic and divine elements which, by

earlier assessments, had been instrumental in designating
an imm, BqillnI was enabled to present the imamate as
depending solely upon the choice of the community (umma).
In that respect the imamate was no longer an institution
possible only in an ideal past when God or the prophet
intervened: in so far as the community had a continuing
existence it could continue to choose an imm.
Bqil1nI discussed the function of the imgm thus:
The imm is appointed merely to deal with
rules, ordinances and affairs which have been
laid down by the prophet and concerning which
the urnma already has full knowledge. He in
all that he undertakes is wakil to the umma
and n'ib to it. The umma is 'behind' the
imm to set him right, to support him, to
remind him, to warn him, to exact rights from
him when they are incumbent upon him, to
depose him and substitute another imm when
he commits something necessitating dismissal.29
The imrn's duties were in sum merely executive, to carry out
certain functions knowledge of which lay, not with the imm
himself, but with the umma, a body deemed capable of recognisirig his errors, setting him right and ultimately of dismissing
him. This authority which the umma possessed derived from
the knowledge of which it was custodian, guarantor and, not
least, expounder. To establish that point BqillnI had
introduced his discussion of the imamate with an assessment
of the sources of knowledge. 30 Those sources were primarily
revelation (Qur'n and Traditions), ijm' and tawtur. Of
these ijm' and tawtur were expressed through the umma
while, as has been shown, the fuqah' had exclusive rights
to interpret the meaning of such revelation as was not
tawtur or such law as was not covered by ijm. It was
because of its preservation of knowledge that the umma
possessed authority. In effect to ShI'I claims that the
29. Ibid., l8'4; cf., also 185-186; Abel, Bagillan, 65-66.
30. Bqil1nI, Tamhid, i60-i64.

35
imrn was maCsam (preserved free from error or sin) and that
the umma was restrained from error because he was 'behind'
them, Bqllln opposed this view, that the umma was mattim
and it was 'behind' the imm.31
According to Bqillnl there could be only one
imm at any one time. This was a ruling designed undoubtedly
to provide a symbol of the unity of the Sunn community in
spite of political fragmentation amongst Buyid and other
princes. It was a tenth century innovation. 32 It suggests
that there was a need for a political focus round which
the Sunns could gather to defend their beliefs (and also,
and not incidentally, the material interests and established
values of the clerical class). That need was generated by
the threat presented by various Shr d l groups, most notably
perhaps the Fatlmlds, but also ImmIs and Zaydis.
In the event of a recalcitrant multiplicity of
imgms, Bqilln stipulated that the people should fight
them, g.talahum al-ns. If thereupon the people were unable
to achieve anything then they were in a state of ghalba and
fitna and had an excuse riot to appoint an imm (tark immat
al-imm). 33 A similar more serious situation was envisaged
if a sectarian group, those who could be designated unbellevers, (kufr, fisg, all) appointed their own imm. In
such a case only the ahi al-Iagg had a valid right to give
the contract. At that point Bqi1ln slipped into the
first person plural:
If we are able we must compel the others to
obey the one to whom we contract the imamate.
If however they repel us and contract the
post to one of themselves then his is no
firm imamate and no obedience Is due to
him. We would be in dr qahr wa-ghalba
and that would be fitna which constitutes
an excuse for leaving - not carrying out the contract.
31. For further discussion of the imm/umma polarity see
below, chap. III, 53_5LL
32. Ibn Taymiyya also noted that the early scholars had not
insisted on one imm: Laoust, Essai, 279-283.
33. BqillnI, Tamhld, 180.

36

o.L3?).3
The situation envisaged might be relevant either to a Fatimid
advance upon Baghdad or to a Buyid appointment of a Zaydi
imm. Similar fears are evident in the discussion of the
umma's right and duty to dismiss the imm, which might arise
e.g. if he became a kfir, or if he were imprisoned under
such conditions that harm might befall the community or
that there was no hope of his escape. 35 Again the threat
which prompted such speculation might be either Fatimid
(Ism C IlI) or Buyid (ZaydI/ImmI). In either case the rule
of law and orthodoxy was in some degree guaranteed by the
power of the people and where that failed the final sanction
was to omit the contract, that is not to have an imrn. The
imamate was thus a title indicating legitimacy (subordination
to the umma) and might under appropriate conditions be
withdrawn.
It can be no coincidence that BqillnI elevated
the Abbasid caliphate to be representative (n'ib) of the
Sunni umma at a time when that institution was devoid of
real power; it was precisely the lack of power there
that permitted the symbolic plenitude. (The European
constitutional monarchs present an analogous pattern of
development.) The Abbasid imrns were portrayed as representatives of the Sunnl umma, focus, as long as the shari t a was
more or less carried out, of Sunni loyalty. The Buyids (the
Tamhld was written for an anonymous amir, a Buyid prince)
were thereby encouraged to benefit from that loyalty by
preserving the imm and acknowledging the sharIa. The
established interests of the clerical elite (and no doubt
other social groups) were likely to benefit from the power
of the Buyids to ward off the Fatimids. Since the Buyids
were in fact carrying out the political functions of the
imam, they were seen to be themselves indirectly a manifestation of shard rule. The shari c a, no longer merely an
ideal, was definitively institutionalised - the end of a
process whose beginnings might be located at the abandonment
of the Mihna. Acknowledgement of the political institutions
as manifestations of a shar'i ideal was, in effect, the
314. Ibid., 181.
35. Ibid., 186.

37
price the Sunnls had to pay for the defence and preservation
of their faith and their interests. It was still possible
to look upon the shari'a as an ideal to be preserved in
despite of actual practice, but on the whole later fuqah'
manifested a primary concern to justify the status quo:
that which was de facto implemented was the sharIa, or at
least a sufficient approximation to
BqillnI's assessment of the imamate modified
but did not abandon the traditional Sunnl assertion of
legal prerogative over monarchical discretion. The imamate
of knowledge, which, for Shfidi and for AsharI, had been
invested in the clerical class, was, by BqillnI, situated
in the umma at large. The umma was, as It were, imm.37
It was the umma which, by virtue of its knowledge, had the
right to choose and dismiss an executive (imm) and to
exact from him obedience to the law. Thus the imamate of
knowledge, the imamate that is of the prophet, 8 devolved
upon the umma, while the imamate of the RshidIn, the political
leadership of the community, passed to the Abbasids. The
greater right, to preserve and propound the law, fell to
the umma; the lesser right, to execute the law and t
represent the umma (wakil, nib) fell to the imrns, the
caliphs. The RshidIn, acting as members of the umma and
participants in the election of the imms, were articulators
of the law and served therefore as precedents; acting as
Imxns (rulers), they too were servants of the law, agents
of the umma. By recognition in the Abbasids of' legitimate
shard imms BqillnI implicitly recognised in the actual
political community a legitimate shar

urnma: it was the

reification of an ideal.
The urnrna, in BqillnI's theory, was presented
not simply as the infallible custodian of the law but also
36. See further, above 20-21; below, 87-88, 117-120.
37. For umma = imm, see iTansbrough, Studies, 514.
38. For the prophet as imm (but clearly riot in accord with
juridical rules) see BqIllnI, Tamhld, 33.

38

as a powerful Instrument to ensure, even to enforce, the

application of that law, (though it may never have done so
In actual practice). General Invocation of the umma yet
did not in the end preclude the fact that the actual
participants in the contract of the imamate were few:
they were the afil al-muslimTh, "those who are the ahi alliall w'al _caqd and entrusted with this business". 39 That
this is a reference to the cu1amt, who were spokesmen for
the umma, might by now be easily conceded, but finds In any
case ample confirmation In the parallel discussion of
BaghddI who refers to the electors variously as: ahi alikhtiyr, ahi al-iJtihd wial_waraC, mujtahld,
lim, 'ulam'
al-umma.
According to Baghdd:
Most of our companions, both mutakallImn
and fuqah', along with the Shi c a, the
Khawrlj and most of the Mu'tazila assert
the necessity of the imamate. They assert
that appointment of an Imm is a duty and
obedience to the one appointed is incumbent.
Muslims cannot do without an imm to put
the law into effect, to Impose penalties,
to raid with their armies, to marry off
widows etc.
This view was ascribed to AsharI who, Baghddr claimed,
had stated that the imamate was a sharl t a mm al-shari',
and Its necessity was known by revelation (samC). Further,
the Companions had achieved consensus on its necessity and
the sharIa Included rules which could not be carried out
except by an imm or a 1kim mm gIbalihi.

This is the
samdl or sharI argument for the necessity of the Imamate:
it states in effect that the shara requires an executive
agent designated imm. Now, that Ashcarl enunciated this
necessity might be true, but it is not evident from the
Nor had BqillnI assumed that
Ibna nor from the Lumac.
39. Ibid., 178.

O. BaghddI, Usl, 280-281; cf., Binder, Ghaz1I, 237.

l. BaghddI, tJ1, 271-272.

39
the imamate was a necessary institution.2 Indeed he
asserted that the umma could under certain conditions avoid
appointing an imm. An imm, for him, was not a leader
whose existence was at all times necessary, it was rather
a title of legitimacy which a leader merited in so far as
he carried out the law or obeyed the umma. The argument
for the necessary existence of the imamate once it came into
existence was exploited, by Baghddi and later writers,
primarily in order to demonstrate the sharci nature of the
ruling power. The logic it generated was of an odd and
unsatisfactory kind. The argument was based on the assumption
that if the shari c a were to be fully implemented it would
require an imm. But that already begged the question of
whether imm was the only title that could be held by one
who implemented the shari c a. It further assumed that the
sharl'a was a practical code which had to be implemented,
thereby neglecting the alternative idea that the sharica
was an Ideal which could only be aimed at. The argument
further confused moral necessity (there ought to be an irnrn)
with natural necessity (there must be, has to be, is an imm,
objectively identifiable in the Abbasid caliph).

Once it
was established that there was an imm, it seemed to follow
that that which was implemented had to be the sharica because It was implemented by an imm. GhazlI introduced
the idea that that which was implemented (by judges and
other officials) had to be recognised as the shariCa: it
followed that the ultimate delegator of authority had to be
recognised as imm. Without an imm there could be no
sharT c a.
All of these arguments were found either explicitly
or by Implication only in the classical theory of the imarnate
as it developed after Baghddl. They stemmed ultimately
from the sarnCl argument for the necessary existence of the
Imm.
By introducing that argument BaghddI Implied that
the de facto powers (Buyids and Ghaznavids) in so far as
t 2. See Ibish's remarks on this omission, Ibish, Bgi1lnT,
101.
L3. This perception is derived from Binder, GhazlI, 235.

140

they recognised the authority of the caliph/irnm and derived

their authority from him (though this was frequently form
without content) were valid shard powers. They, in
carrying out the sharI functions which belonged to the
imm, with his formal consent, were implementing the sharica.
Therefore they deserved loyalty and stability. They gained
the support of an ideology which was effective against
external (sectarian) rivals and internal (sectarian) dissent.
In return they acknowledged the religious nature of the
community they ruled and promoted the authority and vested
interests of the clerical elite. The Abbasid caliph,
presented by the fuqah' and accepted by the rulers as an
imm whose existence was necessary, symbolised the alliance
between religion and state, between jurists and princes.
BaghddI continued to present the imm as, theoretically at
least, subordinate to the umma:
The relationship of the umma to the imm
is like that of the imm to his khulaf',
- 1414
-
-
quat, tummal and sucat.
Society, once it has invested in a symbol, may find
that symbol a natural and inevitable form of self-expression
capable of manipulation so as to articulate even subtle
changes in social structure. That NwardI intended, in his
assessment of tie imamate, to demonstrate the legitimacy of'
the contemporary government has been recognised, 14 but it
can be shown that his theory was markedly different from
that of BqillnI. It may be possible to relate the differences to a different view of the social structure.
It Is noticeable that when it came, for Mawardi,
to the question of election:
There emerge from the people two groups:
one, the ahi al .-ikhtiyr, who choose an imam
for the umina; the other, the ahi al-imma,
one of whom will be appointed. The rest of
the umma, who do not belong to these two
144.BaghddI, Ul, 277-.278.
145.Glbb, Mawardi, 151-154; Theory, 1141-1142; Lambton,
Khilfa, in EI(ii).

L1
groups, incur neither blame nor sin in
the event of delay.6
The community was thus, in effect, divided into three groups.
This division was not only foreign to BqillnI but it
separated the electors from what had been for him the
source of their authority, the umma. Thus isolated, they
had to possess three qualities:
- Cadla (justice);
- C ilm (knowledge, whereby to recognise those
worthy of the imamate according to the
relevant legal conditions);
- ra T y and l4kma (political insight whereby
to choose one most suitable for the post).7
Earlier commentators have assessed these qualities as not
very restrictive. Gibb thought the electors, thus defined,
could include the Buyid princes; while Laoust thought they
could include "toute personallt gui dolt a son position
soclale une autorit dont ii serait maladroit ou dangereux
de ne pas tenir compte". 8 However a stricter interpretation
would be more in keeping with the general assumptions of
the age.
Knowledge of the conditions of the imamate was
a prerogative of the fuqah' and political insight a quality
which might be denied to some of these. It seems likely
that r4wardI meant that the electors should be fuqah'.
It is possible that he meant some of the fugah' to the
exclusion of others.
If rIward then, by separating the electors from
the umma, introduced a new element into the law, he disguised the innovation by presenting it, without overt
polemic, as simply accepted. Similarly he asserted categorically that the imamate became binding either by the
choice of the ahi al-Iiall w'al-agd or by nomination (cand)
by the previous imm. 5
L16.

Now, BqillnI indeed, apropos

MwardI, Alikrn,

L7. Ibid.
48. Gibb, Mawardi, 156; Laoust, MwardT, 28-29.
49. Above,3g.
50. MwardI, A tikam , 6-7.

of AbU Bakr's appointment of CUmar, acknowledged Cand

(nomination), but he had defined it as a particular kind
f' Cqd

(contract). It was carried out by one of the

ahl al-Iiall w'al-'aqd, namely the imm himself, in the

presence of representatives of the community. 51 The symbolic
priority of the umma over the imm was maintained. MwardI
by adducing two methods of contracting the imamate, significantly reduced the authority of the umma.
The reigning imm might, without consultation with
the ahi al-1all wtal_Cagd, nominate (cand) his successor;
thereafter not even the formal acceptance of the ahi alikhtiyr was required - though there was ikhti1f on this.
In the special case of a father-son relationship subsisting
between the imm and his nominee, one of the valid views
there too was that the reigning imm might legitimately
nominate a father or a son to succeed him, without consultation. This was justified "because he is amir al-umma and
n.fidh al-amr lahum wa_Calayhim',2 It thus emerged that the
imm was permitted total independent authority in appointing
his successor. It is a further mark of the independent
authority of the imm, uncircumscribed by that of the
community, that he was permitted to appoint a number of
successors In order. MwardI specifically stated that the
ahl al-ikhtiyr were not permitted to interfere with such
an order because the nomination of a successor ('and) conwhich, once extant, pre-empted recourse to
stituted a na
ikhtiyr. 53 An imm, however, once established, could set
aside the order laid down by his predecessor. This confirms
the absolute authority granted to each im gm in this field,
and emphasises the superiority of his authority over that
of the electors, who were not permitted to Interfere in the
order laid down. Thus, the relatively simple methodological
device of adducing cand as an independent method of appointing
an imm separate from aqd enabled MwardI to reverse the
hierarchic order established by Bqil1nI. For MwardI the
imrn's authority was greater than that of the electors.
51. Bqi1lnI, Tamhld, 201.
52. MwardI, AIkm, 12-13.
53. Ibid., 18-21.

L3
It was not clear in MwardI's original statement
about the qualities required in the electors whence they
derived their authority. He had omitted to suggest, indeed,
had rather denied, that they derived it from the urnma.
Having thus obscured what might be thought the original
source of the electors' authority he went some way to
suggesting an alternative source.
The khalTfa may appoint by na
the ahi alikhtiyr just as he may appoint the ahi alCand [his successors] in which case election
(ikhtiyr)ls valid only of those thus
appointed, just as the accession is valid
only of the nominee he appoints; for these
two functions [designation of electors and
designation of successors] are rights
belonging to his khilfa.4
This not only confirms the subordinate nature of the electors'
authority but might in some.degree imply that they derived
such authority as they had from the fact of appointment by
the imm/khalIfa.
Whereas BqillnI's work implied a hierarchic
sequence as follows:
umma
electors
imam
MwardI's work suggested:
imam
electors
umma
That reversal may be further illustrated from a different
context. Bq1l1nI had suggested that the imm was subject
to error but the umma could guide hIm and set him right.
MwardI defined the first of the duties of the imm as
follows:
Preservation of religion according to its
firm bases and what has been established
by the salaf al-umma.
If an Innovator
arises or a doubter deviates he (the imm)
will reveal the proof (l2ujja) and clarify
5Z4 Ibid., 18.

1414
the truth and exact such rights and
penalties as are required; so that
religion should be preserved from defect
and the umma restrained from error.55
Thus, for MwardI, the urnma was liable to error but the
imm could guide and set aright.
In his discussion of naq al-taarruf- the inability
of the imm to act freely owing to the impingement of Lome
external military or political power - MwardI proposed
two subheadings, liajr and qahr. Under the former heading
he discussed how the imrn should act if the usurper was
just and religious and how he should act if the usurper was
unjust and irreligious. In neither case were the electors
or the umma called upon to act. Under the second heading,
which would seem to be relevant only to a possible Fatimid
capture of Baghdad, (though it might in fact be merely a
remnant of BqillnI's theory), MwardI allowed for some
action on the part of the umma or the electors. 6 BqillnI's
discussion of the same problem had been subsumed under one
heading and had involved the immediate action of the. umma.
MwardI's discussion, by the characteristic introduction of
a subdivision, had reduced the authority of the umma and
rendered it of only peripheral importance.
On the question of whether it was permissible to
address the caliph as y khallfat allah Mwardi said:
He is called khallfa because he succeeds
(khalafa) the prophet in his umma. It is
permissible to call him y khallfat allh57
or simply khalfa. There is ikhtil g f on
whether it is permissible to call him khallfat
allah. Some have declared it permissible
because he undertakes His rights (uqq)
within His Creation and because of [Q6.165J:
He it is who has made you khal'if al-ard...
55. Ibid., 23.
56. Ibid., 30-32; of., Gibb, Mawardi, 159-160.
57. Fagnan corrects this to ya khalflat rasl al1h on the basis
apparently of only one MS, riwardi, Ahkm 3 (tr. Fagnan), 29;
he claims to have checked all of Enger's NSS, ibid., ix-x.

L5
have declared it not
'Iost of the culam
permissible and have called those who use
it iniquitous (fjir). They say that only
one who is absent or dead appoints a khallfa,
while God is neither; and furthermore Ab
Bakr denied the titie.8
Gibb concluded from this that the designation (khallfat allah)
was illegal and Impious. 59 In fact, at most, it was a matter
of Ikhtilf. MwardI's own view was not necessarily that
of most of the Culam: Indeed his work would seem to embody
a novel, if not revolutionary, concept of Imma. The
argument that the caliph might legitimately be addressed as
khallfat allah was provided with a quranic proof and was
related to the supposition that the caliph undertook God's
rights within His creation (li-qiymIhI bi-ugqihI fi
khalqihl).
It is not Immediately apparent what that phrase
might mean. After the discussion of the caliph's title
however MwardI enumerated his duties and concluded as
follows:
If the imm carries out these 1uqg al-umma
then he has fulfilled the liaqq allah in
what Is due to them and In what Is to be
exacted from them; and it is incumbent on
them to grant him obedience.60
Mwardl makes this statement without mention of sources or
ikhtilf: It Is his view of the imamate. He would appear
to mean that the Imm, in carrying out his duties, (to
preserve religion, to defend the frontiers, to impose
penalties etc.), was fulfilling rights due to the umma. In
so doing he was also fulfllllng'a right due to God. God's
right here can only mean His right to be worshipped (taabbud)
It was a right due to
by Implementation of the shari c a.
God that the shara be implemented. It was a right due
to the community that they be enabled to perform their
shard dutIes. The only person who could carry out these
two-fold rights was the imm: he was the essential mediator
58. MwardI, A1km, 22-23.
59. Gibb, Mawardi, 158.
60. MwardI, A1km, 25.

between God and the community. Without the imm God would
not be adequately worshipped; without him the people would
be unable to perform their shard duties and so benefit
from (present) social stability and (eternal) salvation.
The argument is not compelling but it was one of the
results of the statement that the imm was necessary to
implement the sharl'a.
It is now perhaps clear that when Mwardl said,
as an argument in favour of the title khallfat allh, that
the caliph undertook God's rights in His creation, he meant
something similar. God had a right to be worshipped by
implementation of the sharIa, and that, according to the
sarn t i argument for the necessity of the imarnate, depended
on the imm. The irnm was necessary in order that the
sharl t a be implemented. Only the imm, so the argument ran,
could validly appoint qIs, defend the state, impose
penalties and so provide for God's rights within His creation.
If this is correct, It would seem to be the case that MwardI
himself approved of the title khalTf'at allah. Indeed, within
the context of his exposition, this is not surprising (though
his argument In the event was not widely accepted): for
MwardI, the caliph/imm was the final earthly source of all
legitimate authority, a symbol of God's rule (the sharl'a)
and so, also, khallfat allah.
ShfiCi had distinguished between an imamate of
knowledge relating to the law and an imarnate of power relating
to government. It was no part of his theory that the
political state as it existed was shar'I: it might become
so only In so far as it carried out the sharIa, which on
the whole and In general, no doubt, it didnot. The religious
umma (the sharLi umma) existed within and as part only of
the political umma. BqillnI preserved the distinction
between the imamate of knowledge and the imamate of political
power: the umma had the knowledge and the Abbasid caliph
had the legitimate power. In so far however as BqillnI
elevated the imm (political power) to be representative
of the umma (religious knowledge and authority) the political
state became also and fully a religious (sharti)one.
Mward too preserved the distinction between the irnamate of

147
knowledge and that of power, but for him, it would now
appear, it was the caliph who represented the imamate of
religious knowledge: it was the caliph who preserved the
law, guided the umma, exacted God's rights etc. He was
the representative of God. The urnma's authority was
diminished or disregarded.
The value of the caliph/imm was as a symbol: (his
function as the ultimate authority in theory, but powerless
in practice, may be compared with the function of the
British monarch). rlawardl's motive for changing, as he
undoubtedly did, the nature of the symbol requires some
elucidation. Any assumption that he hoped for a re-establishmerit of real Abbasid power may be discountedGl on the
grounds that such was not at this time and had not for over
a century (since perhaps the time of Ibri

R?1q62)

presented
itself as a likely possibility. Indeed it was the very

absence of such a possibility that rendered the caliphate

so apt for such generous symbolic value. The general effect
of Mawardi's argument was to demonstrate the legitimacy 'bf
de facto governments and in this respect he merely enhanced
the efforts of BqillnI and BaghddT. Sunnism allied
itself with military strength in order to defend itself
against sectarian ideology, as represented by, most threateningly, the Fatimids. While still preserving a distinctior
between the symbol of law (the imm) and the de facto
realisation of power, the main aim of the fuqah' was to
demonstrate that the shari c a was being more or less put
into effect, even if this meant, as it frequently did, a
bending of the sharCa to fit the situation. To a considerable extent the symbol displaced the symbolised.6
61. Pace Gibb, Nawardi, 152-153 arid Laoust,
wardI, 11416;
cf. Goldziher, Studies, II, 66/71.
62. Ibn R'iq, the first amir al-umar' of the Abbasids, d.

330/9 14 2.

For details q.v. in EI(ii).

63. Cf. Binder, GhazlI, 239: Recognition of the ShariCa

by the Sultan without obedience to its provisions is
form without content.

An allied effect of MwardI's (as of Bqil1rii's and EaghddPs) argument was that it provided a symbol of the
religious unity of (Sunni) Islam, in spite of political
fragmentation.
There was at this time a threat to the values and
interests of the established scholars other than that which
came from sectarian groups. Their authority within the
community at large was eroded if not usurped by an emergent
group of Hanballs, Traditionists and Ascetics. 6 Relations
between the caliph and the fugah' who formed part of his
entourage on the one hand, and the rabble of Baghdad with
their Hanbal and ascetic leaders on the other were not
always good. And yet these apparently disruptive elements
were not sectarian, they were part of the umrna. It was
perhaps to exclude their influence that MwardI, in
elaborating a theory of imma, neglected or minimised the
authority of the umma.
The value of the caliph as a symbol of the sharT'a
was apparently recognised almost immediately by both princes
and scholars, who, together, supported and preserved that
institution until the heathen Mongols arrived in Baghdad.
It is a striking irony that mainstream Sunnr thought, which
had defined itself in an earlier age against the T1iina,
provided in the eleventh century (too late) for the Abbasids
all that Ma'mn had laid claim to. The scholarly class
sought refuge from external sectarian threat (the Fatimids)
and internal disruption, both sectarian (Shi'I) and Sunni
fT), by articulating a compromise
(IanbalI, possibly also
between jurists and princes, a compromise publicly affirmed
and symbolised in the continued existence of the Abbasid
imam.

See Makdisi, The Sunnl Revival, 16 14_168; Laoust, Les

Agitations, l69-l7.

149
CHAPTER III
THE ABSENT IMA
We will not have a man to rule over us
for a man rules with an eye to his own
interest and becomes a tyrant. We will
have the law for our ruler.1
The question whether it be more advantageous to be ruled by
the best man or the best laws 2 would seem to express a
perennial dichotomy in political thinking. Answer in favour
of the former presupposes adherence to the myth of the
supremely wise man, in favour of the latter, the myth of
an Impersonal, formal and just legal system. The actual
rule of a determinate political entity embodies in practice
a resolution of various forces, manifest as power, wealth,
custom, law etc. Noriopoly of power in Islamic states by
military princes of one kind or another whose interests were
best served by the arbitrary application of force or favour
prompted amongst other classes concern for the elaboration
of codes and laws. The outstanding example of codifyin
activity in Islam was that of the fuqah'. Their elabora'tion
of Islamic law reflected a desire for stability and order
over and against anarchy and expediency, and reflected
further the desire of the clerical class to assert and
preserve their authority over and against that of princes.
By the tenth century the Sunni expression of the law was
also an expression of a compromise between jurists and
princes. At all times however the Sunni fugah' preserved
the belief that the rule of law was preferable to that of
any individual: the ruler is or ought to be merely the
executor of the law. The alternative theory, that the
community ought to be ruled by a supremely wise man, (as
opposed to an impersonal just 1aw) also existed within Islam
and is equally best understood as a response to the arbitrary
and apparently unjust rule of those who administered the
early Arab empire. Those who embraced such an idea in
1. Aristotle, Ethics, 156.
2. See further Aristotle, Politics, 3. 15-18.

50
Islam were the Sha. They proposed that the actual
government should be replaced by the one just and legitimate
ruler, objectively identified, at various times, as one or
another of the (sometimes supposed) descendants of the
prophet.
During the tenth century a particular group of
ShI C a, the Irnml or Twelver Shica, formulated the belief that
the imm, the one just and rightful ruler, was absent or
invisible. They were not the first group to formulate such
a theory. Other imms before the twelfth imm of the Imrniyya
had been thought to have vanished. Such beliefs formed part
of a recurrent pattern within Islam of messianic aspirations.
The hardships and injustice of life during the period of the
imm's absence (the Ghayba) became explicable, and people
lived in the hope of his imminent return, when justice would
replace injustice and so on. The ImmI Shi d i faith may be
assumed to have begun thus as a messianic movement; but its
enduring success demands further explanation.
The varied implications of the imm's absence were
worked out by the ShIT scholars In a series of works on the
Ghayba throughout the Buyid and immediately pre-Buyid period.3
Of these works, the K. al-ShifT by al-Sharf al Murtad (d.
436/10 14 14) is one of the more complex. It was written to
refute the theory of the imamate put forward by the Muctazill
dAbd al-Jabbr (d. 1415/1025) in his MughnL The K. al-Shf
was later abbreviated and adapted by Muhammad ibn 1-lasan alTsT took up
Tsi (d. 460/1067) in his TalkhI al-ShfI.
the discussion of the Ghayba again still later in his K. alThese three works emerged from a polemic milieu
Ghayba.
and represent the latest and perhaps the most sophisticated
assessments of the Ghayba produced by ShI C I scholars in the
Buyid period.
That the Immis argued for the necessary existence
of the imm at all times is well-known. Their argument was
styled the Cag13 argument and is in essence attested as early
as Jhiz. 4 Those who accepted the argument assumed that man,
3. For some early examples, see Kohlberg, From Im.miyya, 5214.
4. Pellat, Ghi, 38-39.

51

without a leader, would be in a constant state of strife

and competition. This tendency to anarchy the ImmIs equated
with man's essential inability to avoid sin (his absence of
isma) and they inferred from it not simply the necessity
for a leader but for a leader who is mam, that is sinless
and infallible. 5 This subtlety had not been achieved - nor
TsI noted:
necessarily desired - by Jl?iz.
It is obvious from the actions and conditions
of all those who claim to be imm that they
are not ma'stm, hence we know that the real
ma t sim must be absent.6
A dispassionate observer might be tempted to conclude that
the absence of the imm is an inevitable consequence of his
required Cirna. The definition of the qualities of the imm
rendered his absence inevitable and indefinitely prolonged.
The ImmIs themselves found the reason for his
absence in the fear which the imm had for his life occasioned
by the domination of the unjust (limTn). 1!urta acknowledged that there was a need not simply for the existence
of the imm but also for his independent action (taarruf):
limIn however had prevented the attainment of that end.7
T 1 too indicated that the mukallafin were themselves
responsible for their not enjoying the benefits of his presence because of their causing him to fear, and their ulm
towards him and their usurping his position (taghallubuhum
ral mawidihi). 8 The reference was partly to the generality
of rnukallafln but clearly it was the secular rulers who were
particularly responsible for usurping the role of the imm.
Thus, in establishing a reason for the imm's absence the
ImmIs were led (almost if not quite) inevitably to the
conclusion that all actual governments were ipso facto rebels
5.

writings seems to cover

The term ma'stm in most ShI
both sinless and infallible. For details of the Sh'T
argument see Murta, ShfI, L_ 6; 'sI, Ta1khI, 297-298;
Ghayba, 3, 15.

6.

TisI, Ohayba, 3.

7.
8.

Murta, ShfI, 36, 58.

TalkhI, 301.

52
against the imm, unrighteous and tyrannical. It is
significant that

TsI considered

that the last imm when he

reappeared would do away with kingdoms and conquer every

sultan; he would spread justice and eliminate oppression.9
-

J>i

That supposition reflected the messianic origins of the

Irnrnl sect and implied that until the advent of the imm
sultans and kingdoms would be inevitably unjust (Jawr). A
general tendency to assume that and to deny the legitimacy of
de facto powers has been noted as characteristic of most, if
not all, immI writers.'0
There was a second cagli argument for the necessary
existence of the imm: it stated that, granted the validity
of takllf and the existence of the shari c a, the latter
required a lfiz ma'stm, an infallible preserver. 11 This
was to be distinguished from the sam'i argument which stated
that certain shard functions required the existence of a
unitary executive authority styled imm. Murtad perceived
that to accept the sam argument would imply that one already
knew the details of the sharT Ca: that would undermine th
fundamental ImmI position that the sharlca could only be
known through the imm. Both he and TST accordingly
insistently rejected the samCT argument. 12 The second taqlT
argument conceded not that one knew the sharl'a but that one
knew there was such a thing, to ascertain details of which
an 1mm was required)-3
9. Id., Ghayba, 200.
10. See e.g. Eliash, Juristic Theory, 17-18, 25-26; Keddie,
Roots, 32.
11. See e.g. TsI, Ta1khI, 297.
12. flurta, ShifT, 15-16, 33_314; Tisi, Ta1khi, 318.
13. It may be noted that whereas BqillnI did not explicitly
make appeal to a sam argument, Abd a1-Jabbr, a near
contemporary did. A Nuctazill origin for that argument
seems not unlikely. The Muttazills, unlike the Sunnis,
did not confuse moral and natural necessity: for them,
the imm was necessary for the implementation of the
sharl'a, but he was not thought to exist in the Buyid period.

53

C Abd al-Jabbr, the Mu'tazil3, and Murtad were

in essential agreement that the sharIa required a lfi
m"sim; they differed on the identity of the
According to Murta:

o>J;.c

6'LP>

Jt(:)
j2
We knowthat the sharIa of the prophet is
eternal and unabrogated, that it will continue
[valid] and uninterrupted, that worship [taCabbud
- by implementation of the sharica] will remain
incumbent on mukallafs until the day of
resurrection.
Granted that there was a shari c a, and that it was at all
times incumbent on mukallafs to act in accord with it,
Murtad concluded that to leave the shari c a without a 1'fiz
would constitute neglect, ihrnal, on the part of God, and
It followed that there must
would be taklTf m l yutq.
be a Ifi and he must be free from error. That the sharica
should be preserved by the urnrria was, according to Murtad,
impossible, for it, the umma, was not matim. Hence there
had to be an imm masim who was 1tfi of the sharPa)5
The logical premises of that argument need not here be
questioned in detail; it may however be noted that the
whole argument was formulated in opposition to a belief
that the umma was the hfiz ma'stm of the sharl'a. It was
therefore a part of ImrnI polemic to belittle the umma: it
was no more than a gathering of fallible individuals who
were no less fallible in their collectivity than they were
1. "The commission of something impossible". The ShICa
believed that God's justice ensured that His commission
to man (takllf) would be in accord with man's capacity.
God may not charge man with something impossible nor may
He require something without ensuring that man has the
means to knowledge of what is required. To commission the
impossible would be evil (qablli) and is not conceivable
on the part of a just God.
15. Murtad, Shfi, 34; TsI, Talkhi, 308.

514

In their individual status)

The polarity of imm and

umma constantly emerged as a central feature of ImmI polemic.

The Irnmls considered that If an amir appointed by the Imm
should err, the Imm would take him by the hand, that Is
lead himaright. This was in opposition to the 1u'taziiI
assertion that if the imm should err the c u1am al-umma
would take him by the hand)' 7 The assertion of the authority
of the umma over and against that of the imm had already
featured as the main point of BqillnI's discussion of the
Imamate. It was also the central feature of the Mu'tazilI
argument, or at least of Abd al-Jabbr's argument. Both
Murtad and T1 objected to 'Abd al-Jabbr's views on
the grounds that he made the umma imm to the imm:
b>-

jj.

\.k . t
&

,i L

Li

'

Lj)

LLt

L J

As to your assertion that the Imm, if he

errs, will be guided by the cularnT al-umma, It
is a clear statement that [the members of]
the umrna are imms of the imm.
Their statement that the umma Is 'behind'
the imm is absurd because, were the
situation as they describe It, it would be
necessary that the umma be Imm to the imm)'8
For CAbd al-Jabbr, as for BqillanI, the umma was behind
the imm to lead him and guide him aright; for the Immls,
the imm was behind the umma to lead it and guide It aright)'9
The means whereby, according to t Abd al-Jabbr
16.

Ibid., 311.

17.
18.

Murtad, Shf,

19.

Cf., above, 3-35 ; see also Wansbrough, Studies,

Ibid.,

142;

142.

Talkhls, 319.
54;

and note further Qummi's emendation/exegesis umma:imm,

in ibid., 162, n. 14 and 2145, n.7.

55
and BqilinI, the umma preserved the shar'a were primarily
ijm C and tawtur. The ImmIs, who were inclined to make
some show of rejectIng the validity of these means, were
required by their interlocutors to demonstrate, in that case,
and in the absence of the imm, how then the sharl'a might
be preserved and known. They conceded that the Qur'n of
itself was not of much avail.
It is not permissible to restrict oneself
to the Book in preserving the shari c a, for,
as to most of the details of the law,
there is no explanation of them in the plain
[part of the Book] ...; there must be an
Interpreter and an expounder (mutarjim,
mubayyiri). If it be said that the prophet
is such, we do not deny it, but for those
who did not witness the time of the prophet
there must be a way for the explanation to
reach them ... [Since tawtur and IjmC are
not valid paths] we are compelled to return
to the fact that there must be a Iiujja who
passes on the prophet's explanation as to the
contents of the Book.2
Further, ijtihd and giys, as elements of legal methodology
were equated by the Shi'I scholars with ann, (opinion, as
opposed to 'jim, knowledge) and rejected.
As to ijtlhd clear proofs reveal the falseness of what you call ijtihd; one of these
proofs is that Ijtihd In the shari'a is
according to you a seeking for a dominant
opinion (ghaiabat ann) in matters which
have no indicator (dalil);

but ann has

no place in the sharl'a and it is not valid

that the ta}irlrn and talilIl of a thing should
be established by opinion.21
As to akhb gr al-hd:
The situation of these with regard to the
20. Murta, ShfI, 36.
21. Ibid., 31-32; cf. also 36, where ijtihd = qiys;
TsI, TalkhI, 306-307, where ijtihd = ra'y.

56
sharl'a is clearer than all that has gone
before: for they do not produce knowledge,
and they are contradictory and they come
down to us with different and opposing
rules 22
partly because his own collections of Traditions contained many that were considered was inclined to take
a less extreme view on this point.23
The ImmIs rejected too ijm and tawtur: these
did not of themselves constitute valid paths to knowledge.
However their rejection was tempered by certain considerations. Ijm', in so far as it comprehended the person of
the imm, could be recognised as a reliable source of knowledge.
Hence, according to us, ijm' i not invalid
(bil): for evidence has shown that amongst
those who constitute ijmt is the mai3m,
ujjat allh, so it is riot possible that ijm
be attained on anything false.2
Ijm then was one of the means whereby, during the Ghayba,
the imgm exercised his function as lfiz al-shara. But
it was the ijm c of the ImmIs that mattered, not that of
the rest of the community. The IrnmTs, though they rejected
tawtur as a reliable source of knowledge per Se, nonetheless,
Murta
conceded, relied on it to establish revealed knowledge (samCiyyt). But, they recognised also, in contrast
to the Sunnis and CAbd al-Jabbr, the possibility that the
transmitters might neglect, accidentally or deliberately,
something that was entrusted to them. So, while it was true
that all that reached a particular generation by tawtur
constituted a Iiujja, reliance on tawtur alone would risk
the danger that part of the sharlca might be lost. 25 In
fact, according to
We could not be sure that what they transmit
22. Murtad, Shf, 36.
23. See below, chaps. VII and VIII,esp. 210-215.
2L1 . Murtad, ShfI, 11; cf., TsI, Talkhls, 311-312.
25. Murtad, Shfr, 11.

57
to us is not the opposite of what they
heard; nor could we be secure against
that happening which is possible [i.e.
that they should neglect or abandon part
of what was entrusted to them] except
in so far as we are certain of the existence
of a ma'sm 'behind' the mutawitirin.26
Tawtur, in fact, became a buija precisely because the imm
existed. As a proof, it was therefore logically posterior
to the Cagli argument for the necessary existence of the
imm. Denial of the validity of tawtur per se and recognition
at the same time of the need for tawtur in order to establish
sam t iyyat provided paradoxically an extra argument proving
the necessity of the imrn:

L)

t;t

If deviation from a transmission is possible

for them [the transmitters] then their need
for him [the imm] is established; so that
he might be behind them to rectify transmission when they deviate, either by himself,
or by those who [or, he who] constitute[s]
a liujja.27
The ImmI position was then that tawtur constituted a
ujja. What arrived by tawatur was to be accepted, for,
If anything went wrong, the imm would have to appear himself, or cause the error to be rectified by "those who/he
who constitute/s a Iujja" That phrase, during the Ghayba,
could only be a reference to ijm
because that, along
with tawtur, was thought to be the only available liujja,
(unless the sufar), the agents of the imm during the
lesser Ghayba, constituted such).28 That which was preserved

26. TsT,

TalkhI, 308.

27. Ibid., 311; cf., also 308-309.

28. I am now Inclined to see in 'he who constitutes a Iiujja"
a guarded reference to acceptable transmitters of a khabar
W&Ilid; see further, below, chap. VIII.

58
by tawtur was of course Traditions from the prophet and
from the imms, all of whom were equally authoritative as
exponents of the law.
The Sunnis and the Mu'taziils (cAbd al-Jabbr)
put forward the argument that the umma was the infallible
Ifiz of the sharl e a. The ImmIs denied this. But the
IrnmIs admitted that during the Ghayba they were (like the
Sunnis and the Muctazills) dependent for knowledge of the law
upon Traditions preserved by tawtur, the Qur'n (likewise
preserved by tawtur) and ijm'. It is clear that the polemic
about who preserved the sharl'a hid a fundamental agreement.
All parties agreed that the law was known by recourse to
revelation (preserved by tawtur) and ijm'. Though the
Immls rejected qiys and ijt1hd they had exegetical and
other methods to extrapolate the law from revelation or
otherwise define it. 29 The insistence by the ImmIs that an
infallible imm was necessary served only to define the
community for whom alone ljmc and tawtur were valid proofs.
The irnm was present in an Immi ijmac and behind Imm transmitters. The polemic was thus far less about the nature of
knowledge than about whose knowledge was authoritative.
The Imml argument was so structured that only the Imm
fuqah' (not Sunnis or Mu'tazills) could have access to the
true sources of authority, namely Traditions from the prophet and imms preserved by the ImmI community. It is to
be noted that the Imrn acceptance of a law based on revelation (Traditions and Qur'an) had been implicitly enunciated
at the time that Kulayni (d. 328/9 L10_l) produced his KfI.
The production of a body of canonical Traditions from which
to derive the law signalled the IrnrnT acceptance of the
idea that the law was known by the preservation and interpretation of texts. The arguments that were elaborated by
later writers such as TsI and I'Iurtad aimed at justifying
a development to which the ImmIs were already committed.
The paradoxical nature of the Imml contention
that the imm was both necessary to Iifz al-shar'a and

29. See below, chaps. VII and VIII.

59

absent prompted from their opponents a number of attacks

which in the event sharpened Shi d i polemic to the realisation that the very absence of the imrn only confirmed
that complete knowledge of the sharl'a was available.
We do not say that we need an imm at
all times in order to know the law, rather
so that we might be sure of its having
reached us. We can be sure of this during
the Ghayba because of our knowledge that
if the transmitters neglected a part of It
which we ought to know the irnm would appear
and himself explain the matter.3
Polemic confrontation over the function of the imm and the
means of preservation of the sharIa concealed in the end
a basic similarity: for Sunnis, Mu'tazills and ImmIs
custody and knowledge of the law lay with the community,
pre-eminently with the fugah' who became thereby Immediate
sources of authority within the community. That was the
major conclusion that emerged from polemic about the Ghayba.
Whereas the earlier manifestations of Shiism had been
associated with the desire to be ruled by a supremely wise
or completely just man, the post Ghayba (ImmT) development
stressed that In the absence of the righteous leader there
was at least a law to be obeyed. The ImmI fuqah', when
they Insisted that the imm was maurn and absent, insisted
too that his function as custodian and enunciator of the
law had passed to them, and that, in carrying out that
function, they shared his ima. It was, I would suggest,
precisely because the theory of the Ghayba could be interpreted to that end that it was accepted and refined by a
group of fuqah' who made this the basis of their own
authority. 31 That the theory of the Ghayba could In this
way become the foundation for an articulation of clerical
authority would appear to be a major factor accounting for
the enduring success of the Immi faith. The IrnmI fugah'
30. Murtad, Shfr,

58;

cf.

Ghayba, 60.
31. For further discussion of all this see below, chaps.
vil-Ix.

60
followed the example of the Sunnl fuqah' in elaborating
a law which was an expression of their consensual interests
and an assertion of their authority.

Discussion of the sources and preservation of

knowledge arose out of the second 'aqil argument for the
necessary existence of the imm (that the sharl c a requires
a Ifi masm). Acceptance that the imm was both ma'im
and absent had pointed also to certain political conclusions,
amongst which was the Imml assumption that all de facto
governments were essentially unjust. That too may be understood not simply as a fortuitous result of the Ghayba (there
might always have been another imm 32 ) but as a factor
accounting for the adoption, propagation and success of
the ImmI theory of the Ghayba.
The question, why the Ghayba, "why the Ithn
asharI Shltites had ... excluded the possibility of setting
up an immr CAljd caliphate", 33 was raised by H.A.R. Gibb
who went on to propose, for an answer, political quietlsth.
Adoption of the doctrine of the Ghayba represented a rejection or deprecation of the right of revolt. A similar
observation was made by Montgomery Watt.

314

Gibb related

the development of the doctrine of the Ghayba to the unrest

and disorder created by Isma'ilism in Western Asia. 35 It
has already been suggested that it was In response to that
unrest and disorder that the Sunnis, from about the beginning
of the tenth century, began to perceive in the actual
political situation a satisfactory approximation to their
religious ideal. At about the same time the ImmIs perceived
that the imm was absent. 6 They thus effectively denied
32. Cf. Kohlberg, From Immiyya, 524, 533-534.
33. Gibb, Government and Islam, 117.

314 . Watt, Reappraisal, 653.

35. Gibb, Government and Islam, 118.
36. For the relatively late date for the acceptance of the
Ghayba and its implications, see Kohlberg, From Immiyya

521_5214.

61

that the political situation, the actual power, could be

an approximation to the religious ideal:

the only just

and rightful ruler was not available. That which was

available was not and could not be just. Thus the need
to preserve values and interests in the face of IsmCIlI
disruption prompted amongst one section of the scholarly
class (Tabarl, Bqil1nI et al..) an alliance with the
princes, and amongst another section (the ImmIs) dissociation
from the princes. For the ImmIs, all governments would be
equally unjust, pending the return of the one true absent
imm.
The ImmIs were however, no less than the Sunrils,
politically quietist: recognition that the ideal state
could not for the moment be achieved (because the ideal
ruler is absent) permitted a variety of approaches to the
de facto power, one of which was co-operation.37
The Sunnis had distinguished between the imm as
repository of knowledge or enunciator of the law and the
imm as political power or executor of the law. The ImmIs
defined their imm as both repository of knowledge and
executor of the law. In his absence his function as
repository of knowledge (lfi a1-shara) fell to the (ImmT)
community more especially the fuqah. Who then, if anybody,
during the Ghayba, undertook his executive duties?
The duties associated with the imm's function
as executor were never precisely defined in the works at
present dealt with. They were enumerated in lists which
almost invariably ended with the words, il 1-khira
(etcetera). These lists included igmat al-iudd (the
imposition of penalties), tanfrdh al-aIkm (putting judicial
decisions into effect?), qismat al-fay' (distribution of
booty), I2ifz al-baya (defence of the heartlands), alt aljum c a (Friday prayer). 8 Discussion of these duties arose
out of the ImmT rejection of the samdl argument for the
necessary existence of the imm. Murta, while noting
that the existence in the sharlLa of such functions did
not constitute a proof for the necessary existence of' the
37. Cf., Ibid., 533.
38. See Murtad, ShfI, O;

Talkh, 318.

62
imam, conceded that it was the imm who must carry out
these functions, (and who must therefore, he claimed, be
ma C tim) . 39 r Abd al-Jabbr had suggested that the immrs,
because their imm was absent had made Iiudd, a1krn etc.
sgi, that is, had caused them to lapse. Murtad initially
denied this:
}Iudd are not sgi during the time when
the imm is unable to appear and impose
them; they are in fact thb1t with respect
to those who deserve them. If the appearance
of the imm overtakes these people then he
will impose Iudd upon them. If, on the
other hand, his appearance does not overtake
them, God will undertake retribution and
forgiveness on the last day. The sin attendant
on delaying 1iudid and preventing their effect
lies with those who caused the imrn to fear
and compelled him to disappear.
Irudd here was used as a paradigm intended to include reference to the various other duties of the imm (tanfldh al-aIikm,
gismat al-lay' etc.) all of which, it was implied, had not
lapsed but were postponed. They remained thbit, but their
execution was delayed until the appearance of the imm. The
sin attendent on this delay lay with the
11mIn, not with
the immrs. Murta however seemed less than satisfied with
his own answer: he 'returned,' the question to his MutazilI
interlocutor.
Tell us about Iiudd in those circumstances
when the ahl al-ikhtiyr are unable to choose
an imm; what do you say, are they sqit or
thbit? If he [cAbd al-Jabbr] says that they
are thbit with respect to those who deserve
them and that the sin ... lies with those who
prevent ikhtiyr; then we say this is our
answer too. If he says that ucid are sqi
as long as there is no imrn to carry them out
then what prevents us from saying that hudd
39. Murtad, ShfI, LEO.

63

are sgi during the Ghayba?

Murta, in other words, seemed not absolutely sure what
happened to the functions of the imm during his absence.
He was Inclined to deny the epithet sgi but would in an
ad hominem argument if pushed accept that term. Whatever
term he used it is perfectly clear that liudd (etc.) were
not validly carried out during the Ghayba. Or, reversing
the argument, that which was carried out during the Ghayba
and was claimed to be budd was not In fact so: the real
shar'I budd waited upon the return of the imm.
argument on the same point displayed a new
subtlety, based on a distinction between suq (lapsing)
and naskh (abrogation).
Q. What is the status of Iudd during the
Ghayba? If, in the form required by the law,
they become sqi then this constitutes naskh
al_shari t a. If they remain valid who is to
exact them?
A. }iudd, once merited, remain valid with
respect to those who deserve them. If the
imm appears and those are still alive he
will administer the penalties ... If however
the opportunity to do so passes away on account
of the death [of the criminal] then the sin
attendent on this lies with those who caused
the imm to fear and forced him to disappear.
This is not naskh of 1udd, for it is only
Incumbent to exact 1iudd when it Is possible
to do so; when something intervenes to prevent this then ludd become sqit. There
would only be naskh if exaction of 1udd
were sqit even when it was possible to carry
them out.41
Murtad's dilemma had arisen out of the fact that he did
not like to suggest that any part of the shari'a should be
sqi: the sharl'a in its entirety must be valid at all
times. By playing on the distinction between naskh and
L O. For the whole of this argument see Murta, ShfI, O-4l.
Lfl TsI, Ghayba, G.

61

suq T' was able to insist on the continued validity

(not manskh) of the sharr c a while conceding that the imm's
functions were sgi1. What seems again to emerge from this
argument is the essential and inevitable illegitimacy of
de facto powers: attempts to impose }iudd during the Ghayba
were illegitimate acts, signs of the usurpation of the imm's
authority. The real shard
udd lapsed pending the reappearance of the imm. It is important however to note that,
while the ShII fuqah denied the shar'I legitimacy of
the de facto powers, they proposed no political alternative.
By stressing that the imm's functions were sqi they could
present themselves as politically neutral: unlike for
example the IsmIlIs they had noalternative ruler in mind.

These three works (the Shfl, the Talkhs al-Shafr

and the Ghayba) were the products of a tolerant political
regime which permitted or even encouraged (up to a point)
sectarian discussion. They were public works aimed at the
heterodox as well as at believers. The arguments they
presented were frequently of an ad hominem nature designed
to answer a particular criticism or to defend a particular
belief. In any polemic encounter the ImmIs would be
challenged on the question of the imamate (as they would be
challenged on any one of a number of themes) and their
response would be partially conditioned by the nature of
the challenge. Because of their public, polemic and ad horninem
nature such works did not explore the full implications within
the community of a particular stance; they were concerned to
defend that stance outwardly against other possible stances.
It will be seen for example that the executive functions of
the imm, though they were indeed partly sqi, had been
to a considerable extent delegated to the fuqah
(or so it
was thought even in Buyld times. That delegation however
received little attention in polemic works. Perhaps its
political implications made the scholars cautious in expressing it. These works however indicate the nature of the
problem which faced the fuqah' when, in a different context,
they elaborated a structure of authority applicable within

65

the ImmI community. Their problem was two-fold: how,

in the absence of the imm, to preserve and articulate
the law and how to ensure (or otherwise explain away) its
execution.

PART II

66

CHAPTER IV
THE JUDICIAL DELEGATION
The term authority may as a matter of conceptual
utility be defined as the right to demand or enforce obedience.
It is thereby distinguished from the power to exact obedience.
P. Winch in opposition to R.S. Peters 1 emphasises that
authority is "conceptually inseparable from participation in
rule governed activities" arid implies always choice. 2 The
exercise of power implies coercion. The relevance of 'rule
governed activities' lies in the fact that the concept of
'a right' only has meaning within a codal system, whether
legal, moral or customary. Different systems of rules, of
whatever kind, current amongst communities or groups, will
make different provision for authority. Weber in addition to
legal-rational and traditional concepts of authority
distinguished also a charismatic type. 3 That was a type of
de facto as opposed to de jure authority: at first sight
certain personalities (Weber cites Christ and Napoleon) seem
to derive their authority not from a system but from something
intrinsic to themselves. Winch objects to this idea and
suggests that all three Weberian types presuppose the existence of a tradition: for him the charismatic type is comprehensible only within arid in opposition to a recognised
system.
In the case of examples such as Christ one might
suggest that certain people recognised his authority not
because of the qualities or acts which he manifested but
because of a pre-existing convention that such qualities and
acts occasiond rights. Authority is not objectively to be
found in this person or that; it is a subjective concept
1.
2.
3.

Quiriton, ed., PolItical Phi1oso1, 83-111.

Ibid., 102.
IbId., 86-87, 107; Weber, On Law in Economy and Society,
33L...337.

k. Quinton, Political Philosophy,

107-108.

67

conferred or recognised in accord with social and group

conventions. Within a given society there might exist a
multiplicity of conventions. In a hypothetical Islamic
city for example there may be a Sunnr ruler but both SunnTs
and Shi c is in the population. Both groups submit to the
ruler's de facto power, but only the Sunnls acknowledge his
authority - his right to wield that power. The ShIs may
in all outward respects obey and submit no differently from
the Sunnrs but they do not recognise his right: they
acknowledge his power not his authority.
Conventional systems implying concepts of authority
are the products of and are fashioned and moulded by socIety
- as well as, once extant, fashioning and moulding. The
existence within one social or cultural complex such as
Islam of different groups with different theories gave rise
to problems, for all groups interacted and few individuals
belonged to only one group. The product of the tensions
thus created was, in the expression of a theory, intellectual
contradiction or casuistry, in the exigencies of daily life)
personal moral dilemmas. Of the systems of authority curt'ent
in the Islamic world only a limited number are available for
study, the product of the literate classes: the poets (Sperl),
the bureaucrats (Lambton), the philosophers (Rahman, Rosenthal)
and the legal scholars. No doubt there could be more groups
and further more subtle subdivisions but it is immediately
to be noted that the interests of each of these groups were
different, the degree and quality of their interaction with
cther groups varied and their contacts with de facto powers
were closer or more distant. Conce p ts of political authority
within each group are likely to be different. Such relativism
if perceived was rarely acknowledged by those withIn a
particular system: for the immrs, the details of their
system, the "rights" for example of the imm, were objective
truths.
A legal system reflects the needs and aspirations
of the society, or part of society, within which it emergs;
as such it is liable to change and develop in response to the

68

contiiigencies of time. Few societies however prior to

modern times have been willing to concede the subjective
or local nature of their legal systems. Where such perceptions have been achieved (as e.g. by sotne Islamic philosophers 5) they have been frequently accompanied by the
elitist view that for the masses at least that truth should
be hidden. A legal system if it was to be portrayed as
something external to and binding upon the whole of society
had to be derived from and linked to something apparently
external, unbiased and unchangingly valid. It might be
linked toamythological past. Such mythology is "exemplary,
paradigmatic: not only does it relate how things came to
be, It also lays the foundation for all human behaviour and
all social and cultural institutiors." 6 Such mythic systems
in fact are frequently local arid Imply acceptance that other
localities with a different Hero/Ancestoi,'Past will have a
different law. "We must obey and do whatever the custom of
the land requires." 7 The assertion that a legal system is
universal mIght be accompanied by some kind of link between
the law and the Intellect or the law and God or some combination of the two, God and the intellect having this In
common that they are 'objective' facts of 'universal' si.gniulcance. Plato links nomos and nous 8 . Philo nomos and logos9
and a variety of theories of Natural Law relate intellectual
principles to God's command.' For the Sunnis the law was

r.

5.

Rahman, Prophecy; Rosenthal, Political Thought;

216-222.
below, chap. VIII,

6.

Eliade, Rites and Symbols, x-xI; cf. Id., The Myth, 3-48.
Antigone in Oedipus at Colonus, Penguin Translation.

7.
8.

Barker, Greek Political Theory, 300.

Goodenough, By Light, 56, 128.

9.
10. Cf. Aristotle: He who bids the law rule may be deemed to
bid God and reason rule: Politics, lO. See also
Friedman, Legal Theory, 17-73.

69

simply the product of God's command - totally free of the

restraints implied by reason. For the Mu'tazills and the
ShI t jS God's commands were in harmony with or supplementary
to reason. 11 For Islam as a whole, as for Judaism, God's
command was conveyed to man via a prophet who was as it were
archetypal law (whether by virtue of blind obedience or
perfected intellect was a matter of shifting dispute in
both religions). In such systems the mythic founder or
renewer of the community is given universal significance
by linking him to God and/or reason.
That law was generally portrayed as an objectively
ascertainable and constant 'fact' can hardly be found surprising, Islamic law may however be more usefully defined
as that which Is written down In works of figh, the literary
creation of eminent fugah') 2 Notwithstanding the apparent
tautology of that definition it defines precisely what is
available to us and serves a purpose In so far as it stresses
that Islamic law Is the expression of one social group with
its own concerns and Interests, variously interacting with
other social groups. Islamic law is not the same as that
which was administered by gIs' courts. About the latter
for the medieval period we know little or nothing. It is
certain that the gI's competence at best covered only a
fraction of Islamic law but even with regard to that fraction
the transposition of the ideal into practice cannot but have
involved change of some kind. Even where it is known that
a particular writer (Aba Ysuf, Ibn Taymiyya et al,) was a
gdI his writings do not represent except dimly and in part
only his judicial practice. Islamic legal works set out a
11. Cf. Wansbrough, Sectarian Milieu, 110-112; also below,
chap. Viii.
12. Occidental commentary on Islamic law has frequently
distinguished between the narrowly 'legal' part and
the rest; see e.g. Schacht, Introduction, 200-201.

70

particular theory of judicial practice but the relationship

of that theory to what actually happened must always be
difficult to ascertain.

Kulayril (d.

328/9 L

O_l), in his collection of

Traditions, begins the chapter on q.ac' with a section
entitled inna l-hukirna innam hiya li'l-imm. Characteristic
of the Traditions there cited is 'All's remark to Shurayh:13
,'

'

1 s

Li

You have taken up a position which none may

take up except a prophet, the waI of a
prophet or a miscreant.
The aim of the section as a whole is to demonstrate that
judicial authority belongs exclusively to the prophet or the
imin (wasi nabi); and conversely that other claimants to
such are miscreants, illegitimate. Centralisation of
authority in the lmm does not however in post-Ghayba ImmI
works indicate desire for or recognition of a highly centralised or hierarchic community, for that centralisation was
coupled with absence: the centre, the source of authority
was absent.
Sections 2 and 3 of Kulayni's chapter discuss the
dangers of q4' with its correlative reward and include
the much cited Sh3'T tradition that there are four types
of
three in hell, one in paradise. Section 1, inna

gcI,

1-mufti min, emphasises the distinction between muftis

and rnustaftls: the latter may safely accept and submit to
all that the mufti enjoins, for it is the mufti who is
13. For details of Shurayh, see Tyan, Histoire, I, 101-lO1.
lL For Sunni assessment of these dangers, see Goldzlher,
Muslim Studies, II, 147_149; Tyan, Histoire, I, 20;
Coulson, Doctrine and Practice.

71

responsible, the mustafti gains salvation by submission.15

Section 5 introduces the theme of working for the profane!
illegitimate government:
Ab Abdullh (Jafar al-Sdiq) was asked
concerning a gI ... who accepts rizg from
the sultan for gag'. He said: That is
abomination (sulit).
This theme Is taken up in section 8, kirhat a1-Irtif'
ha. qut al-jawr. Included in that section are two much
cited delegation traditions which are given here in full.

(j)

j.t d' L
L L
Lr

. (L .
f

.)'
From Ab

&

Abdullh: Let not one of you call

another to litigation before the ahl al-jawr.

Rather look to one of your number who knows
something of our judgements and set him up (to
judge) between you. For I have made him a ____
so seek judgement from him.

(ij)t

k) L? J

L:)I L'

JL

Lii' J,t LULJtcs J1 J L)?'J --'

f'L
La9
JLi
C

ep\>.?

. '

)JU

>i t cu

'

IlL
11

))LL%,

LJ

From Cumar ibn Hanzala: I asked Ab 'Abdullh

concerning two of our companions who are involved
in a dispute over debt or inheritance and who
seek judgement before a sultan or

gis.

15. See below on Ift', chap. IX, esp. 2382L2.

Is this

72

hal1?
Ab Abdu11.h replied: He who seeks judgement
from Tght arid receives judgement receives
only abomination, even If his claim is valid,
because he has accepted the decision of Tght.
God has commanded that (such a one) be considered
a kfIr. I said: What should they do?
He said: Look to one of your number who relates
our hadith, who considers our 1iall and our 1arm
and who knows our a1km; accept his judgemerit.
For I have made him a bkim over you. If he
gives a decision in accord with our judgement and
(the litigant) does not accept it, then it is
God's judgement he has scorned and us he has
rejected. One who rejects us rejects God and
he is Cal 1:iadd al-shirk 16
These are the traditions of ex ante delegation referred
to by Binder in "The Proofs of islam". 17 Now, the opaque
quality of elements of revelation and their susceptibility
to exegesis ( D eut ungs b edtlrftigkelt l8 ) render it invidious
for an observer to assess what meaning or meanings they will
eventually be thought to have within a developing tradition.
Nonetheless certain points may usefully here be elaborated.
Jacfar a1-SdIq Is portrayed as conferring judicial (and to
all appearance only judicial) authority upon certain of his
followers. The delegation takes place from a central but
absent source of authority to a diffuse, numerically unspecified, and hierarchically unranked group. The defining
criterion of that group is their knowledge of the imams'
These delegation traditions may
judgement, i.e. liadlth.
then be taken to represent the conceptual movement from
16. Kulaynl, Furti C al-Kf, I, 357-359; cf. also
17.
18.

Tandhb, II, 68.

Binder, The Proofs of Islam, 122.
Wansbrough, Quranic Studies, 100.

Tsi,

73

infallible ruler above the law to rule of law. 19 The negative

implications of the Ghayba, also discussed above, are
further confirmed here: the actual governments (sultan
and
are ipso facto Tght and their Judgements even be it noted when they are in accord with the law - are
invalid. Stress on the uniqueness of the imm's authority
when he is present ensures that in his absence authority
depends on correct Interpretation of his words (revelation)
and so authority is confined, during the Ghayba, to the
clerical class: those who extrapolate law from sacred texts.
Rule of law replaces the rule of the charismatic leader.
Emphasis on the centrality and superiority of the Imm's
rule, when present, gives place, in his absence, to an
unranked unhierarchic system. Recognition of his rights
renders invalid the pretensions of de facto governments.
The ImmI ShII community reacting against the arbitrary
tyrannies of the Islamic world expressed its Ideal in terms
of a community ruled by a rigorous law, administered by
knowledgeable magistrates and free from the injustice of
de facto governments.
Mulianunad ibn Hasan al-Tsr was a pupil of Mufid
and Murtad. His K. al-Nihya, on furti c al-figh, though
relatively small by general Islamic standards, was yet at
the time of writing the largest Irnm work of its kind.
In the introduction to his much larger work of fur c , the
Mabs, he indicated that he wrote It many years after the
N1hya and further that he wrote it with a particular purpose
in mind. The Sunns had mocked the ImmIs for their inadequacy in legal studies, in particular their paucity of fur.
Since the Nihya was, according to Ts himself, the largest
ImmI work of its kind, the charges clearly had a certain
amount of truth behind them. ShI'i legal studies were still
in an early stage of development. TsI composed the Mabst
however precisely to refute t1ese charges and to demonstrate

QIs)

19. Cf. above, chap. III, 9-5O, 58-60.

7L

that the ShIs too could proliferate furti'. The resultant

work combined immense detail with inadequate organisation
and undoubtedly involved a large amount of borrowing from
Sunni sources, - a point conceded by Ba1irnI who noted that
TsI followed in some of his works maslak al-'mma. 20 Further
there is evidence In that work of deliberate reticence
(tay - it was abed at a Sunnl audience) for example in
the omission of a section on anr arid nahy, and the neglect
of the ex ante delegatton.
In the k.. ai_qat of the Mabs the appointment of
a Judge (tawlIyya) is discussed entirely on the assumption of
the 1mrn's presence. If the 1rnrn knows a person suitable for
a post he should appoint him; if he himself does not know
someone he should consult with ahlal-'ilrn.
If the imm
knows that a particular area is without a q
he must send
one. When the imrn finds someone suitable he should appoint
him, write for him a letter of appointment.., instruct him
to deal with the affairs of shuhd, af.l and wuqf. 21 Later
In the same section 22 some restrictions were imposed on the
Imrn's freedom of choice: If he desired to appoint a g
then if he found one willing to undertake the task voluntarily,
without payment, he should appoint him and should not appoint
one who seeks a rizq: in the absence of such a one he must
appoint a gdl and provide for him (yarzuquhu) from bayt
a1-ml.
Still later the phenomenon of delegation was again
taken up 23 with a view to stressing the Imm's rights to
appoint a judge both In his place of residence and elsewhere.
This was associated with a general assessment of delegation
based on Traditions indicating on the one hand that the
prophet Instructed others to judge in his place and on the
other that others did not judge without his permission. This
might have led on to the question of delegation during the
airnI, Lu'lu T at al-Balirayn, 297; see also
21. Ts, Mabs, K. al_Qa ? , 2-3.
22. IbId., 20.
23. IbId., 22.
20.

below,ll0,n.11;113,n.2
127,n.119.

75

Ghayba; it did riot. If indeed the delegation attested

in the ShI'i Traditions is mentioned in the Mabs I have
been unable to find it.
On the general nature of the incumbency of g'
is, in keeping with its professed aim and its
the Mabs
title, expansive. Qa' 15 a far. 'ala l-kifya; if
carried out by a sufficient number within the community it
is no longer incumbent on the rest. Conversely if the
community do riot carry out the duty to a sufficient measure
they have rebelled (kharaj), they have sinned (athim) and
the imm must fight against them.2k It Is to be noted again
that the imrn's presence is implicit. In spite of Its
incumbency Tsr concedes, in harmony with the Traditions made
available in Shi d i ( as well as of course Sunnr) sources, that
there is a tension between the dangers of ga' and its
rewards. He cites in addition to the Tradition which places
three Judges in hell and one In paradise, the famous
Tradition about a g4I being butchered without a knife.25
The dangers however are not taken very seriously and Ts1
concludes that one who knows the science of judging and Judges
according to al-agg will be rewarded; al-liagg means ShI'
law. Furthermore since qa' Is a far. It is better to be a
g3 than to deal with e.g. a trade which Is merely mubi.
Conversely if one can gain a livelihood by e.g. teaching
religious sciences (a duty which transcends gag') one
should avoid qat.26 Implicit in this is a distinction
between the higher echelons of the clerical elite (who
should avoid qa') and the lower echelons.
The whole question of the incumbency of qa' is
schematised as follows:
a. Those for whom it is l)arrn: that Is one
ignorant or one fsiq.27
24.
25.
26.
27.

Ibid., 2.
ee note 14 above.
Mabs, K. al-Qad', 2.
Note the greater lattItude of the Sunnls: Tyan,
1-listoire, I, 241-242, 244-245.

76

b. Those for whom it is wjib; that is a

fully qualified candidate who is the only
one available; the imm must appoint him,
he must accept.
c. Those for whom it is j'iz: that is one
amongst several who are suitably qualified
for the post. If the imm calls on such a
one to accept the post some say he must
obey but in fact he may refuse. Concerning
those for whom the post is j'iz it may
become mustal3abb under certain conditions:
(i) Income: if a man does not have a
sufficiency (kifya) he should (yustaliabb)
take up a Judge's post because in so doing
he is obeying God in the matter of al-naar
bayna l-ns, and he will receive a rizq
which will suffice him.
(ii) Fame in f ilm: if one is well-known for
one's t ilm so that people demand to be taught
or seek fatws from one then it is mustaiabb
to avoid a Judge's post. But if one is no
well-known it is mustababb to take up a Judge's
post and thereby display one's excellence and
benefit the people. 28
The whole of this discussion including as it does reference
to the imm (paras. b. and c.) makes no specific provision
for the Ghayba and its applicability to that period is unclear. It does however confirm the distinction between
high-ranking and low-ranking fuqah' and stresses the positive
aspects of qaa' as constituting obedience to God. The
negative aspects ,, the dangers of gaa', are not stressed
except insofar as they provide an excuse for dIvIding the
superior fuqah' (who are recommended to avoid those dangers)
from the inferIor.
28. TsI, Mabs, K. al-Qad', 2.

77

The ex ante delegation from Jacfar al-diq,

strikingly absent from the Mabs, had been carried over
into fur' by Ts many years earlier in his N1hya. In
that work it was the only function of the imm which was
specifically delegated in so many words to the ShI'3 fuqah.29
a ' ),
They have delegated that to the fuahi' of
the ShPa,
For the Sunnis the validity of a r
i's judicial
activity depended upon his direct appointment by the de facto
lmm. But for the ShI C rs the relevant appointment had taken
place in the past: there is no reason to suppose that valid
judicial activity for them required any kind of sanction from
the de facto governments. Rather the reverse, it was not
permissible to turn from a faqTh to a mutawall miri qibal allim. 3 Closely associated with the delegation was the
qualifying remark that it was not permissible to undertake
qat if there was fear of harm to oneself or fellow believers.
Muhammad ibn Idris al-Hull (d. 598/1202) added
little to Tsi's discussion of qa' but generally borrowed
from or expanded on the contents of the Nihya or the Mabst.31
Ja t far ibn al-Hasari al-Muhaqqlq al-Hull (d. 676/1277) was
born and spent most of his life in Hula, the centre since
Saljuq times of Imml Sh11 studies. He lived through the
descent of the Mongols on Baghdad and thereafter benefltted
from close relations with government officials not least
Nasir al-DIn
29.

Tsl. 32

He may reasonably have aspired to a

Nihya, 301; see chaps. V and VI for a discussion

of the role of the fuqah' in zakt, jihd, a1t etc.
The right of the fuqah to undertake

dd during the
Ghayba (ibid.) was a limited concession achieved exegetic-

ally by takhI not by delegation.

30. Ibid., 302.
31. Ibn Idris, SarTir, 161, 191.
32. Cf. Mazzaoui, Origins, 2 !k27, 83.

78

greater degree of practical discretion to implement the

shari ta than his forebears. He introduced in discussing
gag' an interesting innovation lnsofar as he linked that
function to the Implementation of penalties (igmat al-1udd);
that is, he would seem to have read the delegation traditions
as referring not simply to judicial authority but to a
rather wider band of delegated functions. For him during
the Ghayba the fuqah' are permitted to undertake both
gag' and Iiudd as long as they are safe from the mischief
of sul.n aL-waqt. He stresses that permission is restricted
to those suitably qualified. 33 In his Shar'i' al-Islam he
distinguishes fairly clearly between the nature of judicial
appointment prior to and during the Ghayba:
a. The validity of an appointment depends on
the permission of the Imm or one delegated
by him (man fawwa.a ilayhi al-imm). If the
such an
people of a district choose a
appointment is not valid. [This is for the
pre-Ghayba period].
b. But during the Ghayba the qa' is efCective
of a ShI d I LIh who possesses the qualities of
jft34

MuIaqqIq then cited the delegation Traditions given above.

It would appear that whereas while the irn.m is present
a judge requires to be specifically appointed, during the
Ghayba no specific appointment Is required. The possession
of knowledge confers the right to be a judge. Judicial
authority is thus informal - does not require initiation,
confirmation or appointment - and diffuse - Is present in
a number of individuals equally, if they have the necessary
qualities.
Hasan ibn Ysuf ibn al-Mutahhar al-Hull al-'Allma
33. Muhaqqlq, Shari', I, 314143145; Mukhtaar 115, 279.
314. Id., Shari', IV, 68; for the qualities of ift', see
below, chap. IX, 2)412142.

79

(d. 726/1325) whose influence at the court of Oljaytu is

well-known 35 had perhaps even more than a1Muhaqqiq some
right to hope for the discretionary lattitude that would
permit a greater degree of implementation of the sharica.
He reiterates the view that the imrns have delegated Judicial
authority (kmwa-qa') to the Shi t i fugah', that is,
those who are ma'rnr1 arid possessed of appropriate know1dge.6
Some insight into the workings of this delegation is
provided in an assessment of how to choose a valid judge.
If it happens during the Ghayba in a
particular land that there are two fugah'
each of them possessed of the qualifies of
fatw and ukm the choice lies with the
mudda'I to appeal to whichever he wishes.
This is true also if there are a.great number
of fuqah' appropriately qualified. If they
(the litigants) accept both the fugah' as
judges and they each give a different opinion,
that of the a'laii and azhad is effective.37
This passage provides evidence of the informality of the
system envisaged. The knowledge of the faqih together with
his being chosen by the people (with of course his own
consent) are criteria sufficient to create a Judge. The
element of popular 'election' in this process is significant.
A lengthy passage from the Qawid al-AIikm may
sum up 'Allma's position.
a. Appointment (tawliyya) is valid only with
the permission of the imm or his n'ib. [That
is, while the imm is present; cf. from the
TaIirTr: tawliyya if the imm is present belongs
exclusively to him except Insofar as he gives
specific permission...] The people of a district
35. Cf. Balirani, Lu'Iu'a,
36. cAl1ma, Ta}rIr, 158,
37. Id., Talirrr, 181.

224-226.
180;

Qwadid,

45, 223.

80

gI

[that is,
may not appoint their own
again, while the imm is present].
is effective
b. During the Ghayba the
of a faqh jmi' li-shar)1 al4ft'. If
there are many such [reading ta c addada for
ta tadhdhara] the muddaT chooses whichever
he prefers, but if one of them is superior
(afal) it is incumbent to apply to him
during the Ghayba, even if the rejected one
is azhad.8
The transition is again clear from a highly
ceritralised hierarchical system (a) which is displaced
during the Imm's absence and gives way to a diffuse and
emerges in the same way as
uricentralised system. The
the rnuftl, requires the same characteristics and is to be
judged in the same way, the adam being preferable to the
azhad. 39 It is clearly envisaged that for any given roup
of people one or more qualified fuqah' may emerge as
There is rio central authority to which all
potential
qaIs are subject. It may be assumed that the manner of
assessing a qualified faqh was, as it was for the mufts,
various forms of public witness supplemented by personal
choice. The formal denial of this for the pre-Ghayba period
might be a reflection/rebuttal of some errant attempt to
assert for that period what could only be acceptable for the
period of the Ghayba.
On the general nature of the duty (far.) to undertake qad ' 'Allma manifested an increased interest in the
various implications of the problem. He suggested that the
incumbency of qai' 'ala l-kifya was based on amongst other
things the fact that It involved an element of preservation
of the order of the world and justice for the mazlm. These
were Cagli arguments not cited by earlier writers. He
further introduced a distinct modification in the conditions
under which it became mustaIabb to take up judicial posts.

gI

38. Id., Qawid, 223; cf. TalirIr, 181.

39. See below on ift.', chap. IX, 240-2Lt2.

81

Earlier writers, following Ts had asserted that where there

was a plurality of qualified aspirants it was permissible
to take up such a post; it became mustaltabb if there was
lack of money or of fame. 'Al1ma claimed to have demonstrated the isti1)bb of undertaking gad' for those who
trusted themselves to carry out its conditions:
-aJ

fl L>fl

J; L

(%1

That qualification replaced the more rambling qualifications

of TisI. CA1lma quoted from the Mabs
Tus T s opinion
about istibbb and rejected it saying: Our opinion is
better.

bLJ,L,

sJ,

In the Qaw'id a1-A1km he stated simply that

tawliyya (accepting judicial authority) was mustababb for
one who trusted himself to carry out its conditions,
c al! l-a'yn, and wjib Ca1 l-kifya.
LY

c.r

(Sc)
2.

Qa' is mustahabb tal l_a c yan, a subtle enough but

significant advance on previous assessments. Thus CAllma
insisted more strongly than any previous writer on the
general istiI2bb of qa'. His discussion however like
earlier ones included various references to the imm so
its applicability to the period of the Ghayba remains
problematic. In spite of this he provided a clearer assessment of the ImmI concept of judicial authority than earlier
writers, he displayed considerable sensitivity to the
ulilitarian need for magistrates (to preserve the order of
the world) and he laid great stress on the desirability of
accepting such a position. His interest might be explained
14Q Above,
75-76.
41. CAlla, Ta1rrr, 179.
42. Id., Qaw'id, 223.

82

by his presence at Oljaytu's court and his witness to the

sultan's conversion. It is not however to be thought that
a tolerant and formally Shidi ruler was thereby a legitimate
ruler. Such a one might be immensely preferable to an
intolerant Sunni ruler but he remained technically a usurper
of the imm's authority; he might be worthy of the profuse
praise which 'Allma on occasion heaped on Oljaytu, he was
nonetheless a
Muliammad ibn MakkT al-miii al-Shmi, al-.Shahid
al-Awwal, (d. 786/l38-5) was executed in Damascus under the
sultanate of Barqq. He compiled his most famous work of
fiqh, al-Lum t at al-Dimashqiyya, it is said in response to a
request by Shams al-Din a1-wT, with a view to the practical
implementation of Sh1 c 3: ugh in the new Sarbedarid state.
The brevity of that work and its practical aim, led him to
express the elements of judicial theory not only more
succinctly than earlier writers but also with a minimal concern for the problems of the pre-Ghayba period.
Qa' is the duty (waifa) of the im gm or his
n'ib [while the imm is present]. During the
Ghayba the qa' is operative of the faqih jmic
li-shar'ii al-ift'. He who turns away from
such a one to quct al-jawr is rebellious. The
The wilya of a qi Is established on the basis
of common report (shiy) or the witness of two
just men.
This constitutes all of Shahrd I's discussion of the delegation
and attendant problems. Clearly all fully qualified fuqah'
have the right to qa'. Their establishment in the enjoyment
of that right depends on common report or witness. Thus
explicit what was earlier less clear, that the emergence of a
qi depends on some form of popular 'election' - admittedly
informal, a by-product of fame.
143. For 'Allma's praise of Oljaytu see e.g. Minhj al-Karma, 2.
1414.

Shahld I, in al-Rawat al-Bahiyya,

914-95.

83

'All ibn Husayn al-Karaki al-. Amill, al Muhaqqiq

al-Thnl, (d. 937/1530-1) Shaykh al-Islam under Shah Tahinasp
added nothing of significance to 'Allama's discussion of
His younger contemporary Zayn al-Din ibn 'All alqa'.
mill, al-Shahid al-Thni, (d. 966/1559) who spent most of
his life in Ottoman territory and was finally executed there
His commentary on Shahid I's
proves more rewarding.
remarks about judicial authority provides a definitive
clarification of the precise nature of that authority. He
specifies that it is while the imam is present that qa'
is the duty of the imam or his n'ib.
N'ib then here, as
in most contexts if not otherwise qualified, is a reference
to a contemporary of the lmm. During the Ghayba the qa'
is effective of a faqih who possesses the conditions of ift'.
These conditions Shahld 2 discusses in considerable detail.
Thereafter:
If a mufti is confirmed to be possessed of
these various qualities it is incumbent on
the people to take their cases to him, to
accept his word and cling to his judgement,
because he is manb mm al-imam r ala 1_tumm.
That recognition of a suitably qualified faqlh as appointee
of the imm in a general sense (cal i-'umi:im) was justified
by reference to the two delegation Traditions cited above.6
It may be noted here pre-empting a later exposition 4 that the notior
of the faqih as 'general' deputy of the imam was associated with
a conferral of virtually all the executive functions of the
imam (most of them originally saqit) upon the faqlh. In order
to justify this considerable extension of clerical authority
the specificity of the delegation Traditions was denied; they
were read as being 'general' not 'specific' in their reference,
'amm not kha. Change and development in the substantive
content of furii' was achieved exegetically.
5. For further details on the lives of Karaki and Shahid II,
see below, chap. VI, 158-160 and 165-167; also, Appendix.
L 6. Shahid II, Rawa, 9L_95.

147 . Below, chap. VII.

Shahld II stated that the people should apply to a

fully qualified muftl for judicial decision because he is
appointed by the imm tal l_ c umm. The judicial competence/
office (wilya) of such a qa
was to be established by common
report (shiyC)
Common report according to Shah!d 2 meant
either the information of a group (jam'a) whose truth-telling
prevailed as a supposition:

i' (JI

1Ll

Dy

or the witness of two just men. That witness was sufficient

even if it was not (formally given) before a 1kim (judge,
qr); in fact such witness was sufficient to establish the
judge's position for all those who heard it.

J, (i

The wilya of a

>

. zLI? O',

&J ' >

Lr'L}>'
1f
was not to be established by the report

of an individual, nor by his own claim even if supported b

qar'in, nor by written support (wa-li b'il-kha) whether
with security from forgery or with the likelihood of such.8
Judicial authority according to the evidence of
ImmI jurisprudential works is then derived from an ex ante
That delegation is relevant
delegation from the sixth imm.
to all Shi d I fuqah' who fulfill certain qualities: the
shar'i al-ift', that is Imn, cadla and knowledge of ul
al-figh. The authority in question Is diffuse and quite
independent of any formal act of legitimation or initiation.
For the period dealt with here there is only a rudimentary
notion of hierarchy. A preference is specified for the more
knowledgeable but this would appear to refer to the more
knowledgeable within a particular local community; the idea
of a central most knowledgeable faqlh had not yet been formed.
The development of a central marjac al-taqld, though logical
enough, would appear to be a modern phenomenon, dating from
perha p s the Qajar pjQ9 Even now it is by no means an
L 8. Shahrd II, Raw.a, 95.
49. Cf. Hairi, ShiCism and Constitutionallsm,
Lambton, A Reconsideration.

63;

see also

85

established institution. In any case the matter of knowledge was not susceptible to tests, it was the opinion of
the community that counted. The establishment of a qi
legitimately in office (wi1ya) was an informal process
depending on public opinion and public practice. Shahld 2's
stress that the witness of two just men did not mean that
formal witness which required validation before a judge confirms that the emergence of a new judge was something quite
independent of existing judges. There was formally at least
no higher authority which might impinge in any way on the
establishment of judicial competence. A mujtahid possessed
of Imn, cad1a and the necessary knowledge gained the wilya
of a judge as a result of shiy'. The place of common report
(shiy') as well as the element of choice available to litigants gave the process a certain "populist" quality. It is
probable however that the opinion of the clerical classes
themselves would tend to prevail over those of the masses.
In general the manner and the success of conversion of legal
theory Into praxis remains difficult to assess. Islamic law,
as preserved in individual works, is the expression of an
ideal more or less distant from 'reality'. The aspiration
to implement that ideal was affected by political developments and had, I have suggested, an effect on its substantive expression; that, however, did not effect its fundamental
character as ideal. Something more about the nature of the
ideal and the compromises it envisaged may be inferred from
observation of the relationship depicted in legal works between
the ideal judiciary and the de facto power.

It has already been suggested that the doctrine of

the Ghayba necessarily involved the Imms in a general resection of de facto governments. That conclusion was also
reflected in the ImmT concept of judicial authority: a
judge derived his legitimacy from the delegation of JaCfar alSdiq; the de facto power was at best superfluous and insofar

86

as it claimed to 'appoint' judges inevitably a usurper. It

cannot however be overlooked that there exist Shi c i texts
which would appear to confer some degree of authority on
particular kings, or even on kings in general. The Kitb
al-Naqc, cAllmals eulogies to Oljaytu, and some of the
writings of MajlisI 50 spring to mind. These might be
accounted for in a variety of ways, as taqiyya, as courtly
etiquette, possibly as evidence for the existence within
Shi'ism of more than one theory of authority. 5 ' Taqiyya
may be used too often as an excuse for disregarding or reinterpreting ShI'I views. It seems not unreasonable to suppose
that a particular de facto governor may be recognised relative
to those before or after him as just but relative to the imm
might still nonetheless have to be defined as
lim. This can
produce inconsistencies but is not I think a mark of hypocrisy:
it is a matter of register. Exordiums or polemical works
( c Allma t sMinhj aJ-Kar.ma or the K. al-Naqd) may contain a
great deal of praise for kings who are tolerant or have
adopted Shl c ism but In the more rarefied air of defining the
shari c a the invalidity of de facto rulers is inescapable.52
Distinctions between church and state, spiritual and
temporal have been found to have limited relevance to Islamic
contexts. 53 A distinction between sacred and profane may be
more apt; the former is to be understood as applicable to
acts and institutions which are thought to conform to the
sharl c a, the latter to those which fail to do so.
The
essentially profane nature of the actual governing institution
was already implicit in the delegation

Tradition from 'Umar

50. For ajlisT, see e.g. Majlisl, 'Ayn al-}ayt, L 87 ff,

51. Cf. Chardin, Voyages, V, 208-212.
52. Note N'InI's attitude to de facto powers in Hairi, ShI'ism,
103. Kohlberg is Inclined to see the K. al-Maq. as
exhibiting tagiyya, Taqiyya, 140l_!402.
53. Cf. e.g. Lambton, Quis Custodiet, 126.
54. Cf. above, 20-21, 28.

87

ibn Hanzala. If an ImmI had recourse for judicial purposes

to the sultan or qIs he was dealing with
he received
only sulit. Scholarly exegesis of that Tradition proposed
the existence of a class of qIs (sic) who did in fact
possess valid judicial competence. There were then real
(sacred) qTs and unreal (profane) qTs. The Tradition
Indicated further that the judgements of the profane QIs,
- those who derived their 'authority' from appointment by a
de facto power - were invalid, even if they were technically
correct (wa-in kna 1agguhu thbitan). All the legal
theoreticians in rendering that Tradition Into furC proposed
a simple dichotomy: there were either
limTn or suitably
qualified ShICr fugah'. The existence of a just government
was not envisaged. TsT in the Nihya stated that if an
Individual called another ImmI to a faglh mm ahi al-1iagg
to judge between them, and the other preferred to go to an
official of the unjust government the latter was mutaCaddin
li'l-laqg, transgressing the law. 55 ShahTh II went rather
further:
He who turns from [a faglh possessing
shar'i al-ift'] to the gut al-jawr is
fasig, for such an action is according
to us a kabrra.6
And Shahid II cited as evidence for that the maqbl Tradition
of Umar ibn Harizala. It would appear to be the case that
even relatively acceptable and in other contexts praiseworthy sultans (Oljaytu, Tahmsp) in this shar'i context fall
under the heading of iimrn. The sacred community thus
insisted on its autonomy: there was no legitimate authority
external to that which emerged within the community. The
sacred community was not to be confused with the actual polity.
The Sunnis tended, certainly from the time of
3qillnI but probably from as early as the beginning of the

55. si,

Nihya, 302.
56. Shahld II, Rawla, 95.

-___

88

tenth century, to recognise in the de facto government a

sufficient if imperfect implementation of the shara. 57 To
justify their support of the government they suggested that
not to recognise its legitimacy would entail the illegitimacy
of all judicial activity. 8 For them the validity of a
judge's competence depended on the fact of his appointment:
all authority was delegated. 59 While expressing a nostalgic
sense of what the shard
judge should be they were willing
to bend the shari c a in such a way as to provide legitimacy
for the actual, and Inevitably Imperfect, practice of the
de facto powers. A judge should be both knowledgeable and
C dj1. However if he lacked these qualities but was appointed
by the governor (suln lahu shawka etc.) he remained a
shard judge; his judicial competence was valid. This was
true of ShfI'I as well as Hanafi and Mlikl writers. Ghazll
freely conceded that "tout II nomm par un homme puissant
peut rendre valablement des decisions, pour cause de ncssit."6
Justification for the bending of the shara was found in the
concepts of carra and mala1a/maslih al-ns; the aim was
to demonstrate the shar'I nature of the actual political
entity.
The IrnmIs in contrast to the Sunnis developed a
theory which denied any significance to the de facto appointment
and placed exclusive stress on qualities. They insisted on
Tmn, cad1a, and shar'i al-ift' as the essential prerequisites for judicial competence. Conversely, loss of
these qualities, for whatever reason, became in the absence
of the imm the only cause of dismissal. NuIaqqiq expressed
his conviction that voluntary dismissal, resignation, was
57. See chaps. I and II above.
58. Cf. Watt, Reflections, 21-23.
59. Cf. Tyan, Histoire, I, 135_1 14 24; Lambton, Quis Custodiet, I,
125-126, 132-135.
60. CIt., Tyan, Histoire, 258; cf. also 24l-212, 21114_2145.
See also Nawawl/Shirbini, IV, 37.

89

not permissible "because the judge's office (wilya) is

established by divine law (sharcan) and may not be brought
to an end by whim (tashahhiyan)." 61 There were however
certain eventualities which would lead to dismissal even
if the irnm himself did not witness that dismissal. These
were fisq (loss of t adla) and junn (loss of knowledge).
C Al1ma extended that category by adding ig1m' or nisyn.G2
There was further a possibility of dismissal for a subdelegated judicial authority if his delegator for whatever
reason lost his authority; but sub-delegation would appear
to be a phenomenon relevant primarily to the time of the
imm's presence. 6 During the Ghayba there was no conferral
of authority from a determinate superior; a faqlh emerged
into judicial competence as a result of general or even
popular recognition that he possessed the appropriate
qualities.
61. Mulaqqiq, Sharit, IV, 70-71.
62. Ibid.; 'Allma, TalirTr, 180-181; Qaw'id, 223_22!L.
63. Tsi, Mabs, K. al-Qa', 22-23; Muliaqqiq, Shar'i, IV,
69-71; 'All.ma, Ta}irTr, 180-181. It is to be noted that
not only do these discussions of sub-delegation contain
references to the imm, but that Shahid I and ShahTd II
in the Lumca and the Rawa fail to discuss the problem
of sub-delegation at all. That reflects its irrelevance,
as a shard phenomenon, to the period of the Ghayba. Nonetheless biographical works suggest that Murtadj and Ts
appointed deputies to undertake judicial functions. (See
e.g. KhwnsrI, Rawdt, 201: Sa1r al-Daylami appointed
by Murta; ibid., 355: Ibn al-Barraj,deputy to TsT.)
These appointments may be interpreted as "official" and
profane; i.e. that r!urtad. and T 1 were acting as agents
of the Buyid government. More likely however they were
informal "apointments' T by recognised leaders of the
community whose influence was widely accepted: it remains
technically true that the validity of judicial competence
was dependent not on the appointment but on the possession
of shar'il al-ift'.

90

If

ShI'I

fuqah' took up judicial posts with

de facto governments this was a concession to the exigencies

of political life not a recognition of legitimacy. The
hierarchic structure of de facto governments, even that which
marked the religious bureaucracy of Safavid Iran (Chardin,
Savory) was not condoned by, hardly even mentioned in, Safavid
works of figh. The tension between, on the one hand,
legitimate judicial authority arising within the ShI'I community
and independent of external sanction and, on the other hand,
illegitimate but pervasive political power, must have presented
to the community and to individuals some Intellectual and
moral dilemmas. Resolution of these dilemmas was sought in
the partial integration of the legitimate system with the
illegitimate. A fully qualified faqlh might take up office as
a ____ on behalf of a governor/sultan/shah in which case his
acts would be valid, not because of his appointment by the
governor but because of his qualifications for the ex ante
delegation. The complex of notions relevant to the integration
of the legitimate and illegitimate systems was made fully
available in TsT's Nihya. Though the formulation was
thereafter streamlined the content remained much the same
up to Safavid times. (It is to be noted that all the major
works of fiqh discuss the principles of governmental service
twice; once in the k. al-amr w'al-nahy with specific reference to gad' and tiudd and again in the k. al_maksib/buyC
with reference to the general principles of taking up any
official posts. The analysis which follows draws on both
sections and must be understood to go beyond the strictly
judicial sphere.)
In the k. al-maksib Ts contrasts the just sultan
dil who orders
with the j'ir. The just sultan is he who is
the ma C rtif and forbids the munkar who puts things in their
places etc. That this is the Imm seems to me inevitable
though the context in isolation may (deliberately?) permit
ater discussion of the same
of other 1nterpretations. 6L
6. For the identification of mir bPl-marf and nhT an
al-munkar as imm, see e.g. TSI , Tibyn, ad Q9.122.

91

problem indicated that the suln 'adil was the imam ai-ai.6
Shahld II whose remarks on the general problem of taking up
official posts betray his considerable concern and interest,
relates those remarks entirely to the su1n j'ir; he displays no interest in the sultan
di1. 66 It would appear
that for all the ImmI fugah' the su1in 'dil was the imm
and remarks about service with him were of merely formal
interest. T 1 ' assertion that service with him is permissible, desirable maybe indeed wjib 67 was generally confirmed
and prompted minimal elaboration.
Service to the su1n al-jawr became mustahabb only
under certain conditions, viz.
i. that it does not entail evil action (gabTli)
ii. that the officer sees his way to any of the
following:
a. igmat al-tiudd
b. amr wa-nahy
c. distribution of khums and zakt
d. silat al-ikhwn6
With specific reference to a judicial post T s1 notes that
accepting such a post is not jiz unless the appointee is
certain (ya c zim) that:
a. he will not transgress the wjib
b. he will be able to judge by the immr code
c. he will put things in their correct places,
e.g. ada q t, khums etc.6
Under such conditions service with a jir became mustaliabb
(general) or j'iz (gag'); the potential contradiction went
unresolved. It is to be noted that such a governor though
65. Muhaqqiq, Sharic, II, 12.
66. See especially the relevant section of Shahld II, Maslik
al-Ifhm, I, 167-168.
Nihya, 356.
67.
68. Ibid.
69. Ibid., 303.

92

referred to as j'ir would have to be favourably inclined

to ShPism to permit all this.
If the conditions were other than those mentioned
above then it was not permissible voluntarily to expose oneself to office; equally if one doubted one's capacity to
carry out the task.
However there might be compulsion.
Under compulsion, to take up office, even without hope of

69a

achieving the above ends, was permissible. There were degrees

of compulsion. If the potential harm arising out of rejection
of the compulsion was slight then it was thought better (awl)
to avoid office and accept the harm. Otherwise if there was
fear of harm to one's person, one's family, one's goods or
some of the believers (mu'minln = ShI'a) then it was permissible to accept office and try to put things in their places
to whatever degree was possible, especially as regards regulating the rights of the ImmTs (qa' 1uqq al-1khwn) and
lightening the burden of taxation laid upon them. A judge
should as far as possible use the ImmI code (shari at al-Tmn)
but might use other codes If compelled to do so. In all cases

there was no taqiyya in matters of killing (qati al-nufs)

( qa4')

or bloodshed (safk al-dim')(general). (Again the

disjunction of these two concepts went unresolved).70
Such in essence was TSI'S attitude to the de facto
government. It was a concession not a conferral of authority.
The reference Is to tyrants (j'1r) even where they permit
the full implementation of ShI
law. The aim was in general
to lighten the burden of tyranny that might fall upon the
ShIIs and to maintain legitimacy for the ShI C T community
while avoiding overt opposition to the governmental system.
Basically where It was possible to implement Shi law and
avoid sin taking up office was permissible and mustaliabb;
under compulsion, to take up office was permissible even if
it meant recourse to taqiyya and concessions to tyranny.71

69a. Ibid., 303 and 356.

70. Ibid., 302-303; 356-357.
71. Cf. also Ibn Idrls, Sar'ir, 161, 203; Mu1aqqiq, Shar?it
II, 12; ila, TalirIr, 158, 163; Shahid II, aw1a, 90, 95.

93

Not unexpectedly it is Shahld II's discussion,

rendered urgent by the contemporary situation, which finally
provided the most interesting insights. The impressive
competence manifest in the movement of his thought renders
desirable extensive quotation. He comments in the Maslik
al-Ifhm on Muliaqqiq's second (more general) reference to
the problem of taking up office:
An element of the problem has been discussed
before ..., to the effect that it is permissible
to undertake udd under the conditions there
specified. In this section [the fugah'] state
the absolute permissibility of taking up office
[In general](wa-gad alaq hun g jIwz al-tawliyya)
and the desirability (istibbb) of undertaking
that. That Is as long as there is security from
committing what is forbidden, and ability to
carry out al-amr bifl_macrif. But a necessary
corollary of that condition [security from
committing ... and ability to carry out...] is
[in fact] the necessity of taking up a post.
For it is incumbent on one who is able to under-

take al-amr bil_macrif to undertake It even If

a 1im has not appointed him.
Shahld II here is suggesting In opposition to earlier thinkers
that it might In fact be wjib to take up office - as long
as one can implement sharI rules. The question then arises
as to why earlier fuqah' have neglected to state this.
Possibly the reason for the absence of [statements
indicating] wujb is [a.] that (the officer] might
take on the appearance of being n'ib to a

lim

and [b.] the general nature of the prohibition

of working with them (limIn); if that [prohibition]
does not in fact reach the degree of outright denial,
it is nonetheless not less than a judgernent that it
[taking up office] is not wjib.

9)4

c._Wt

>

c,
72.

Taking up office in sum is not wjib. (Note the

exegesis of a prohibition as indicating not wjib.) Shahid
II has however managed to express his personal conviction
that it might well have been, while at the same time he has
provided adequate justification for the view of earlier
thinkers. The first point in that justification condenses
precisely the matter at issue in this section of the present
work. If a fully qualified fagTh takes up for example
lim and acts completely in accord
judicial office under a
with the law, he will take on the appearance (ra) of n'ib
lim. Legally however his authority derives from
to the
his being nib to the img m. The sacred and the profane thus
overlap and to the casual observer it might seem that the
profane is real. The problem is related to the ontological
value of religious law. As far as the theoreticians were
concerned a judge gained judicial competence as a result of
the ex ante delegation; that was reality. If he exercised
his competence through or within the de facto governmental
system then he might seem to be a deputy of the ruler; that
was mere appearance. Prevention of that appearance constitutes
a reason for not declaring it wjib to take up office with
lirnn.
Reference to a lim who appointed Imm fuqah' to
office and permitted the full implementation of the ImmI law
can in historical terms only imply Tahmsp. Shand II, though
resident in Ottoman territory, on the 'whole condoned the
taking up of office within the Safavid political system. He
was aware of the problem of legitimacy which might arise out
of the apparent recognition of de facto powers: he nonetheless proposed, if' he lid not push too far, an argument
suggesting that it might be wjib to accept official posts.
72. Shahld II, Maslik, 167-168.

95

He confirmed the traditional view that it was j'iz to do

so and perhaps mustaliabb. All of this indicates a certain
degree of support for the new dynasty.
On the question of taking up office when it was not
possible fully to order the macrif etc or when office might
entail transgression of the law, Shand II agreed that this
was permissible only under the usual conditions, viz.,
compulsion arising from fear for self, family, goods or
other mu'minTh. Some new distinctions and clarifications
were however introduced. For example Shahrd II distinguished
a particular situation wherein orders to carry out what is
Iiarm may not be associated with the actual work of office,
as, for example, if a
lim compelled a person to accept
forbidden goods (as payment); the execution of his official
function might still be left to his own discretion. In such
a case accepting the office remained jiz, mustaabb or even
wjib and did not wait on compulsion. Shahid II further noted
and commented on a point small enough in itself but not
previously mentioned, namely that permission to act under
compul,sion included both the actual perpetration of the act
and causation (tasbib), as when an individual gives a fatw
or a command: it is permissible under certain conditions to
give a fatw which is not fully in accord with the law.
t1

})i

Lt
73

The general tendency of Shahld II's effort is to

argue in favour of taking up office. He commented also on the
disjunction between bloodshed (dirn') and death (qati) as
constituting the limits of taqiyya.
Dim' might include
both killing and wounding, jarI and qati.
Shahld II blamed
TsI for introducing that termnology which he thought wrong.

Qati is the correct condition and the only limit to taqiyya.

Thus, under adverse conditions, having accepted office, an
official might act under taqiyya in discord with the law: the
only exceptionto this were acts which resulted in deaths;
bloodshed of a lesser kind might be permissible.
73. Ibid.

96

Shahid II was in general concerned to sanction and

encourage the taking up of official posts, and to clarify and
broaden the scope of tagiyya. The practical imperatives which
must have conditioned his discussion were partly those relating
to the establishment and support of the Safavid state: there
lim who permitted and encouraged the
indeed there was a
amr bi'l-ma'riif etc. but there too practical politics might
still require the perpetration of illegal acts - sanctioned
and condoned by the doctrine of tagiyya/cornpulsion. However
that doctrine was not less relevant to the difficult situation
which Shahld II faced in Ottoman Turkey. He himself took up
an official post as a teacher in a madrasa at Batalbak,
deliberately indeed it Is said searched out that office by
travelling to Istanbul and applying to senior officials there.
Biographers claim that he taught the five C!) schools of law,
though whether the teaching of the fifth, the Shr d l, could
be officially undertaken at that period in Ottoman Syria may
well be doubted. His post nonetheless gives a peculiar
relevance to his assertion that where necessary ift' could
be undertaken under jaqIyya: that was clearly a personal as
well as a general and political issue.
A full discussion of taqiyya would require a wider
reference than the present one. It is however the case that,
in works of figh, taqiyya becomes a significant issue primarily
limIn. The issue arises
in the context of working with
out of the duality of the ShI'i world view. On the one hand
they perceive the Shidi community as autonomous with its own
legitimate magistrature. This is contrasted with the illegal
political structure and illegal magistrature which derives
Its "validity" from the ruling monarch (lim, j'ir). Where
the two structures can be made to overlap it is mustal)abb
for the legitimately qualified to exercise his function
through as it were the illegal system. Though he might
lim this is not a conferral
thereby appear to be n'ib to a
of authority upon that system; that remains per se illegitimate. When the two systems, the legal and the illegal, come
into conflict, as when the illegal government requires actions

97

which are contrary to the law or prohibits actions which

are necessary, tensions are set up whIch can only be
resolved by the suffering of harm or by recourse to taqlyya.
The general ShI'i position is that if the harm is small
enough it is better simply to suffer it. But for major harm
threatening oneself or other believers it is permissible to
so regulate one's affairs as to mitigate or remove the clash
between the legal system and the illegal polity, even where
this means transgressing the law - except of course in killing!
bloodshed. Thus in works of fiqh tagiyya provides an
essential element of elasticity reducing the tension between
the ideal legitimate Shi d i community and the illegal political
structure, the sacred and the profane.

There must exist in the State, as a legal

association, a power of final legal adjustment
of all legal issues which arise in its ambit.
The legal association will not be a single unit,
and law will not be a unity, unless there is
somewhere one authority to which crucial differences ultimately come, and which gives, as the
authority of last resort, the ultimate and final
decision. Different social groups may press
different views of what is, or ought to be the
law; it is even possible that different departments of the State may hold, and seek to enforce,
different notions of what is legally right;
there must be a final adjustment centre. That...
is the sovereign.75
Barker's remarks on sovereignty serve by contrast to emphasise
what Islam was not. There was there no State as legal
association, and no power of final legal adjustment. There
7. See Kohlberg, Taqiyya, for a general assessment of taqiyya.
75. Barker, Principles, 59-60.

98

were rulers whose power rested usually on force; and there

were different social religious and political groups with
different views of what was or ought to have been the
relevant norms. There was no this-worldly sovereign. Even
the clerical class who depicted the sacred umrna as a minutely
detailed legal association failed to provide - and did not
necessarily desire to proide 6 - the sine qua non of practical
Implementation - a final court of appeal. The systematic
works of Islamic jurists were private works which might as
being convenient or excellent become in practice for a 'cime
or for a particular group authoritative: there was however
no provision for a final or definitive statement of the law.
Every renewed appeal from established authority back to the
sources was a denial that the law could be defined. In
practice later governments especially the Ottoman so controlled
and ranked the religious classes that for those aspects of
the religious law that were actually and officially implemented
a final arbiter seemed to exist. The fact however that a
government chose to accept as definitive the view of for
example a particular Grand Mufti did not make that view a
statement of what the law was. From the point of view of the
scholars the law was not so easy to pin down. The Sunnis in
general did at least recognise the legitimacy of the government and the validity of the fatws of rival schools and mufts.
For the ShT c a the de facto hierarchical system even in Safavid
and

ShII

Iran was simply Illegitimate. The individuals within

76. Cf. Coulsori's remarks on the nature of Islamic law,

Doctrine and Practice, 219; see also Becker's words,
cited by Wansbrough, On Re-composing, l63-l6. The sharT'a
as pious Ideal was affected certainly by events, and
was, by Sunnls In particular, so moulded as to provide
some kind of sanction for what actually happened.
Compromise however subsisted with a nostalgic sense of
the ideal, and, frequently, with an underlying acknowledgement that the ideal was not only in practice unattainable
but also in theory not even susceptible to definition.

99

it might or might not be fully qualified and validly

competent judges, but the element of hierarchy found no
sanction in legal theory. This had certain repercussions
on the practical efficiency of the shard courts in so far
as they were permitted to exist.
Consultation was a generally recognised preliminary
to the judicial decision. 77 Tyan points out with regard to
the participants in the concilium (mashra) that "[leurs]
avis ont un caractre ssentie1lement consultatif; la
dcision appartient toujours au iI seul." Ts1 in the
Mabst states for the Immls that there must be present at
the judge's session ahi al_Cilm mm ahi al-I)aqq; for the
Sunnls (cinda al-mukhlif) one representative of each madhhab.
He asserts that according to the Sunnls no-one may reject a
judge's decision even if in his opinion the judge has decided
wrongly (bi'l-bil); for if a judge's opinion is based on
ijtihd he must act on it and none may oppose it - except
indeed if the judge's decision is opposed to nas al-kitb or
sunna or ijm' or qiys which has only one possible outcome.8
Sunnl decisions may thus be of two types: either part of an
area of undisputed certainty or part of a more doubtful area
where more than one decision is possible. In that more
doubtful area (ijtihd) a decision may be thought wrong but
cannot be challenged. The principle was enshrined in the
slogan kull mujtahid muIb - anathema to the early ImmIs.
Tsi claims for the Shi d is that a decision may either be in
accord with al-liaqg or otherwise; he denies the area of
ijtihd.
We say if the judge's decision is correct (in
as ba al-agq) his ruling is effective (nafadha
1ukmuhu) and none may gainsay it, but if he is
wrong those present [the members of the conciliurn]
must inform him of his error.
For the ShIdis there are no i.ndeterminate cases, only right
or wrong; "there is in our opinion no qiys in the sharc nor
'77. Cf. Tyan, Histoire, 315-322.
78. TsI, Mabsl, K. a1-Qad', 4-5.

100

ijtihd and not every mujtahid is mu3b."79

Shi d i judges according to Ts unlike Sunni judges
may not act by tagild nor seek fatws; however if they are
doubtful about a decision they should consult with ahi al-'ilm
that they might enlighten them about the dalil - otherwise
they should not consult.

J DI J) c

Li

;'j
ShI

CI

80
decisionsmust be either right or wrong. Perception of

prior error entailed reversal or annulment of prior decision:

If a judge considers that he himself or that
another judge has erred he must annul the
earlier decision and recommence the judging on
the basis of what he knows to be right.
For some [clearly Sunnls] if the error is in an
area of ijtihd81 then he may not annul the
earlier decision though he is bound for future
cases to follow his later ijtihd.82
Not the least beneficial aspect of the Sunni dictum that kull
mujtahld muIb was that it provided some degree of judicial
stability; a decision once formally given and established as
ijtihd could not be reversed. Shi t i judicial decisions on
the other hand were, it would appear, always subject to review
and since there was no hierarchic system, no higher court
whose decisions were binding on lower courts, there was no
obvious end to litigation. TSI specifically stated that when
a new Shi d i judge took up a position it was not incumbent to
79. Ibid., 4; see also 5.
80. Ibid., 6. Information about consultation and the nature of
the judicial decision Is repetitively and rather Incoherently
presented over pages 146 of K. al-Qa2'.
81. The text reads t 'is not" but the 1 is certainly a misprint.
82. Tsi, Mabsiiil, Qad', 6.

101

follow the decisions of the one before him - though he

might do so. If it became clear that the earlier judge
had judged a case bi t l-1aqq the new judge should confirm it,
otherwise he should annul

T'

discussed also the question of what ought to

happen if a judge told another judge of a particular decision.

For the Sunnls the situation would appear to be that if the
information is passed directly from judge to judge the
second judge will act according to the decision of the first;
likewise if two witnesses bear witness to a prior ruling it
will be put into effect by the later judge. For the Sh'Is
however:
that which our madhhab requires is that it is
not incumbent on the later

gcT

to act in accord
with the opinion of the earlier at all, for
there Is no dalil indicating that.8

Western observers accustomed to the idea of legal sovereignty

might be tempted to call this judicial chaos. It was however an almost Inevitable result of the theory that all judges
were equally legitimate, equally drawing authority from the
absent imm, equally without an evident superior. Initial
refusal by the ImmIs to follow the Sunnls in recognising
an area of Ijtihd made the matter worse.
This almost absurd degree of Independence for every
Individual judge became a permanent characteristic of the
ImmI theory. Muhaqqiq confirmed that perception of prior
error, whether perpetrated by oneself or by another judge,
whether qat'T or ijtihdl (!) required annulment of the prior
decision.' 5 By the time he wrote the Mukhtasar however he
seems to have perceived that there was a need for some stability.
[a] A judge does not decide on the basis of
Information provided by another judge, nor on
the provision of evidence (bayyina) as to the
validity of the judgement of another.
[b.) However if one judge decides between litigants
and confirms the judgement and causes [witnesses]
83. Ibid.
8. IbId., 22-23.
85. Muhaqqiq, Shar'i, IV, 75-76; cf. for Muhaqqiq's use of the
term ijtihd, below, 236 n.33.

102

to bear witness to himself [his status];

and if [subsequently] two witnesses bear
witness as to the decision in the presence
of another [judge] it is incumbent on
the latter to put the decision into effect.

7L(L

t
, ,i- >

*-

o

1

[a]

> )*

''

86.

Statement b. here is actually a contradiction of statement

a. and represents a determined effort to encourage judicial
stability. That it was incumbent on a judge to accept a
properly authenticated decision by an earlier judge would or
the whole seem to be an aberrant view within the SI-Il c i system.
The theory of ijtihd, initially rejected by the
immls, was later, primarily through the efforts of CA1lma,
adopted and restated as a central part of their legal and
social theory. Adoption of that concept by C Allrna rendered
him able to introduce it as relevant to the stability of
judicial decisions. But his argument, lengthy and subtle,
finally led him to restate the traditional view that all such
decisions might be changed by one who considered them an error.
If a judge gives a decision which is opposed
to the Book or sum-ia mutawtira or ijm, or
in effect if he opposes a dairl qat

it is

incumbent on him and on other judges to annul

the decision (wajaba ... naquhu) . . . But if
[the judge t s decision] is opposed to a dalli
annI it may not be annulled ... except where
the decision is a khata' in so far as he reached
it neither on the basis of a dalil qai'i nor
annI, or jfl so far as he did not fulfill the
conditions of ijtihd.
86. Id., Mukhtaar, 283_2811.

103

That would seem to suggest that there is for ShI'Ts an area

of ijtihd wherein if a decision is made, though one might
disagree with it, it is not to be annulled. c A11ma however
proceeded to recall that reconsideration of a problem
demanded that a mujtahid should follow his later ijtihd.
And that led to the following:
It Is not necessary for a judge to follow the
judgement of one who preceded him nor the
judgement of any other Ikim.
/ L9

L L

cuL (..ei9

There is some contradiction here. cA1lma elaborated:

If [in spite of Its not being necessary] a
judge follows earlier decisions he should
consider the earlier judge:

If he was
one of the ahi al-ijtihd he should not
annul those of his decisions which are awb;
but he should annul others - [unsolicited] if it Is a matter of Iagq 1i'1lh like citq
and alg, but if it is a liagg dam he
should annul it only on application.
Now assessment of what was awab was not easy. There was
an area of legal decision qualified as being dependent on
dalil qaI within which there was general agreement on what
was awb.
The precise limits of that area were never defined:
it probably got smaller with the passage of time. Within the
area of ijtihd (which tended to expand with time) it was a
matter of reasoned opinion as to what was awb.
A11ma's
remarks would seem to imply finally that if a judge disagreed
with another judge he might or should annul his judgement.
He went on to propose another oddly casuistical case and then
conceded that all judicial decisions were subject indefinitely
to review:
If the decision is, according to the first judge,
an error and,according to the second,correct
then there is dispute as to whether it should be

ldL

annulled, if the first judge is one of the

ahi al-ijtihd. The best view is that for
every judicial decision that appears to a
judge to be an error whether he himself
was the judge or someone prior to him, he
should annul it and recommence the Judgement
on the basis of what he knows to be liaqq.

Let (J 7

j")t J9

8
c A11ma displayed on the whole some concern that judicial
decisions within the area of ijtihd should be recognised as
valid; but he finally affirmed that in fact all fully
qualified mujtahids could at will rescind earlier decisions.
If a judge did not recognise In a prior judge one
of the ahl al-Ijtihd he should annul all of his decisions
en bloc even if they were right (awb).

Allma admitted

some doubt there arising out of the fact that legal rights
were correctly apportioned:
C:-

)5' ;

qj

) )

88
The ruling however emphasised that the major
qualification for judicial office was not appointment but
knowledge. A government appointee might act as a judge,
might indeed ensure that de facto his decisions prevailed,
might in fact implement the correct decisions, but those
decisions would be from a religious point of view null. Common
to all the ImrnI fuqah' was a ruling that if a former
litigant claimed to have been adjudged bi'l-jawr then It was
87. 'Al1ma, Qaw'Id, 225-226.
88. Ibid.

105

incumbent to re-investigate the case. 8 Bi'l-jawr probably

meant by a government court. Such decisions might always
be reversed by fully qualified religious judges. That
reversal however even if it represented the 'true' state of
affairs might not prevail over the actual. There was always
to a greater or lesser degree tension between sacred reality
and profane events.
The unwillingness of Imamrs to provide for a final
decision in judicial matters affected also their discussion
of correspondence between gIs (kitb qin il qin).

1 discussed the matter at great length, and not without

considerable confusion and repetition. 9 All the furt' he

adduced .however would seem to be relevant only to the Sunnis;

for the Shica correspondence between
was not permissible.

qIs

However all these fur' are not relevant to

us, because, as we have demonstrated, correspondence between
is not permissible.

q3rs

(ot%L&
91
'Allma too rejected outright the idea of conveying
a judicial decision by letter. But provision of bayyina (=
two witnesses) as to a judicial decision contained in a letter
altered the situation:
If the bayyina bears witness to the judgement,
and if the judge had caused [two witnesses] to
witness the decision, then the best view is
that the second judge should implement [the
decision of the first ]; because of the need for
ithbt in far lands and because of the impossibility
of transporting the original witnesses

Ts, Mabs, Qa', 6;

90. TsI, Mabs, Qad', 11-13.

89. e.g.

91. Ibid., 12.

c A1lma, Qaw'id, 226.

106

Allma claimed there to have avoided transgression of the

revealed text which prohibited only recourse to written
evidence per Se.
Transferral of a decision however only
made it possible that a judge might - and 'A1lma thought he
should -confirm an earlier decision. It was not incumbent
to do so.
If the two witnesses bear witness before the
second judge he should put into effect what
was decided by the first, not because he
[the second judge] has adjudged the validity
of [the earlier decision] rather the usefulness
lies in the ending of litigation.
/

92
The ending of litigation was in fact a central
problem within the ImmI judicial system. Muhaqqiq and 'A11ma
showed some awareness of the problem and some concern to
obviate it; they recommended in general an acceptance of
earlier decisions (if properly carried out) but they did
not in the end limit the unlimited competence of every
qualified individual to review earlier cases or to express
a new opinion. Presumably it was only the temporally later
decision which was for the moment valid.
Such a system it is clear could not be practically
efficient. The comments of Barker cited above show that for
him the absence of a final legal authority simply means that
a legal system cannot work. Such evidence as is available
from other sources suggests that in practice the ImmI
religious courts did not work. Chardin noted that most
Persians preferred government courts to the gst courts,
and noted, concerning the latter, that their cases could rarely
be brought to a final decision. "La facult de plaider est la
92. CA11ma, QawCid, 230.

107

plus grande du monde en Perse, et de plalder sans fin, soit

au mme tribunal, soit devant les autres, et . plus d'une
tour." 93 Later he states that court cases in
douzaine tour
Persia though easy are not more common than in other lands,
"parce gue les procs vont a la prompte ruine des plaideurs,
cause de ce gu'il faut donner pour gagner les juges,
tant
gue parce gu'on n'est pas stir aprs gue les procs sont finis
S.

S.

gu'on ne soit pas, des le lendemain, tire en cause a un autre

tribunal pour les mnies affaires
The failure of the courts to work in practice
reflects not so much a deficiency in the law as simply one of
its characteristics. In spite of aberrant attempts to so
constrain or depict Islamic law as to make of It a blue-print
for practical politics, it constantly re-emerged as In fact and not less grandly - the developing product of pious and
wondering speculation. Its ambiguities, its recalcitrance to
definition, its generous ikhtIlf and open-ended problems were
on the one hand essential to its flexibility and diachronlc
development, but reflected on the other hand its transcendent
and Intangible divine origins.

93. Chardln, Voyages, VI,

94. Ibid., 95-96.

91.

108

CHAPTER V
C0UNITY FINANCES
MwardI said of zakt that it was paid 'ahratan iiah1ih manatan li-ahi al-sahmn. as a purification for the
donor and a support for the recipient. 1 It has then a dual
aspect. As a 'political' tax it provides for the transfer of
wealth from certain productive classes of society to certain
poor or non-productive classes. As a religious duty it is of
essentially the same type as a1t, iiajj etc., a farIca tal 1t ayn.
Like these it is a ritual whose correct performance
provides religious reward (jaz'); and like these correct
performance involves an attention to precise details of quantity
(nab etc) timing (a1-1aw1) and intention (niyya) which may
be irrelevant or even inimical to the optimum fulfillment of
the political aim.
The zakt donors were defined as those possessed of
productive wealth derived from pastoralism, agriculture or
trade., 2 The recipients were divided into eight categories,
five of which may be subsumed under the headings of the poor
the needy or those under particular difficulties; the other
three were the tax-collecting and military classes (al-'mi1n,
fi sabil allah, al-mu t allafa).
The Sunnis in general
acknowledged that the de facto power had a certain right to
collect and distribute the zakt, "because they (the wu1t) are
the trustees (umana') over the collection of zakt from the
donors on behalf of the recipients." 3 Shafidl however
perceived that the political rights of the wulat might conflict with the individual need for religious reward. For the
latter it was necessary that the zakat be paid at a particular
1. Mawardl, Ahkam, 195; cf. also, Aghnides, Finance, 323, n.j.
2. See, for further details, mostly common to SunnTs and Shidis,
Aghnides, Finance, 203-295. For those who possessed sufficient
wealth of the specified types, zakat was wajib. There were
categories of wealth on which zakt was mustaliabb.
3. Shfi t i , Umm, I, 70.

109

time (al-Iiawl): Shfi'i indicated that if the tax-collector

did not arrive at the appropriate moment the donor should
distribute zakt himself.
Likewise with regard to niyya
ShfidI considered it preferable (wa-aIabbu ilayya) that the
individual should undertake his own distribution in order to
be sure about the discharge of his religious duty (fa-yakn
C al yagin mm adtih).5 There was thus some tension between
the political rights of the ruler and the religious needs of
the individual.
The ruler was entitled to use one part of the zakt
collected - that of the mi1n - to finance his tax-collecting
bureaucracy.
&.i'J

(y't ukvi

It may be given to the aides in the administration

of the
1.-adaga.6

wlI

There was however for ShfiI a possibility, when

the donor distributed the zakt directly, that the share of
the Cmj1 might be sqil. He further considered that the
share of the mu'allafa might be sgi. 7 On the whole, for the
Sunrils, the system of religious taxation implied the existence
of a government who collected and distributed. Of what was
collected the government might legitimately use 3 out of 8
parts for bureaucratic and military purposes. Shfi', who
on the whole puts forward an idealistic view of the law,
stressed that the individual need for salvation might pre-empt
the right of the ruler to collect and distribute. He also
thought that two out of the three parts dedicated to the
government might under certain conditions lapse. He set out
the rules as ideals without any reference to the actual
practice of the administration. Later writers manifested a
rather different concern to provide the greatest degree of
4. Ibid., 15.
5. Ibid., 19.
6. Ibid., 61.
7. Ibid., 68; cf. Ab Ysuf, Kharj, 81.


110

letitimacy for the actual practice of de facto powers.8

defining the ImmI attitude to zakt, declared
that in the absence of the imm or his appointee (man nasabahu
al-imm) the shares of the suCt/milin, the rnhfallafa and
jihd were sqi1. 9 The term man naabahu al-imm refers
presumably to a contemporary of the imm, certainly in this
context to an entity deemed absent. During the Ghayba,
then, the ImmIs emphasised the dissociation of the sacred
from the profane by denying precisely those shares which
would have served the administrative and military needs
of the government. They were enabled thereby to affirm
their independence of the political institution and to stress
the religious rather than the political aspects of zakt.
Later, in the Mabs, Ts confirmed that the shares of the
mu'allafa and the

mi1 were

sgi,

but, by providing an

extended definition of sabil allah, he was enabled to affirm

its contemporary operancy) Sabil allah was there defined
as including the building of mosques, the upkeep of bridges
and other mali1i) 1 The scholarly class thus had a claim
on zakt only in so far as religious buildings were concerned. Muhaqqiq confirmed that the shares of the mil
and the mu'allafa were sqi, accepted the extended definition
of sabil allah and added to it the consideration that jihad
- 12
might be possible even in the absence of the imam.
Allama,
a generation later, not only accepted the possibility but
inferrred from it that the share of the mu'allafa too
could not continue saqi:

)I,P

r
8.

13

Below, 117-120

9. TsT, Nihaya, 185.

Mabs, k. al-zakat, 17, 18.
10.
11. Possibly an indication of Sunn influence on ShI'I fur;
- cf. Abu Yisuf,Kharaj, 81.
12. Muhaqqiq, Shara'i', I, 162. Tusi also accepted the validity
of defensive jihd, but this did not affect his discussion
of zakat.
13. cA11ma, TaIirr, 68.

111

The share of the Cmj1 however would appear to have been

left as TsI had defined it:

L&' (.LifU
l4 ' )' -r'
The share of the mu'allafa, the

'

s'3, and

the ghzI are sqi during the Ghayba except when there is need of jihd.
It is difficult to see how the exceptive phrase can have
that share remains sqi.
any relevance to the

s'P;

There is nonetheless an element of ambiguity about the

phraseology which may have been deliberate. The ambiguity
was emphasised, possibly exploited, by Shahld 1.
jL.

15
During the Ghayba, no s'i and no
mu'allafa except for one/those needed.
That removed the specificity of the need (Jihd) and left
it an open question as to whether there might not be a
need for the Sad during the Ghayba. Shahld II was quite
explicit. Commenting on the exceptive clause, illa li-man
yu1itj ilayhi, he states:
That is the faqh if he is able to appoint
a s'T to collect zakt; and if Jihd is
necessary during the Ghayba and ta'lTf
is required that is permissible by means
of the faglh or someone else (bi'.l-faqih
wa-ghayrihi) 16
By the sixteenth century the three zakt-recipient categories
which by T1 had been declared sqit had been re-established
as operative. That change took place exegetically and
14. Id., Qawid, 22.
15. Shahld, I, in Rawa, 51.
16. Shahld II, Rawda, 51.

112

without polemic. There was no overt appeal, as justification

for development, to revelation or to ijm; there was rather
a gradual reassessment of the meaning of revelation, or of
the dicta of earlier jurists. Adduction of the concept of
1ja as justificatory might be compared with Sunni recourse
to artira. 17 The final Imml assessment of the eight categories paralleled the Sunni assessment except in this that
whereas for the Sunnis the major executive agent (for
organisatiori of collection and for jihd) was the wl or
governor, for the ShI'Is it was the fagh. Legal developments
in this field benefited the fugah' in at least one other way.
Although there had been amongst the ImmIs various views as
to who was included in the categories of the fugar' and
the maskln the opinion prevailed that it meant a person
whose income was Insufficient to cover annual expenses for
himself and hisfamIly. Shahld II added to this the consideration that If someone was incapable of earning a living
because of his endeavours in religious knowledge ( C ilm diPi)
he might receive zakt, even if he could, by leaving his
studies, earn a sufficiency. In the section on zakt al-fira
he added that special attention was to be given to people of
pre-eminence in cilm and zuhd (ahl al-fal bi'l-'Ilrn wa'l-zuhd):
preference was to be given them over other degrees. The fugah'
thus not only finally controlled the collection and distribution
of zakt but were also major recipients.18
Developments in legal theory reflect change in
the aspirations of the religious class. Those in turn were
affected by historical events and by ideological (legal)
conditioning. The early ImmI Image of the sacred umma, as
presented In the Nihya, is relatively free from concepts of
military or bureaucratic organisation; Shahid II on the
other hand had perceived the community as ideally possessed
of a tax-collecting bureaucracy and a military force, both
17. Above, 28, n.18; 87-88.
18. Shahld II, Rawa, 9, 51.

113

under the control of the fuqah'. It is certain that in

Buyid times the fugah already considered themselves as
judicial authorities, and no doubt were able within the
Shi c i community to control some part of zakt payments.19
In the following centuries, especially under tolerant and
favourably inclined governments (the early Mongols, Oljaytu)
it is probable that they consolidated their hold on their
followers and increased their control of material wealth.
That was reflected in the increasing stress in legal works
on their general authority. By the early sixteenth century
when Tahmsp promoted the cration of . frna1ly Sh

state,
their espirations and perceptIons had been sharpened to a
point where they preser.ted themselves as the rightful
authorities within such a state. The Safavid theory differed
significantly from the early expression In so far as it
offered a determinate body which ought ideally to replace or
at least control the actual government, in order to implement
the sharica.
The early doctrine of sug
had probably
developed as a means to avoid overt expression of political
opposition.

According to Shfi

God had imposed on all

Muslims a charge on their wealth on behalf of other Muslims.20

Zakt in general was not to be paid to non-Muslims. 21 Just
19.
20.
21.

Cf. QazwTnl, Nag., 1614.

Shfi C

I,

Umm, I, 70.
For both Sunnls and ShI'Is there was much discussion
about whether the mu'allafa might or might not be nonMuslim. See, e.g. Shfidi, Umm, 61 and 72-73; the
different opinions there expressed indicate the composite
nature of the Umm. See also TsI, Mabs, Zakt, 17;
note that Shafici is cited as source of this discussion:
his opinion thereafter became part of ShI'I ikhti1f.

11

as performance of alt was an outward symbol of adherence

to the Islamic community so was payment of or acceptance
of zakt.
Participation in that ritual displayed membership of the community. The ImrnIs were concerned not with
islam as the relevant category but with Imn; that was
specified by all writers as a necessary condition in the
recipient of zakt al-ml. TsI suggested that In the
absence of ImmI recipients zakt al-fira might be paid to
mustac'ifIn in general. Ibn Idris objected to this view
and Mu1aqqiq after initial hesitation confirmed Ibn Idris's
opinion. Thereafter no ImmI faqlh appears to have denied
that zakt al-ml and zakt al-fira were to be paid within
dr ai-Imri. Any ImmI who knowingly paid zakt wrongly that is to a non-Immi - failed to gain reward. Zakt was
a community defining ritual.22
C Ad1a was a quality which might be required in
the recipient of zakt.
It was a quality Involving manifestation of positive virtues and was the opposite of flsg,
a quality acquired by the commission of one or more kab'ir,
or persistent commission of sagh'ir. To stipulate in the
recipient of zakt the quality cadla suggests that the
community within which that ritual was carried out was perceived as a positively virtuous community in which every
individual assiduously obeyed the law. To stipulate absence
of flsq permitted a broader definition of community, including
those who were if not conspicuously
dil yet not fsiq. The
matter became controversial at about the time of 'Allma.
He stated In the TalirIr al-A1ikm:
T 1 and Nurtad have specified Cadla
but others have rejected this opinion;
the latter view is stronger. The others
have specified avoidance of kab'ir.
In
our opinion it Is permissible to give to a
fsig as long as he is mu'min.23

22.

TSI, NIhya, 185, 192; PIabsu, Zakt, 16;

23.

Ibn Idris, Sar'ir, 106, 109; MuIj.aqqiq, Iukhtasar, 59.

c A11ma, TaIrIr, 69.

115
The identification of "the others was provided in the
Mukhtalif al_Shi c a. In that work CAllma cited
Ab
l-Salah, Ibn al-Barrj and others as specifying some variant
of the root 'd 1. Mufld and Murtad both demanded positive
virtues:
iyna, riazha, taqi, C aflf
According to CAl1ma
Ibn Idrls did not specify Cad1a. But that was misreading
or misrepresentation for Ibn Idrls had in fact demanded
ne adla aw }2ukmuhu".
The only writers that cAllma finally
adduced in favour of his opinion were CAll ibn Bbya and
his son al-adtiq, of the generation prior to Tsi. Even
there the evidence was negative: They did not specify cadla,
but nor did they specify the absence of this quality. In
the face of an immense weight of tradition cA11ma chose to
deny that C adla was a condition required in the recipient
of zakt. Avoidance of kab'ir was necessary but not absence
of fisq.25
The point was obviously thought important, and the
law was obviously capable of modification. That modification
was achieved again not by appeal to revelation or 1jm
rather by asserting (unconvincingly?) the absense of ijmc
and appeal to the dicta of earlier jurists.
Allma for
reasons which are not Immediately apparent wished to deny
that cadla was a condition for receiving zakt. To that
end he searched the sources for justification. It Is manifest
that his opinion was not forced upon him by the overwhelming
evidence of acceptable ul.
Shahid II also treated the question of cadla as
a contentious issue. In the Maslik al-Ifhm he put forward a novel argument. All Imml fuqah' accepted that
children of mu'minln may be given zakt to the exclusion of
children of kuffr. Shahid II noted that children may not
possess the qualities either of adl or fisg.
If they do
not possess Cadi they should not receive zakt. But everyone
agrees they should receive zakt. Therefore Cadl cannot be a
condItion.26 The logic is fair enough but the reader is not
2LL Ibn Idrls, Sar'ir, 106.
25. A1lma, Mukhtalif, II, 11.
26. Shahld II, Masalik al-Ifh.m, I, 61.

116

convinced that he was led to his opinion by the logic,

rather his opinion required justification.
Shahid II's conimentary on the LumCa apropos of
c adla may be given in abbreviated form as illustrative of a
manner of juridical argument.
[Shahid I said:] c Adla is to be considered
in other than the mu'allafa ... arid it is
said (qua) that that which is to be
considered is [only] avoidance of kab'ir.
[Shahld II added:]
As to the mu'allafa it is because their kufr
prevents cadla
Consideration of Cadla in the Cmi1 is
generally agreed (mawcic a1-wifq).
As to the consideration of Cad1a in the
others, that is only one of various views,
though Murtada claimed ijm'
Some stipulate [only] avoidance of kabir
even if the saghir lead to the designation
fisq. That is because of the text (na)
which denies (zakt) to shrib al-khamr ...;
it does riot indicate a general denial to
fsiqs.
Other kab'ir have been added [to
wine-drinking] by a process of muswt.
That however is a matter for consideration
as muswt is denied and qiys rejected [by
the Imamus]
If Cadla were to be considered children
would be denied zakt 27
Shahld II like cA1lma wanted only kab'ir to be taken into
He argued that
consideration as denying the right to zakt.
ijm, though it had been claimed, had never existed on this
issue; he questioned the interpretation of a revealed text
by earlier fuqah'; and he worried aboul children who might
have to be denied zakt.
27. Shahld II, awc.a, 50.

Commendable care for the welfare

117

of children, and for the precise meaning and exegesis of

the sources may have been the only factors which led him
to his conclusion; the argument however may also and
perhaps better be understood as post facto justification
of an opinion which emerged for other reasons.
While the precise motivation behind CA11ma?s
and Shahld II's argumentation cannot be discerned it may
nonetheless be remarked that removal of Cad1a as a condition
broadened the scope of patronage of those who distributed
zakt - the fuqah'.
Further, again, from Tsr to Shahid II
there was a change in the underlying vision of the community.
writings implied a community of believers, (mu'minrn)
T 1'
whose actions were in accord with their faith (cad1a), a
community without a formal bureaucratic or military class.
Shahid II's believers were acknowledged as participants
in the community even if their actions did not reach the
He stressed the general authority
high standards of Cad1a.
of the fugah' over the bureaucratic and military classes.
That view did not reflect any legitimacy on the Safavid
state but it did rather suggest that most of the people
within that state might be brought within the ambit of
practical shard politics; or at least, in this context,
they would be subject to those elements of control and
patronage implied in the theory of zakt. The fuqah' might
hope within the community to exercise at least a part of the
authority they conceived themselves to possess, and to make
this a basis for consolidation and expansion of their
practical control of affairs.

For both Sunnis and ShI'is niyya (intention)28

is an essential element In the attainment of religious reward
(jaz' /ijz'). CA1lma stated that if a man gave away all
28. For niyya, see Schacht, Introduction, 116-118.

118

his wealth and forgot niyya he gained no ijz'. 29 Shfil

explained that since adaqa might be either incumbent (farc)
or voluntary (taawwae) it is not permissible that a man gain
ijz' for distribution of zakt (the far) except he express
niyya to the effect that it is such. 3 Niyya distinguished
the formal ritual of zakt from superorogatory alms-giving.
There was at least one modification to the requirement of
niyya, namely that if the will took zakt from a person without
his consent that person still gained ijz'. ShfiCl was led
from there to conclude:
If the wall takes from a man his zakt, whether
the man at that point does or does not express
niyya, whether the man is willing or otherwise,
whether the will expresses niyya or not, it
provides ijz
for the donor; just as 1Jz'
is procured in the matter of distrbution of
zakt whether the one who distributes it is
his [the donor's] wall [= agent] or the sultan.
Shfidl however does not in his fur' provide for an unjust
sultan; the implication is that the zakt will reach Its
legitimate recipients. He added that it was preferable that
the individual should undertake his own distribution so as
to be sure about the correct discharge of the duty of zakt.31
Mwardl stated that payment of zakt (on hir
goods 32 ) to the w1l might be either wjIb or mustaliabb; It
was a matter of ikhtilf. 33 That was a development from
Shfi c Fwho had considered it preferable for an individual
to distribute zakt himself. Mwardl acknowledged that if
the ultimate distribution was unjust (not in accord with the
law) the individual should not in fact pay his zakt to the
29.
30.
31.
32.

'A1lma, Talirlr, 67.

ShfI'l, timm, I, 18-19.
Ibid., 19.
For zh1r and btin goods, see Aghnides, Finance, 296-297.

33. Mwardl, Alikm, 195.

119

c mi1 but should hide it, or, if it was taken willy

filly,

should repeat the zakt in order to gain reward.

Mawawl
(d. 676/1277-8) in the Minhj al-libTn admitted that

while an individual might distribute zak g t himself or

through his wakfi it was better for him to give the zakt
to the imm (i.e. de facto ruler) unless he was j'ir.35
The commentary on this by Shirblnl, (d. 977/1569-70)
though complex, is illuminating on the tendency of Sunnl
legal theory to acknowledge the authority even of tyrants

(j'irTn).
hir goods [may be distributed by individuals]
according to the new view, [which is derived] by
giys from [the situation of] bin goods [which
may at all times be distributed by individuals];
the old view is that it is incumbent to deliver
[zakt on
hir goods] to the imm or his n'ib
because of ... (Q9. 103) ... The apparent view
[arising from the Quranic verse] is that it is
incumbent. This [is the case even] if he [the
1mm] does not demand it. If he does demand it
it is incumbent to surrender it to him even if he
is Ja'ir, as a display of obedience ... The j'ir
imm is joined to the others [in this rule]
because of the prevailing of his authority and
because of his not being dismissed for jawr.
For, if they refused to surrender it to him he
would fight them, even if they claimed that
they would deliver It to the recipients themselves,
because of their refusal to display obedience.6
Here it is unambiguously stated that it is incumbent to
pay zakt to the ruler even if the ruler is j'ir. His
jawr would imply that the zakt thus paid might not reach
its shard recipients, it remained valid zakt. On nlyya
314 . Ibid., 209.
35. Nawawl/Shirbrnl I, 1402.
36. Ibid.

120

Shirbini acknowledged that it was required, for ijz',

at the moment of payment to the sultan. However there was
a variant view:
the donor [who pays the sultan without
expressing niyya] gains ijz' because custom
indicates that what the sultan takes and
distributes to the recipients is nothing other
than incumbent zakt. This custom removes
from this religious duty the requirement of
niyya.
Here Shfi'I's original argument for the necessity of niyya
- to distinguish between the far and the nfi1a - is turned
back to front. Since ShirbnI deems it incumbent to pay the
zakt (far) to the sultan (even if he is ja'ir) that payment
becomes itself the sign of its being the zakt (far) and it
no longer has to be distinguished by niyya.
The whole development from ShfidI to ShirbInT
reflects the transition from idealism to pragmatism. The
right and duty of the individual to gain reward for himself
by paying zakt was gradually remdved and transferred to the
ruler. In the final view of Shirblnl the mere collection
of taxes by a tyrant ruler, (who might not assess them
correctly or distribute them justly) constituted zakt.
Neither the donor nor the sultan was required to undertake
niyya. Yet the donor gained ijz'. This was nothing more
than a bald assertion that religious reward was gained by
mere submission to the de facto power of the actual ruler.
His deviation from the ideal (jawr) did not preclude the fact
that what was put into effect was a sufficient approximation
to the shari'a for the members of the political community to
gain religious reward. The effect was certainly to render
sacred (shard) the ordinary acts of the people (who paid)
but it also rendered sacred the political acts of the ruler
(who exacted). The whole wss an expression of the support of
the religious classes for the de facto power.
37 Ibid., 1403.

121

For the ImmIs, while the imm is present zakt

should be paid to him or his suct or might possibly be
distributed by the donor himself. According to
T when the
su c t do not come or at a time when there is no imm (the

Ghayba) it is incumbent on the donor to undertake the

distribution of zakt by himself. He should not pay zakt
to the suln al-Jawr.
'-

J ) )

OJ'

--

jw

J Li >

This was qualified however by the acknowledgement

that the donor might distribute through his own waldi.

If

the individual undertook distribution without the intermediation

of the imm or the s'I - as must be the case during the
Ghayba - it was thought better not to appoint a wakll because
the individual could be sure of his own actions but only
doubtful about the actions of others. On the other hand it
was perm1ssble to deliver zakt to a trustworthy member of
the Shi l l fraternity and in fact It was better (afal) to hand
It over to the Culam? because they were more knowledgeable
about the 'places' of zakt, (i.e. the appropriate recipIents).8
There is some confusion here as it is thought 'better' not to
appoint a wakil; and yet 'better' to pay through the Culamaf
Since the imm was absent and the s'l not relevant to the
Ghayba the role of the 'ulam' was only explicable as trustworthy wakll to the donor.
In spite of this there is for the Buyid period some
evidence that the faqlh was already seen as to some extent
replacing the imm. According to r4urta:
It is better to disburse zakt . .. to the
imm or his khulafa' representing him; and
if that is impossible It is related [i.e. in
a Iiadlth?] that it should be disbursed to
trustworthy fuqah'
38. See, for all this,

Tsi, Nihya,

(ad niyya), 16, 20.

39. Nurtad, Jumal, 269.

185; rtabs, Zakt,. 13

122

And for Muf Id:

God has made it a duty for the urnma to carry
zakt to the prophet or the irnm, his khallfa,
...; if the khallfa is absent the duty is to
carry it to one of the khallfa's associates
appointed by him; and if the sufar' between
him and his people are not available it is
incumbent to carry it to the trustworthy
fugah'
5L

' i'

(,

L)Il L
C

c)
L'

c1414

fl
J' LL
cJ \Jii \
c>4
'1' }\ 40

The incumbency there suggested by Mufid was not

afterwards accepted. Nonetheless it was obviously considered
preferable in Buyid times that zakt should be transferred to
its recipients via the fugah'.
It is apparent from the
remarks of Murtad and Mufld that in some undefined way they
took over in this field the role of the imm or his representatives. They were not themselves however referred to as
representatives of the imm but rather as a kind of residual
authority which became effective precisely in the absence
of representatives. Tsi's detailed fur' provided for
distribution either through the imm or his Sad (both absent
or sgi) or the wakll of the donor. Presumably the fagih
acted as wakll to the donor.
Muhaqqiq too specified only three possible distributors: the imrn, the 'mil and the donor. In the absence of
the imm zakt was to be paid to a reliable Imml faqlh for he
was more cognisant of its 'places'.
Since the faqlh was
neither imam nor S a c i ( that category was sqi according to
0. Cit. in 'Allma, Mukhtalif, II, 16.
4l. Muhaqqiq, Shar?ic, I, 164-165.

123

-
Muliaqqiq) he must again have acted as wakil to the donor. L2
The faqlh however even if his role was formally identifiable
only as wakll to the donor performed the same functions as
the saci who was direct appointee of the imm. The impingement of some confusion on Muiaqqiq T s thought may be illustrated
by a comparison of a particular remark in the Shar'i' and
its parallel in the (later) Mukhtaar.
(Shar T 1 C )
J ,), AfJ)1.L i .-4,' W)I 1LJ' , t.
(Mukht a ar)
J) W .>
j&b
,)
'-'
If the imm or the s'T/faqIh takes the zakt

t
1

the donor's duty is fulfilled even if the goods

subsequently perish.4
MuIaqqiq was beginning to see the faqlh as taking on the
function of the s C I, direct representative of the imm.
'Al1ma reaffirmed in his Qaw c id al-AIikm the
traditional view. The zakt might be paid to the imm (absent),
the 'mil (still sqi?), the poor (the recipient) or a wakll.
During the Ghayba it should (yusta1abb) be paid to a faqTh.
Presumably again the faqlh acted as wakll (to the donor).
In the Tabrir however A1lma complicated the issue by specifying not four but five modes of distribution. It might be
passed either directly to the recipient, or to the sa'i, or
the wiI (= imm?), or the wakil, or, and this is new, the
Ikim. 45 It is known that during the Ghayba the
kim was a
faqrh acting as direct appointee of the absent imm. Earlier
assessment had not distinguished the role of the faqh in
distributing zakt from the role of a wakil to the donor.
'A1lrna's list at least suggested that the faqlh acted by
right of his representation of the imrn. The period was
obviously one of transition. Both Muaqqiq and 'A11ma
considered that zakt should be paid to the fuqah'. There
was some doubt as to In what capacity they undertook to
receive and distribute that zakt.
42. Cf. id., 14ukhtaar, 60.

Lt3.

Ibid.; and Shar'i', I, 165.

4L 'Allma, Qawacid, 23.

145.

'Al1ma, TalirIr, 67.

1214

According to Shahid I:
c!_& jZjJ

(LJ

L( Jt

iI

J'

'4. 3f -'

4',

-9)?

y' i

L
1

146

[a.] It is incumbent to pay zakt to the imm

on demand whether by himself or through his sI.
[b.] It is said likewise to the fagrh.
[c.] Payment to those [all of them] voluntarily
is preferable.
[d.] It is said incumbent.
That represents considerable change. Rule a. was traditional.
Rule b. was a direct transferral to the faqlh of authority
The
which traditionally belonged to the imm or his
faglh had a right to demand zakt and it was incumbent to
pay him. The justification of that view was obscured in
Rule c. was again not novel: it had
the anonymous gTla.
always been thought preferable to distribute akt through
a fagrh.

Rule d. suggested it might be incumbent. Shahid

I, anticipating the formation of a Shi d i state under the

Sarbedarids in Khurasan, produced a ruling whereby zakt
could be a compulsory tax, administered not by the government
but by the fuqah'.

Mufid it is true, some three and a half

centuries previously, had already stated something rather

like this but that statement had proved premature: it had
not found general acceptance. Centuries of legal effort
and a gradual refinement in the concept of clerical authority
had provided a context, a structure and a justification which
made Shahld I's assessment seem both appropriate and convinc ing.
Shahld II witnessed from his base in Ottoman Syria
the establishment of Shl n ism as the official religion of Safavid Iran. That did not in itself represent the official
Implementation of the shari'a but it offered opportunities
and inspired a final and comprehensive analysis of the nature
and extent of the faqlh's authority. Shahid II's commentary
on Muliaqqiq's Shartic, with regard to zakt, may be
146. Shahld I, in Rawa, 50.

125

described as an incorporation of the faqlh into all those

rules which for Mu1aqqiq had been relevant only to the imm
The clarity and integrity of his view however
or the s'I.
is best appreciated in the Rawat al-Bahiyya.
It is necessary to pay zakt to the imm
if he demands it in person or through his
s C I, because of the necessity of obedience.
It is said it is necessary to pay it to the
faqlh sharI during the Ghayba if he demands
it whether in person or through his wakil,
because he is n'ib to the imm like the

sI,

indeed stronger, (li-annahu n'ib a1-irnm

ka v 1s c T bal agw).
If the donor refuses (kh1afa) and distributes
the zakt himself he gains no iJz

because
of the refusal [to pay the faqlh] which
corrupts his act of worship (li'l-nahy al-mufsida

(sic) li'l'ibda).

To pay it to them

voluntarily without demand is better because

they are more knowledgeable of its 'places'
It is said ... it is incumbent to pay, without
demand, the imm or his n'ib or, during the
Ghayba, the faqlh ...; [ those who say this]
cite as evidence ... (Q9.103).., for, rendering
it compulsory for him [the imm] to collect
requires that it be compulsory for them [the
donors] to pay, and the n'ib is like the munawwib.
-
147
The more prevalent view (al-ashhar) is istilbab.
The more prevalent view is not necessarily Shahid II's.
This passage asserts categorically that the faqlh may demand
zakt and if he demands it, it is incumbent to pay him. He
may demand it in person or through his wakll. So it is
envisaged that the faqlh might send out tax-collecting deputies
to collect zakt. It might be incumbent to pay zakt to a
147.

Shahid II, Rawla, 50.

126

faqlh even if he does not actually demand it. Shahid II

justifies this view with a quranic quotation and with the
assertion that the n'ib is like the munawwib. That is,
the faqlh (nib) is like the irnrn (munawwib) and possesses
accordingly all of his rights. The two propositions
rendered available in this passage that the faglh is n'ib
to the imm and that the n'ib is like the munawwib
constituted the foundation of Shahid IPs vision of the
execution of the shar'a.
The seeds of that vision were
not absent in Buyid times but it took almost six centuries
for the full implications of the imm's absence to become
clear. The Buyid fuqah' perceiving that their interests
were bound up with the preservation of the shari'a as ideal,
free from involvement with the de facto power, had declared
several aspects of the sharIa sgi. By the time of Shahid
II it had been perceived that most of the sgit functions
should in fact be under the control of the faqih. That
development probably reflects a real consolidation of authority
under tolerant governments and was fostered by the elusive
possibility of actually forming a t state' subject to sharI
rules and governed by the clerical class.
It will be noted that as the faqlh asserted his
right to administer zakt the individual lost his right to
distribute it himself. It was gradually perceived, by the
Shi c a, that the authority delegated by Ja'far al-diq to
the fugah' covered not only strictly judicial activities
but also control of zakt.
This was only part of a gradual
process whereby most of the s q it functions of the 1jrm were
brought into the sphere of ga'. That process while enhancing
the authority of the clerical classes, restricted the degree
of participation of the non-clerical individual in his own
salvation. Some indications in early texts (Tsi, Ibn Idris)
that the individual might be personally responsible for the
fulfillment of the religious duty of zakat were overlaid
and finally removed by the imperatives of clerical control.
A similar process was evident in Sunnl sources.8
L8 Above. 117-120.

127

There, the right to collect arid distribute granted to the

accepted sharLi executive, the de facto ruler, pre-empted
the right of the individual to assess and distribute for
himself. For the ShIs the accepted shard executive was
manifestly not the de facto ruler but the faqlh.

The
developments in Sunn theory were accompanied by a removal
For TsI
zakt might be paid either through the wakil of the donor
or through the imm/sdI.
In the former case both the
of the need for niyya on the part of the donor.

donor and the wakrl were required to express niyya: if either

of the two neglected it there was no ijz'. In the latter
case - a merely theoretical possibility - the donor had to
express niyya, but not, necessarily, the imm. If however
the 1mm took zakt by force, then the unwilling donor gained
- 149
ijza'.
Niyya then during the Ghayba - when the imam cannot
take zakt forcibly - is absolutely essential on the part of
the donor and is required twice if the zakt is paid through
a wakil. The faqlh in Buyid times was probably thought of
as wakTl, in precisely this sense. 'Allma in the Ta1irr
agreed with TsI. 5 In the Mukhtalif he disagreed. Where
payment was effected through a wakll niyya was required only
once either by the wakfl or the donor, not both. When payment
was effected through the imm/s'I CAllma denied any
distinction between compulsory and voluntary donation: in
either case niyya was not required on the part of the donor.51
It thus fell for 'Allma that whether the faqlh as distributor
of zakt was wakil to the donor or n'ib to the im g m, the
burden of niyya could be removed from the donor and transferred to the faglh. The religious reward (iJz') of the
individual might be vicariously acquired simply by his
149. TsI, Mabs, Zakt, 13; cf. Shfi, Umm, I, 19;
another example of SunnI/ShfidI influence on Shi'I furi.
50. C A11ma, Ta1rIr, 67.
51.

Id., Mukhtalif, II, 21.

128

submission to the authority of the clerical class.

According to Shahld II:
t7,'

tUt

aJW j, ,bJ3 t <',V,>

LjV CC j-i; ' ,' ;; 4at

cJ't-r->

C_-,

cJ

>-

L> >'
lit C
Though complex both the terminology and the content of that
passage are important. It may be analysed as follows:
a. Niyya is incumbent with regard to
zakt al-fitra and zakt al-ml.
b. [It is incumbent] from either the donor
or his wakil, [not both].
c. [It is incumbent] at the time of payment
to the recipient or to his [the recipient's]
wakil.
d. [The recipient's wakil may be] 'umiim,
in a general way, i.e. the imm or his n'ib
____ or his n g 'ib khs
e. [Or the recipient's wakll may be] khus,
that is his [own personal] wakll.
f. If the donor does not express niyya on payment
to anyone other than the recipient himself or
the recipient's personal wakll
g. [i.e. if the donor fails to express niyya
when he pays to the imm or the n'ib 'grnm
or the n'ib kh] and if the receiver
subsequently expresses niyya on disbursement
[the donor] gains reward.5
It is here confirmed that if the donor pays through the faglh
(the n'ib , rnm of the imm) - and according to Shahid II
it is incumbent that he in fact do just that - he need not
undertake niyya because he will gain reward on the basis of
52. Shahld II, Rawda, 51.

129

the faqih's niyya when he distributes the zakt. Just as

the SunnPs religious reward depended finally simply on
submission to the authority of the de facto governor, the
ImmI's religious reward depended simply on submission to
the authority of the clerical class. 53 They depicted
themselves as responsible for the collection and distribution
of zakt; they further analysed the tax-situation in such
a way that their control of zakt was uncircumscrlbed by the
rights or duties of donors. The donor became most certain of
reward by unqualified surrender of his zak g t to the fuqah'.
The faqlh in this assessment of zakt-distribution was seen
to be not wakll to the donor but wakTl to the recipient, a
position acquired by virtue of his status as n'ib to the
imam.
The fugah' were permitted to distribute the zakt
In a rather arbitrary fashion. It was preferable (mustal2abb)
to pay something to each of the eight recipient categories.
But it was not incumbent to deal equally with each group, in
fact it was better to give preference to the 'superior',
bal al-afal taf.Il al-murajJah. That, from Shahid II,
should be compared with his incorporation of the clerical
class Into the recipient categories of fuqar' and maskln.
It is permissible to pay the zakt to only one of the eight
categories, or even to only one individual within a category.
Further it was permissible to provide ighn', that is to
provide more than a sufficiency: there were no limits to
how much could be given as long as it was paid at one time.
The fugah' gave themselves a great deal of discretionary
power In utilising and distributing the taxes, payment of
which to them they thought Incumbent.
Shahid II's terminology in that passage should be
53. The general trend of clerical thought may be illuminated

by consideration of the words of the Grand Inquisitor

514 .

to Christ in Dostoevsky, The Brothers Karamazov, I, 2914-305.

Shahld II, Rawa, 50-51.

130

noted. He refers to the n'ib

mm of the im gm.

That
locution refers to the faq
if fully qualified in accord
with the (judicial ) delegation from Jac far a1-diq. The
significance of the term n'ib kh

is not here unambiguously

evident. Later discussion will demonstrate that it designates

the same entity as was earlier designated man naabahu al-imm;
that is, a contemporary of the imm appointed by him for one
or more purposes.55

Khums was, like zakt, a ritual duty incumbent on

individuals, a farTda al 1-'ayn. Distinguished from the
other 'ibdat by being as to immediate aim human rights and
not divine ones (aqg li'1-damiyyin nQt liagg 1i'11h)56
the final aim of both khums and zakt was nonetheless divine
reward and more certain salvation. The quranic basis for
khums was found in the verse

(8ii):

Know that whatever you acquire [as material

gain] a fifth belongs to God and to the
Prophet and to those related and the orphans
and the poor and the wayfarers.
The ImmI fugah', notwithstanding an early- or pre-Buyid
assertion of a five-part division, generally inferred from
this a six-part division of khums.
After the death of the
Prophet three parts belonged to the 1mm and the other three
parts to the poor, orphans and abn' al-sabTi of the Ban Hshim.
With respect to the last three categories, except for an
isolated taraddud expressed by Nuhaqqiq, Imri (ImmI faith)
was a condition required in the recipients. 'Adla was not
generally required. Need (fuqr) was specified in all three
categories. Tusr alone was doubtful If fuqr was required in
orphans; subsequent fuqahT considered that it was. The
55. See below, 1148 and 150; 152-153; 163-165.
56. E.g. Ibn Idris, Sarir, 106.

131

t wayfarers ?

were deserving if they were in need during their

journey even if they had a sufficiency in their home land.

Whereas for Sunnls khums was a concept applicable
only to booty gained in war, the quranic phrase m ghanimtum
mm shay' was taken by the ImmIs to refer to seven categories
of goods. These categories did not change through the
centuries. Those cited by Shaykh-i BahI Cd. lOi3/i6o4-5)
are exactly the same as those provided by Muhaqqiq and were
all available in a slightly less systematic form in TsI's
They are, from c Al1mats QawE'id: 1. booty taken in
Nihya.
war; 2. minerals; 3. treasure-trove; LI. what is taken
from the sea (pearls or treasure); 5. the profits of trade,
agriculture and craft (tijrt, int, z1rt);
land If bought from a Muslim;

7.

6.

Dhimm

tiall goods mixed with I)arrn.57

5.

was subject to al-Iawl, I.e. It was an annual tax.

There was considerable ikhtil g f on the question of nab as
applicable to categories 2. 3. or 4. Khums was due on booty
Category

taken In war only when the war was legitimate and that,
according to TsI, had not occurred since the time of 'AlI8
With the gradual reassessment of jihd as legally viable
during the Ghayba 59 it is possible that khums on booty became
also potentially If not actually viable.
As long as the imm was present he was personally
responsible for distributing khums. There were no restrictions
on what he might do with his own shares; he might spend them
on whatever he wishes, family expenses, whatever burdens he
had to bear, (m yalzimuhu mm taliammul al-athgl) or providing
for others (ma'unat ghayrihi) etc. The remaining three
shares he should distribute to the appropriate Banr Hshim
recipient categories according to their needs and annual
expenses calculated on a basis of moderation (igtid). It
57.
58.
59.

214; cf.
Tsi, Nihya, 196-197.
Mabs, Jihd, 9.
Below, chap. VI, 1117_158.
Al1ma, Qawid,

Bah'I, Jmi, 101-102;

132

was thought that if there was a surplus after disbursement

to the recipients then it belonged exclusively to the imm:
Conversely if there was a deficiency
kna lahu khssatan.
he must make it up from his own wealth (? share of khums):
4It

If that meant that he was bound to use his share of the

khums to provide for deficiencies in the recipients? shares
there was some logical inconsistency in TS'5 statements,
since he had also stated that the imm might do with his
share whatever he wished. The question was to become
controversial. 60
Anfl was a term used by Immi fuqah' to indicate
goods which belonged exclusively to the mm. Amongst those
were various types of land and material wealth as well as the
imm's share of khums.
tisufruct of the imm's property
without his permission was thought rebellious. The imarns
however had provided their followers during the Ghayba with
a dispensation (tarkhls) as to mankili, maskin and matjir.
There was considerable speculation as to the meaning of that
dispensation. The idea emerged that ImmIs alone were
permitted to trade in certain goods which ideally belonged
exclusively to the imm.61 The dispensation did not include
Sunnis; so their trading in such goods confirmed them in
rebellion. Both Ibn Idris and CAllma added to their assessment of that dispensation: let none imagine that if he
make a profit in such trade he should not extract the khums.
The dispensation in other words did not effect the operation
of khums.
It provided a legitimate basis for the daily
life of the ShIa, while emphasising the essential rebelliousness of Sunnl activities. Khums as a tax remained valid even
in the absence of the imm.
was uncertain about what should be done with
khums during the Ghayba and pointed to the absence of a
specific revealed text, on this matter. He suggested four
60.

Mabs, Zakt, 20-21.

61. Ibn Idris, Sarir, 116;

Allma, Qawid 2; Ta1irr, 75.

133

possibilities.
1.

All goods are to be considered in the same way

as maskin, matjir etc. That is, a waiving of khums

completely in view of the recorded dispensation. This view
was dismissed as being opposed to i1itiy ( t'caution") and
constituting the usufruct of property without the owner's
specific permission.
2.

Khums is to be preserved as long as the donor

is alive; when death approaches he should appoint as waI
over the goods a reliable member of the Imm fraternity
and so on until the goods may be delivered to the irnm.
3. Khums should be buried because the earth will
expel what is in it on the advent of the imm.
ti. Khurns should be divided into six parts; the
three parts belonging to the imm should be buried ( option
3) or consigned as wadT c a to someone trustworthy (= option 2).
The other three parts should be distributed to the appropriate
recipient s.62
TsI then indicated very firmly his preference for option .
He did not indicate who should undertake the distribution of
the Bani Hshim's shares. 6 It is nonetheless highly probable
that by virtue of their influence and their knowledge the
fugah' were able for khums as for zakt to exert considerable
control over the matter of distribution. No role however at
this stage was specifically asserted for the fuqah' as such.
Trustworthy ImmIs (who might frequently be members of the
clerical class) were under certain circumstances to look
after but not to use the imm's goods.
Ibn Idris is renowned in ImmI biographical literature
furi c . Indeed scarcely
for his unmitigated attacks on

TSI'S

a page of his Sar'ir a1-Is1m does not contain some carping

comment about TSI'S inconsistency, his reliance on akhbr hgd,
his going beyond
hir al-tanzll or Cumrn al-ya, his deviation
from bar'at al-dhimma. In spite of this, large sections of
the Nihya or the Mabs
were quoted verbatim by Ibn Idris.
It is clear that TsI whether attacked or followed was the
62. For waiyya and wadIa see Schacht, Introduction, 119-120,157.
63. TsI, Nihya, 200-291;

'iabs, Zakt, 21.

134

dominating influence. The discussion of khurns in Sar'ir

al-Islrn displays a polemic intensity rare in

ShI'I works

of figh and manifestly reflects a real contemporary concern

for an issue not unapt to rouse controversy, namely control
of an extensive body of wealth. The centre of the controversy
for Ibn Idris lay in the rule summed up by T
1 as that,
while the imm is present, if there Is a surplus of the Bani
Hshim's share of khums it belongs exclusively to the imm
and if there is a deficiency he must provide for It.
According to Ibn Idris:
If there Is a surplus he [the irnrn] becomes
the Ifiz and mutawalli for the preservation
of that on their [the BanI Hshim's] behalf.
He may not appropriate any of It for himself
for the right is theirs and he may not take
any of their property.
Ibn Idris here was actually contradicting

TsI

and other

earlier writers. This he acknowledged:

[As for] that which Is found in the writings
of some of our companions to the effect that:
the surplus belongs to the 1mm exclusively

TsI];

(kna lahu khatan)[- a quote from

the meaning of that Is that the imm may look
after it, protect it and manage It but he may
not own it.
Ibn Idrls defended his interpretation of his predecessors
by adding a grammatical argument which suggested that It
was or might be permissible to use terms indicating ownership
to denote a relationship which was less than ownership.
Having thus re-interpreted his predecessors he was enabled
to state:
There is no khi1f amongst Muslims in general
nor amongst the ShI c a in particular that with
regard to the shares of [the BanI HshIm]
the imm has no right to any part of them,
rather he has his shares and they have theirs.
That flagrant disregard of the 'plain' meaning of his

135

predecessors' words Ibn Idris further defended by careful

exegesis of the quranic na

(Q8.41), which enabled him to

claim the Qur'n together with ijm as dalils supporting

his view. He acknowledged then that there were two Traditions
in Tusr's Tandhlb which might support the opposite view but
they were according to him akhbr

mursal not rnusnad

and in the isnd of at least one of them there was a cursed
kfir.
Since the generality of Traditions may be said
to be in favour of separate ownership by the imm on the one
hand and the Bani Hshim on the other Ibn IdrTs claimed in
support of his view sunna as well as Qur'n and iJm'.
To this he added adillat al_ c ugili in so far as it is not
permissible to utilise without permission the property of
another. That was a purely rational perception of right
action.
To that four-part proof supporting his view Ibn
Idris added 11itiy, caution, with respect to bar'at al-dhimma.
Bar'at al-dhimma refers to the most economic fulfillment of
religious duty, without the addition of extra details which
might be irrelevant or inimical to God's command. Caution,
1htiy, requires the fulfillment of the command and the
eschewing of extras.
i

'

.t-) L

--' LI

L< )'
\j
1a

4.)

J 05LU 6

If there were in this matter only the way of

i1tiyt with respect to bar'at al-dhimma it
would suffice. For, the dhirnma is dependent
on the wealth [the khums] and its reaching
its rightful owner; if the donor ensures
that, then the fulfillment of his duty
[bar'at al-dhimma] is certain but if he gives
it to other than the owner then there is
opposition and the fulfillment of his duty
is not certain.

136

There, in fact, Ibn Idris betrayed his real

concern: the problem was not whether the surplus from the
BanI Hshim's share belonged exclusively to the imm, but
what the donor was to do with his khums during the Ghayba.
Ibn Idris wanted a strict separation of the imgm's share
from the Bani Hshirn's share.
We have not found any work by our companions
in which after mentioning this problem he
does not say ... that half the khums should
be bequeathed (y) to its owner [the imm]
or preserved for its owner (yutifa) or stored
for its owner, with a variety of
as
expressions. If these writers had desired
that someone other than the imm should have
a right to it during the Ghayba, or that it
should be consigned to someone other than he,
they would be contradicting themselves.
Ibn Idrls desired that the imm's share should not during th
Ghayba be distributed to someone else, above all not to the
Bani Hsh1m; it was precisely the confusion of their share
with the 1mm's that he most inveighed against. Presumably
the opponents who called forth this extensive piece of
polemic figh had advocated the distribution of all the khums
to the Bani Hshim.6
When he came to discuss what precisely to do with
the khums during the Ghayba Ibn Idrs followed indeed paraphrased indicating four possible options and his very
decided preference for the fourth. Three parts should be
distributed, the other three parts preserved, passed on etc.
Before leaving the subject Ibn Idrls produced a flurry of
lengthy quotations from Mufid, Murtad and TsI designed to
illustrate that none of them suggested that the imm's share
during the Ghayba should be distributed to the BanI Hshim.6
6I. Ibn Idris, Sar'ir, 114-ll5.
65. Ibid., 116-118.

137

The evidence would seem to suggest that Ibn Idris

was right: the earlier fugah' had suggested that the imrnhs
share of khums should be preserved as waiyya or wadl c a until
the advent of the imm. Ibn Idris's polemic indicates the
emergence of a novel contrary view that the imm's share
should also be distributed to the Bani Hshim. That view
was justified by exploitation of the unresolved ambiguity
of expression of earlier fuqaha': did the imm possess his
share of the khums exclusively or was he required to provide
from it for the deficiencies of the BanI Hshim? The polemic
suggests a struggle but the nature of the struggle remains
unknown. It is probable that a certain income accrued
to the clerical classes from both zakt and khums through out the Saljuq period. Part of that income was no doubt
distributed in accord with the shari ca, part of it certainly
went into the upkeep of mosques and madrasas and personnel.66
The technically unusable part of that income, the irnm's
share of the khurns may have been stored as treasure in a mosque
or shrine. That part, Ibn Idris suggested, should continue
to be stored in perpetuity. The opposing group suggested it
should be distributed to the Bani Hshim recipients. That
would be beneficial to the fuqah' in general (if they were
the distributors) as increasing their practical control of
a usable body of wealth. It might have been particularly
useful to those of the clerical class who were also Sayyids
(BanI }ishim). 6 Ideologically any claim by the fuqah' to
control the imm's share of khums differed from their control
of zakt in that khums unlike zakt was actually the property
of the imm. By claiming the right to distribute khums the
fuqah' asserted a certain relationship between themselves
and the imm.
and disregarded Ibn Idrls,
r'IuIaqqiq followed
66. Cf. e.g. Qazwn, Naqq., l6.
67. For the relationship between Sayyids and CulamT, see
Lambton, The Persian Ulam', 250.

138

in acknowledging that while the imam was present any surplus

in the Bani Hsh1m's share became his, and any deficiency
had to be supplied from his share. 68 This, not unnaturally,
in view of Ibn Idrls's polemic, was a preliminary to the
production of a new option regarding what was to be done
with the khums during the Ghayba. Muhaqqiq cited the four
options presented by T SI and Ibn Idris and added that the
1mm's share too might be distributed to the Bani Hshim.
Justification for that view was of course that it was incumbent
on the imm to supply any deficiency. 6 Both the premise and
the conclusion had been denied by Ibn Idr3s. Muhaqqiq
confirmed his preference for the distribution of the imm's
share in the Mukhtasar. 7 He like T 1 and Ibn Idrs before
him failed to specify who was to distribute the Bani Hshim's
share but of the imm's share he had this to say:

Lji

LM

th; V ut

y.
That passage is hardly perspicuous but the following translation is proposed:
He must distribute the imm's share ... who
possesses [judicial?] authority by right of
niyba just as he undertakes fulfillment of
what is incumbent on the absent [imm]; [or:
just s some-one undertakes the duties
incumbent on one who is absent (- a reference
to wakla?)].71
That is admittedly tendentious but the ambiguity is Muhaqqiq's.
There is little doubt that the person in question was the
faqlh but the tortuous reference to him reflects perhaps a
68. Muhaqqiq, Sharhic, I, 182.
69. Ibid., 1814.
70. Id., Mukhtaar, 614.
71. Id., Shari', I, 1814.

139

real problem, namely how to justify the clerical class's

claim to administer wealth which unambiguously belonged
to the imm.
The implications of Muiaqqiq's locution are
that the acknowledged judicial niyba had a general not a
specific import. The 1mms had delegated also to the fugah'
control of their property or at least of the distribution of
khums.
A specific text (khs) was perceived to have a more
general (rnm) denotation.
'Allma in his TaIrIr repeated verbatim regarding
the distribution of khums Mul3aqqiq's man ilayhi al-Iiukm biliagg a1-niyba. 72 In the QawCid however he more bluntly
asserted:
t' -s- cg

W - - -

During the Ghayba ... the only one who may

undertake the distribution of the Imam's
share Is the 1kirn.73
It was in other words the fagTh who q ualified for the ex ante
delegation of Jadfar al-Sdiq who had also a right to deal
with and distribute the imam's khums.
It was in his capacity
as judge (1kIm, n'Ib a1-Imm) that the faqTh distributed
the khums.
The superficial tendency of the jurisprudential
argument was to assert the right of the fuqah' because of
the acknowledged delegation. It Is not impossible that a
de facto or generally accepted control of khurns by the fuqah'
made this ritual a suitable context for asserting that the
niyba of the fuqah' was a general one. CAllma did not
specify for his view ijmC, it was rather the strongest
available option (aqw).7
It may here be noted that the accepted mode of
passing on Intact the im gm's share of the khums had been
wasiyya and wada.
Juridically the ultimate residual
authority over both of these was alegitimate judge. In the
72. 'Allma, Tal)rlr, 7-75.
73. Id., Qawid, 14L1.
714 . Id., TaIirIr, 714....75.

140

various events of a wa

or mida c dying, disappearing or

proving incompetent or corrupt the relevant goods fell
under the control of the judge. In the course of generations
it would inevitably happen that acknowledged judicial
authorities would find themselves in control of a body of
wealth derived from khums payments: as such they would seem
to be acting as n'ib to the owner of that wealth: not the
donor, but the absent imrn.
It was perhaps thus in response
to a de facto situation that the law developed in the way
that it did. The assertion that the fuqah' could actually
distribute the wealth they controlled was a legal innovation,
perhaps again justificatory of what actually happened, but
certainly a distinct increase in the discretionary authority
of the clerical class.
Shahid I stated about khurns:
It is divided into six parts, three of which
are for the imm to be paid direct to him, or
to his nuwwb as long as he is absent, or to
be preserved (yu1fa) ...
That illustrates the general tendency to recognise in the
Cu1am ? deputies (nuwwb) to the imm. The terminological
distinction between nib khss and n'ib CyUfl does not
however appear to have been available to Shahid I. He did
not specify in the LurnCa how the nuwwb were to use the imm's
share.
Shahid II in the Maslik a1-Ifhm took up the
problem of the distribution of the khums. That problem was
related to the question whether, if there was a surplus in the
share of the Bani Hshim, it belonged to the imm. If it
did, then it became permissible to argue that, likewise, the
share of the imm during the Ghayba belonged to the Ean Hshim
and so could be distributed.
Shahld II commented on the
assumption that a surplus in the share of the Bani Hshim
belonged to the imm as follows:

75. Shahld I, in Rawa, 53.


]J41

(
>

1ii

L 5

iit

'-

'> " >

.9

'

)'

)
-

Sj

Ibn Idris denied that view because of its

being opposed to

hir ai-ya [i.e. Q8.'l]

which requires for each category a specific
share. But the exegesis (tafTl)[of the verse]
is definite as to the sharing [of the wealth
between the categories].
The going beyond [the specific content of]
relevation here constitutes [a.] naskh al-qur'n
by khabar al-d or what is less than that and

[is based on][b.]the loss of the usefulness

of distribution.
The resolution is then [in view of the fact that
a. and b. are insufficient?] that Ibn Idr3s's
view is reasonable were it not opposed to the
views of our great companions. 'Allma was
uncertain for this reason.6
That text again Is by no means easy to understand but if
correctly translated here It represents an interesting example
of the arbitrary nature of juridical argument. Shand II
confirmed on the following page that it was in his view
Incumbent on the imm to make up any deficiency in the share
of the Bani Hshim. That, for him as for earlier fugah',
constituted the essential premise which justified the
distribution of the imm's share during the Ghayba by
appropriately qualified clerics. He appears however to have
conceded that his view could only be supported by isolated
Traditions - which could not technically abrogate the Qur'n and by an intellectual argument which again of itself was
insufficient to modify a revealed text. Ibn Idrls's opinion
76. Shahld II, Maslik_al-Ifhm, II, 68, line 20 ff.

l42

was then muttajih, reasonable, were it not for the fact that
great fuqaha' had opposed it. There was in the matter no
question of ijm' it was avowedly a matter of ikhtilf. The
assertion of 1khtilf rendered Shahld II's view a permissible
one but it remained indefensible by strict reference to the
four ul.
To remedy that the opinions of major fugah'
were presented as sufficient per se to justify a juridical
rule. T 1 had put forward his opinion in a spirit of
speculative exegesis as to what might be incumbent on the
imm if he were present. He did not consider that the imm's
share of khums was to be distributed during the Ghayba. Later
fuqah' exploited that ruling in order to justify their
distribution of the imm's share in this period. Shahid II
remained apparently unable to account for that ruling. The
only discernible motive for its preservation was that it
served the interests of the fuqah'.
Shahid II's later analysis in the Rawat al-Bahiyya
of the rules relevant to distribution of khums sufficiently
indicate his assessment of the scope of the fugah''s
authority.
[The imm's share] is to be conveyed to him
if he is present or to his nuwwb; they are
al-fuqah' al_ C udij i al-immiyyTn, those who
possess the shar'ii al-fatw, because they
are his waklls. Thereupon, it is incumbent
on them to act in accord with their madhhab:
those who consider it permissible to distribute
it to the recipients by way of fulfilling
[deficiencies], which is the prevalent view
amongst the muta'akhkharTn, may distribute it
as they see fit generously or otherwise; those
who do not accept that view should store it
for the im gm as wadi c a until his appearance
Thus, as long as the imm is absent; or it
should be preserved (yutifa) by the donor of
khums as wadlca just as we have described for
the n'ib [ = faqlh].
The donor may not

1143

distribute the imrn's share of the khums

to the recipient categories under any
circumstances, may not in fact give it to
anyone other than the
kim shard. If
someone does undertake that he is held
accountable (min).
It is apparent from the absolute nature
of Shahid I's statement that it should be
handed to the nuwwb that no aspect of that
share during the Ghayba is Iiall to anyone
other than its group [farigihi = ? the group
in charge of khums, the fugah'].77
That passage as usual in Shahid II clarifies the unresolved
ambiguities of earlier writers by stressing the authority
of the clerical group and minimising if not in effect denying
the rights of the individual donor. It may be noted that
even if the latter stored his goods as wadi c a for the imm
they would in the long run fall into the hands of a judge.
The terminology here applied to the fugah' is clear and
reflects on the one hand the absence of any limitation to
their deputed authority and on the other the judicial origins
of that authority. They are n'ib and wakil to the imm,
must possess the shari al-fatw, (i.e. those specified in
the delegation traditions) and are the 1k1rn shard. A
1kim shari is a real judge whose authority derives from
Ja'far al-Sdiq's delegation and is to be distinguished from
the appointee of a de facto power, who might or might not be
a Ikim shar'T.

Shahid II portrayed the financial affairs of the

Shi c i community as being ideally completely subject to the
authority of those members of the religious class who had a
right to judicial authority. The idea had emerged as early
as Mu1aqqiq, but for him only with regard to khums, and
77. Shahld II, Rawa, 53.

1)414

then uncertainly. The application of the concept of judicial

niyba in such a way that it became the basis of the clerical
class's control of all shar'I finances, zakt as well as
khums, is barely discernible in earlier works.
The details presented in this chapter were chosen
to Illustrate both the fact and the manner of juridical
development. As to the former, even if change was from
generation to generation subtle and gradual, the difference
between T 1 and Shahid II is sharp and clear enough. The
manner of development may be summed up as exegetical and
terininological innovation. The motives of development are
not readily extrapolated from jurisprudential texts but may
to some extent be inferred. The doctrine of Isq
elaborated
by Murta and

1 suggests a cautious approach to the de

facto political power and implies the absence of any aspiration
to oppose or replace the extant power. That caution reflects
T

the extreme degree of tension which subsisted through the

BuyId and early Saljuq periods between the Sunril majority
and the minority ShI'I community. By late Saljuq times that
tension had to some extent been dissipated. The K. al-Naqd
presents a picture of the Shi t i community as
on the whole flourishing, rich and prosperous, with a considerable number of mosques and rnadrasas, 8 abundant wealth to
(c. 560/1165)

be distributed in the form of zakt and khums 79 and increasing

influence in court circles.80 Subsequently under for example
the caliph al-Nsir and then under the Ilkhanids tolerance
and patronage from the de facto governments provided the
Shi c i community with opportunities-to consolidate its internal
social structure and to establish its control of material
wealth. Throughout this period friction between Sunrils and
ShI'Is though endemic does not appear to have been so violent
or so Intense as during the Buyid period. That may partly
78. Qazwlnl, Nag, 147-148.
79. Ibid., 1614.
80. Ibid., 53511, 71.

1145

be related to the fact that the ShI d is had established

themselves in certain areas as the dominant community,
notably in South Iraq, Bahrayn and the East coast of the
Arabian peninsula, parts of central and northern Syria
(Aleppo, Bacalbek, Tripolis) and scattered areas of Iran.
The Syrian community was already well-established in Buyid
times8l and might be related to the pro_Shi c i tendencies
of the Hamdanid princes. Generally however the history
of that community is still more obscure than that, obscure
enough, of the other Shi'I communities.
They existed nonetheless and apparently flourished.
It is probable that the internal affairs of the community were
managed by legitimate magistrates who emerged locally in
close association with noble families (Sayyids), themselves
constituting minor and local dynasties.82 The re1ationshi
subsisting between the Shi d i community and the de facto
power might be compared to that described by Goitein as
subsisting between the Jewish community of Egypt and their
government. 8 Both the Jews and the ShI'Is constituted a
community within a community, variously linked to the political
structure but preserving a great degree of independence from
it. The Shi d is had their own law and their own magistrates
whose authority was no doubt rendered more effective by
careful cultivation of good relations with the actual ruler 8L
and acquisition of his practically useful albeit technically
illegal decrees of appointment. KulaynT's delegation Traditions
It may be thought reflected already a need for formal
judicial sanction for the community's daily activities such
as marriages, wills, resolution of dispute etc. The real
81. As witness the letters of

and Murtad to Aleppo,

Tripolis, Tyre, Sidon, Damascus etc.; KhwnsrI, Rawt,
385, 582.
82. See e.g. Qazwlnl, Nag, 229ff.

83. Goitein, A Mediterranean Society, Vol. II.

814. Qazwni, Naq, 53_514, 71.

1146

authority of the clerical class, in the course of time,

with successive tolerant governments and declining friction
with Sunnis, became actually greater and was bolstered by
the acquisition and control of material wealth in the form
of awqf as well as zakt, Ithums etc. Legal theory both
justified and promoted that authority. The reality of power
in turn Influenced and dictated developments in legal theory.
The juridical developments described above display an
increasing trend to express, to defend and to promote the
authority of the fuqah'.
That was a reflection of their
aspirations certainly but also of the reality of their
achievements. Thus amongst the motives of juridical development were the justification and promotion of the real effective authority of the Shi'I fuqah' within their community.
The relationship between the sacred community (the
dr al-Imn) and the profane power, as depicted in juridical
works, underwent little or no change from Buyld to Safavid
times. The ruler was always
lim/j'ir. The ShII community
If they submitted to his power did not concede his authority.
Shard activities took place within the community of believers.
Tax-collecting other than that undertaken by the fuqah' was
essentially profane and Irrelevant to the question of
salvation.

lL7

CHAPTER VI
THE GENERAL DELEGATION
The executive functions of the sharica which for
the Sunnis devolved upon the de facto ruler, for the ShIca
were to be implemented large:L y , if not completely, by the
fugah'. The single most important exegetical support for
the supposed delegation of authority to the Shi d i clerical
class was found in the delegation Traditions from Jafar aldiq, application of which in the earlier iadith collections
was exclusively to judicial authority. Exegesis of those
Traditions proposed a legitimate judiciary whose control of
communal affairs owed nothing to the existence of a profane
power. That judiciary, one aspect of the clerical class,
was able through the centuries not only to extend in practice
its real control over communal affairs but also to justify and
enhance its authority by 'depicting it as essential to the
fulfilment of various shar t l dut-les. The manner in which the
spheres of zakt and khums were brought under the control of
kim shar'I has been illustrated. Similar developments
the
might be found in other areas of juridical discussion.
However, the extent and limitations of clerical authority,
together with the exegetical and terminological devices
which were the instrument of juridical development may best
be illustrated by analysis of ShI' attitudes to jihd and
Friday prayer.
Shahrd II wrote concerning jihd:
It is of various types:
1. jihd against mushrildn which is begun
to call them to Islam.
ii. jihd against those kuffr who threaten
Muslims in such a way that they fear the
conquest of their lands and possessions,
even a little.
iii. jihd against those who desire to kill a
nafs mu1taram [?=Muslim] or to take his wealth
or enslave his women ...

1L8

iv. jihd against bught

It is Incumbent 'all-kifya
It is conditional on [the presence of] the
imm Cdjl or his n'ib kh; that is the
one appointed (manb) for jihd or for what
is more general than Jihd. As to the n'ib
'rnm such as the fagTh he may not undertake
[jihd] during the Ghayba - in the first type
[i.e. Jihd for conversion].
But this [the presence of the imm etc.] is
not conditional for the permissibility of
the other types of' jIhd, nor for the attack
of an enemy on Muslims whereby they fear for
bay.at al-islrn ... Hence that [type of jihd]
Is incumbent without the permission of the irnm
or his n'ib. It is understood from the
condition (al-gayd)[i.e. attack on Muslims and
bayat al-islm] that the enemy Is a kfir for there Is no fear for Islam from a Muslim
even If he is an innovator (mubdic)[All of this
Is a reference to type ii.],
s%

,; ,

J%

c..

'I
)I

.i

14% c.r

cj)

t)'

01% , &

5)

a J' n.'

(S'

L% t.)'
' a

However if they fear for themselves they must

undertake defence (dif') or if there is fear
for some Muslims he (sic) must undertake it.
And if one is Incapable it Is incumbent on his
neighbours and so on; and this is not jihd.
[All of this is a reference to type iii.]
Rib: that Is guarding the frontiers of bild
al-islam in order to provide warning as to the

lLt9

situation of the mushriks.. . - It is musta1abb,

definitely, always, with the presence of the
imm or during the Ghayba. It must be for at
least three days ... and at most forty days;
if it is more than that it is joined to jihd
in the matter of treligious] reward (thawb),
but it does not go beyond the designation (waf)
rib:
It is iarm to stay in bild al-shirk for one who
cannot display shi'r al-islrn, i.e. adhn, alt
etc ......The writer [Shahld I] in what is
related from him added to bild al-shirk bild
al-khilf wherein a mu'min cannot undertake shir
al-Imn, as long as he is able to travel to a
country where he can.
Those to be fought: a. the liarb ... [i.e. nondhimn kuffr]: this is only incumbent after the
call to Islam ... and the caller is the imm
or his n'ib
b. the kitbI, i.e. Jews,
Christians and Zoroastrians
Chapter on a1km al-bught ... It is incumbent
to fight such a one if the mm (sic) appoints
(idh nadaba al-imm) until he returns to
obedience to the imm ... Fighting him is like
fighting kuffr, as to its wujb Cal l-kifya
and the preceding a1.ikm.1
The full implications of that discussion require extensive
comparative analysis. The third type of jihd was in
earlier works referred to as muliraba and envisaged a
situation wherein a person going about his peaceful business
whether in dr al-1arb or dr al-slm is attacked and
required to defend himself. 2 That this was not jihd was
1. Shahld II, Raw.a, 86-89.
in Nih.ya, 291; cf. also Muhaqqiq,
2. T 1 refers to dif
Shar ! i c , I, 309.

150

generally accepted. The first type of Jihd, for expansion

and conversion, was universally held to be sgi by all
ImmI fugah, and is clearly so considered here by Shahid II.
The nib kh
was a person appointed by the
imm for Jihd or something more general than Jihd: the
term n'ib kh
refers to a contemporary of the im gm or to
the sufar'. If that be doubted it should be noted that the
term was prefigured barely a generation earlier by Karaki,
Shaykh al-Islam to Tahrnasp, who had noted simply:
jihad is incumbent only on condition of the
imam or his nib; the meaning of na'ib here
Is the na'ib specifically designated by the
Imm during his presence and with his ability
to act [freely]; it does not mean the na'ib
absolutely.3

&f,'j' (

)1U

C..JL

'

Defensive jihad (type

c' 1-'

c..s

defined by Shahid II
in terms of Muslims and bayat al-islam. He specifically
ii) is

denies that there might be fear for Islam from a Muslim: the
enemy envisaged Is a kafir, to be distinguished from a mubdic.
There was no serious threat to Islam at this period from
kafirs: the Ottomans in fact were expanding dar al-islam
vigorously in Europe. Shahid II's remarks there were perhaps
necessary but largely academic. On the baghi (rebel) Shahid
I had written:

Shahld II's commentary was as follows:

LL

LJLP

Shahld I's expression was open to interpretation. ShahTd II

chose the narrowest possible interpretation and relegated the
3. Karakl, Jami c , 187.

151

struggle against bught quite firmly to the period of the

imm.

He did not even add any reference to the n'ib. It

was the imm alone who could organise the fight against
bughit.

More than five centuries earlier

the same view. 4

Shahid

1 had indicated
II's subsequent note that fighting

a bghl is like fighting kuffr might, by a determined

exegete, be taken as containing a potential reference to the
Safavid Ottoman struggles, but that would be very much in
opposition to Shahid II's remark that defensive jihd
referred only to kfirs as such and uslims do not constitute
a threat to Islam even If they are mubdi C . Later fugaha'
elaborated and acted upon the rules relative to defensive
jihd; none of them seem to have provided for or sanctioned
a struggle between Muslim sects or states.5
Shahid II in spite of the acknowledged anti_Shidi
activity which existed during his lifetime in Ottoman territory6
did not provide legal shar'i sanction for Tahmsp's wars.
In a spirit of what would appear to be largely theoretical
speculation he left open a role for the faqlh in the sphere
of defensive jihd but made no provision for fighting between
Muslims. While it Is true that Shl'l Traditional literature
provided a great mass of detail on 'All's struggle against
his Muslim opposition, the struggle against the bught during
the Ghayba does not appear to have been a prime concern of
the Shi c I fuqah', not at least in the context of fiqh writing.7
The Immi fuqah' were, prior to Safavid times, not in a
position to advocate military struggle against the numerca1ly
4. Tsl, Nihya, 296-297.
5. See Larnbton, Jihd; Kohlberg, Jihd; Hairi, Shiism, 88-90,
ll6-l9. The 'ulam' of Iraq and Iran in the early 20th
century only declared jihd against non-Muslim invaders!
oppressors. They constantly asserted pan-Islamic unity
even to the extent of addressing the Ottoman sultan as caliph
or amlr al-mu'minln; Hairi, ShIism, 88-89, 121.
6. See Eberhard, Osmanische Polemik.
7. Pace Kohlberg, Jihd, 69.

152

vastly superior arid militarily immensely stronger Sunnl

'rebels'. MwardI and other Sunnl writers were much more
free to stress the rulers' right to fight ahi al-baghy.8
The establishment of Safavid power might have occasioned a
change in ShIi law but there is no evidence that it did so.
This might partly be a result of the unwillingness of the
clerical class to interfere in military affairs, as witness
their constant refusal to sanction offensive jihd; but
might also be a result of real Inability beyond certain
limits to engineer change In juridical rules. TsI's earlier
definition of the struggle against bught had specified the
activity of the imgm and had not provided for the activity
of an appointee. 9 Later fugah' seem to have accepted the
implied sug of that particular type of jihd.

TSI'S

assessment of jihd in the early 11th

century had already indicated that, for conversion, it was

sgi, arid that fighting for al-imm al-j'ir was a sin.
JIhd depended upon the presence of the Imm or his appointee.
For TsI both the Imm and his appointee were absent. The
identification of T1' 'appointee' is not clear but may
reasonably be inferred from his Kitb al-Gha y ba. That work
has a section in whichare discussed the various representatives
of the imm) These were of two kinds: there were the
wakils appointed by the imm for various purposes while he
was still present, and there were the sufar' appointed
during the period of the lesser Ghayba. These were I believe
the people

TsI meant

when he referred In his various figh

works to man naabahu al-imm. For Ibri Idrs likewise,

unambiguously, the nuwwb al-imm were not extant during the
period of the greater Ghayba. 11 Mu1aqqiq introduced the
term niyba with clear reference to the fuqah' in the context
of khums, and related that niyba to the judicial function of
the fugah'. In other spheres, however, notably that of jIhd
8. cf. Ibid.
9. TsI, Nihya, 296-297.
l0.Id., Gha y ba, 209ff.
ll.Below, 161-162.

153

the term indicating appointee, however phrased, is best

understood as a reference to an absent entity. The same
is true of CAllama
Both of these stated that jihd was
dependent on the imam or his appointee. Both seem to have
considered jihd, (for conversion), sqi) 2 The various
terms indicating appointee of the 1mm in ShI'i legal works
up to and including those of CAl1ma are best understood as
referring to contemporaries of the imrn or to the sufar'.
Where these early writers wanted to specify a function for
the fugah', as for example In the spheres of khums, zakt,
Friday prayer, they explained that function; they never
assumed that the term or phrase indicating appointee would
be simply understood to include the faglh.
When Shahld I used the locution a1-imm aw n'ibuhu
with reference to ,jihd, he seems to have meant in conformity
with tradition that offensive jihd was sqI. He still
understood by n'ib someone actually appointed by the Imm,
someone not present during the greater Ghayba. When he
wished to specify the fagTh he did so explicitly, either
avoiding the formula 'the imm or his appointee', 13 or by
glossing It: e.g. ad a1t al-jumta:
Friday prayer is not performed except with
the imm or his n'ib, even the faqlh,
(wa-law kna faqihan) ... during the Ghayba)
Or ad khums:
It Is paid to him while present or to his
nuwwb while absent)5
12. Mu1aqqIq, Shar?ic, I, 307; Mukhtasar 109. Cf. Lambtori,
182-183; Kohlberg, Jihd, 80-81; (the evidence
is rather against the supposition that the fuqah' gave
themselves authority over jihd in this period.)
See also CAllma, QawCid, 38.
13. E.g. ad Iudd and ga', Shahid I, in awq.a, 90.
14. Ibid., 34.
15. Ibid., 53.

1524

In general arid unless otherwise specified the locution

'the imrn and his appointee' is a reference to the period
of the imm's presence or to the lesser Ghayba. The
possibilities of ambiguity were however immense and there
is every reason to believe that some at least of the later
fugah
(cAllma, Shand I) took a perverse delight in
rendering their legal thought with all the rich obscurity
and generous potential that they found in revelation.16
Deliberate ambiguity was a part of the Islamic legal tradition.
Much later, N'nI's attempt to relate his view of constitutional government to the acknowledged rights of the fuqah'
resists all efforts at precise analysis. 17 There were
elements of the Iranian constitution which, following the
same tradition, deliberately exploited ambiguous phraseology
in order to undermine the traditional judicial authority of
the clerical class. 18

Early Imm Traditional literature, as presented

for example in the KfI of Ku1ayn, expresses a distinctly
negative attitude to military jihd.

The spiritual struggle

was presented as the higher form of jihd) 9 Most Traditions

portray the imms as disapproving or denying the duty of
military jihd during the period when they lack real power.
Such Traditions as might appear to condone jihd are placed
under tendentious chapter headings which pre-empt interpretation,
- a technique familiar from Surinl Traditional works. A
16. Cf. Wansbrough's comments on deutungsbedrftigkeit,
Quranic Studies ,, 99-102.
17. Hairi, Shl t ism, 193-1924.
18. Ibid., 215-216.
19. See also Lambton, Jihd 182; Kohlberg, Jihd, 66.

155

particular Tradition for example in which Ja t far a1-Sdiq

discusses certain aspects of jihd without condemning it
is placed byKulayil under the heading Bb al-ghazw maca
l-ns idh ukhlfa Cal l-islm. 2 The idea of fear for
Islam was not however part of the Tradition: the rubric
betrays Kula3nI's desire to interpret and classify the
Tradition in such a way as to harmonise it with the otherwise
accepted belief that jihd was dependent on fear for Islam.
The only Tradition concerning murbia cited byKu1ayn was
placed under the same heading, thereby on the whole reflecting
a negative attitude to that activity. The implication would
appear to be that murbia too depended on fear for Islam.
T 1 in the Tandhb produced two further Traditions more
distinctly opposed to murbita and set all three under the
heading BAb al-murbita.21
The translation of Traditions into furti' insI's
Nihya confirmed the invalidity of offensive Jihd. Defensive
jihd was specifically declared permissible during the Ghayba
if there was fear for bay.at al-islm, hawzat a1-is1m or
mu'minTn. The intention should not be jihd with an unjust
imm (al-imm a1.jtir) but simply defence. Murbita during
the Ghayba according to TsI technically did not have the
value It had while the imm was present:
(

4)

1,31.6 L

However under certain conditions, after vows, it became

permissible either to perform or provide for murbita.22
Ibn Idrls followed T1 closely on the matter of Jihd.
Offensive jIhd was not valid during the Ghayba. Defensive
JIhd was. He argued more strongly than T 1 in favour of
mur g bita and suggested that all vows associated with that
should be kept. 23 Muhaqqiq took up a similar position. He
20. Killayril, Furt' a1-Kf, I, 332.
21.
Tandhib, II, L2J43.
22. Id., Nihya, 290-291.
23. Ibn Idrls, Sar'ir, 156.

156

declared murbita mustaliabb even when the imm was not

present, which was distinctly different from Tsi t s view.
His assessment of defensive jihd does not include reference
to mu'minin, but only to bay.at ai_isim.2 Such development
as took place from TsT to Muliaqqiq was partly undoubtedly
a result of the controversial and idealistic uli views of
Ibn Idris but also reflects political developments. The
Islamic world of Tisi's time was under no threat from nonMuslims; it was actively expanding under the Ghaznavids in
India and by peaceful conversion in Transoxanla. By the time
Muhaqqiq wrote) the Mongols had emerged as a real threat to
bayat al-1slm. It is true that the Immi community in
southern Iraq actually benefitted from the tolerant attitude
of early Mongol governments, and that Muliaqqiq himself enjoyed
the friendship and respect of Nair al-Din Tsi; but clearly
the sense of potenti
threat to the Islamic world would be
more present in the thirteenth than in the eleventh century.
That would account for Mulaqqiq's greater stress on the
desirability of murbia.
At the same time a reduction in
tension between Sunni and Shi d i communities, and the sense
of communal interests produced by the Mongol onslaught were
conducive to the removal of the idea of defence of rnu'minin
from the discussion of defensive jihd. It was not thereafter
to reappear. In discussion of defensive jihd the Shi'is
seem consistently to have transcended their narrow interest
In dar ai-rmn and to have asserted their common cause with
dir. a1-islm.
For Al1ma and Shahid I offensive jihd continued
sgi1. Defensive jihd was defined as relevant to an attack
by kuffr on dr al-islam (cAllama) or fear for bayat alislam (Shahld I). Murabia was increasingly a duty deserving
considerable reward. The struggle against bughat though
possibly ambiguous because of the potential identification
of bughat with kuffar is nonetheless best understood as,
like offensive jihad,saqit. 25 Karaki's Jamic a1-Magaid,
24. Muhaqqiq, Shara'i' I, 309; cC. Mukhtaar, 109.
25. See 'Allama, Qawa'id, 38-39; Shahld I, in Rawda, 86-89.

157

a commentary on t Allrna's QawCid and his most famous work

of figh, though it contained in the exordium high praise
for the new Safavid dynasty which upheld the Shi c i religion,
provided in the sphere of jihd nothing that could be interpreted as favourable to or justificatory of the Safavid
struggle with the Ottomans. Offensive jihd was related
firmly and unambiguously to the period of the imm's presence.
No comment was added to Allma's ruling that if kuffr fall
upon dr al-is1rn it becomes incumbent to fight them; rio
attempt, in other words, was made to stretch that terminology
so that it might refer to the Ottomans. Bught it is true
were defined as kuffr but the struggle against them depended
on the presence of the imm or his ri'ib. If Karak had
meant by nib either n'ib 'mm (faglh) or something else
(a Safavid ruler?) he might have found some way of indicating
his view. In fact he left the term n'ib in that context
unqualified. Generally, I have suggested, that term unless
otherwise qualified designates a sqi entity. Further, in
the field of jihd, Karaki had earlier specified that the
n'ib in question was the one manss bi-khussihi ll uhr
al_imm.26 A positive assertion of a role for the faqh in
some aspects of jihd did not come until the time of Shahid II.
Karaki certainly and Shahid II probably were concerned to
support the Safavid dynasty; It is not possible however to
suggest that either of them provided any degree of shar"I
legitimacy for their wars, nor indeed strictly for their
bureaucracy. Shahid II did acknowledge a possible role for
the faqlh in defensive jlhd - against non-Muslims. While
discussing zakt he specified that during the Ghayba Jihd
and ta'lIf might be valid bi'l-fagh wa- ghayrihi. 27 The
indefinite ghayrihi might seem replete with exegetical promise
but it probably meant nothing more than that If Islamic
territory was attacked by non-Muslims anyone who witnessed
or suffered that attack should immediately take steps to
defend Islam, even without waiting on the authority of the
ideal organiser, the faqih. That the faqlh was the rightful
26. Karakl, Jmi c , 186-188.
27. Above, 111

158

authority in charge of defensive jihd is clear from Shahld

II's discussion; no writer prior to him had expressed
clearly that idea. That juridical innovation was prompted
on the whole less by the pressure of real events than by
the sense of juridical economy and elegance manifest through-.
out Shahid II's Rawat al-Bahiyya.

Recourse to the pivotal concepts of n'ib khss

has so far in the present work been demonstrated
and n'ib
only in the writings of Shahid II. Some attempt has been
made to assess what these terms implied.28 For a fuller
understanding of that development however It is necessary
to note that they were prefigured in the juridical works of
Karaki. It is well-known concerning the latter that he held
a position of considerable power at the court of Tahmsp.
He is presented in biographical works as being particularly
involved in the propagation of the ImmI faith and in taxcollecting activities. Indeed he wrote a controversial work
on kharj, entitled Qatat a1-Lajj fT Ia11 a1-Kharj. That
work remains at present unstudied but would not appear to
contain any strict legitimation of official tax-collecting
activities. Hairi says of it that in it "unlike many other
culam
[Karaki] declared it legal to spend money which was
gained through taxation by an oppressive ruler." 29 The ruler
was still oppressive. It seems probable that in it he
discussed the problem of how to resolve the tension between
the actual political entity and the ideal shard community.30
In biographical works Tahms is presented as having said
to Karaki: "You have more right to mulk than I because you
are n'ib to the imm; let me be only one of your 'umml to
28. Above, 83,125-126, 1 28 -130, 13a-139, 142-143, 147-150, 152-154.
29. Hairl, ShI'ism, 61-62; (my italics).
30. Cf. above, 85-97.

159

carry out whatever you command or prohibit." The same works

also provide a letter in Persian from Tahmsp in which he
urges obedience to the command (Iu1an?) of the mutahids who
are guardians of the law and in particular to the seal of the
rnujtahids, the inheritor of the knowledge of the prophet, the
nib of the sinless imrris etc., Karaki. In his letter
Tahmsp provides as 'source' for his views the maqb1 Tradition
of C umar ibn Hanzala.31
Now, that Karakl was Shaykh al-Islam, a man of power
and authority at the court of Tahmasp, involved in the
propagation of the ShI d I faith and also in the assessment of
taxes, there is no reason to doubt. But that Tahms wrote
that Persian document or considered himself tmj1 to Karaki
seems less likely. The Persian document takes the form of an

ulI

argument in so far as it sets forth a ruling and

justifies it with reference to a Tradition. While Karaki

in order to support his argument about Friday prayer did infer
from the magbl of C tjmar ibn Ianzala a 'general' delegation
he did not apply that concept to any other spheres, not at
least in the Jamic a1-Magid.
The perception that that
Tradition, together with the concept of general delegation,
was applicable to khums, zakt, jihd etc, that it implied a
general necessity of obedience to mujtahids, belongs to a
slightly later generation than Karakl.

It is possible then
to suggest that the idea embodied in Tabmsp's purported letter
are too well-organised to belong to Karakl's generation. In
any case the ulI format, the citation of a Tradition to
support a 'juridical' view, suggests not an official courtly
source but a clerical source for the letter. Likewise the
assertion that a king might be tmil to a muJtahid is .an idea
which was current in Iran while Chardin was there, 32 but it
was not expressed by either Karakl or Shahld II. It is
probable then on the whole that these and possibly other
elements in the biography of Karaki were invented at a later
date. Karakl became the protagonist of these inventions because
31. KhwnsrI, Rawdt, 1402406.
32. Chardin, Voyages, V, 216.

160

he had had a very real and considerable degree of power.

The biographical details of his life provided a precedent
Indicating what authority the clerical class ought to have,
and associated their supremacy with the 'great' days of
the early Safavid period.33

A full understanding of how the concept and

terminology of clerical nIyba developed requires a brief
survey of juridical dispute as it emerged around the problem
of Friday prayer. In the Buyid period it was unclear whether
Friday prayer was or was not permissible during the Ghayba.
in the NIhya specified for Friday prayer the presence
of the Imm 'dI1 or man naabahu ai-imm. 3 In the chapter

T'

on al-amr w'al-nahy he qualified that by stating that It Is

permissible for the fuq.h' to gather the people for all
prayers including the Friday prayer and the two cIds.35
teacher
That permission however was not specified by

T1'

Murtad

6 who was portrayed later as a major proponent of the

anti-Friday prayer view. Ibri al-Barrj, a pupil of

commenting on Murtad's Jumal, equally made no provision for
the organisatlon of Friday prayer by the fuqah' during the
became famous for
Ghayba. 37 Salr, another pupil of
his explicit comment that: 'During the Ghayba the fuqah may
pray the two G rds and the Istisq' but as to the jumca, they
may not do

The Buyid fuqah' then, even those who were

came to no
closely associated with one another and with
definite conclusion about the permissibility of Friday prayer
during the Ghayba. It is probable that in the following period
33. For further details of Karakl, see Glassen, Schh IsmT1,
265-268. For the post of Shaykh al-Islam, see Lambton,
Quis Custodet, II, l39l142.
324 . TsI, IIhya, 103.
35.
36.
37.
38.

Ibid., 302.
E.g. in I'Iurta, Jurnal, 121-122.
Ibid., 123-125.
Cit., Ibn Idrs, Sar.'ir, 161; see also Sa1r, on Friday
prayer in KhwnsrT, al-Jawmi' al-Fiqhiyya.

161

view prevailed and the Friday prayer was actually

performed when circumstances and governments permitted in
Shi t i mosques. Ibn Idris, true to form, disliked TsI's
view on the grounds that he could not find, and

T had
not produced, any justification for it. He quoted a passage
from TSI'S Khilf in which at some length T1 insisted that
Friday prayer depended on the presence of the imm or one
appointed by him for that purpose.

1 claimed ijm' alfirga on that matter and denied the existence of any dalil
which contradicted that view. Then in reply to a rhetorical
question which he asked himself

declared that It was

in fact permissible and desirable for villagers, countrydwellers and believers (li-ahi a1-gary w'al-sawd w'almu'mlnln) to pray the jumca, "as if the imm had appointed
someone to pray with them." ( L
Ibn Idris described that as an astonishing
reply and proposed as the correct alternative that Friday
prayer was only permissible If there were amongst the people
nuwwb al-imm or nuwwb khu1af'ihi.

TsI's view

according
to him lacked the necessary dalil, and was indeed selfcontradictory. Furthermore by application of the juridical
'proof' of bar'at al-dhimma Friday prayer could be shown
to be invalid; for, while It was at best doubtful about
the validity of Friday prayer It was certain that during the
Ghayba It was permissible to substitute for it the
(four
rakcas).
The uhr is four r.ktas and Is as to the
dhirnma certain. Those who claim that two
rak'as (= Friday prayer) dispense with the
need for four rakcas require a dalli: for
we do not turn from what is known on the
basis of what is merely a matter of opinion,
nor on the basis of akhbr which necessitate
neither knowledge nor action.
Ibn Idrls then quoted Nurtad and Salr as significant fuqah'
who had denied the validity of Friday prayer during the

162

Ghayba. That for him was the only correct view.39

Mulaqqiq on balance supported

TsI

and opposed Ibn

Idris.
If the imm is not present nor the one
appointed f or prayer, but ijtimC
and the
two khubas are possible, some say it is
mustaliabb, and some say it is not permissible.
14Q
The first view is better, (ahar).
That, it should be rioted, actually concedes that the one
appointed for prayer Is not present; In spite of this
Friday prayer might validly be carried out. For CA11ma:
It Is permissible for the fugah of the
ahi al-Iiagg to gather the people for prayers,
all of them, the five incumbent, and the CIdS;
this is mustaiabb and definite as long as
there Is no fear. As to the Jum'a our Culam?
are at variance on this. Some have declared
It permissible. Salr and Ibn Idris have denied
it, and this is strong, (wa-huwa gawI).
He In other words favoured Ibri Idris. Shahid I expressed his
opinion as follows:
t4 - ,

ti,.s,

cLL
L fJ

(J'>

L)'

Friday prayer Is not performed except with

[the presence of] the imm or his n'Ib, even
if it is the faqlh, as long as ijtim' is
possible during the Ghayba.2
This differed from for example Muhaqqi q 's view in that It
recognised the fagih as a kind of n'ib in this context to
the Imm.
In his Dhikr Shand I acknowledged that permission
from the imm was required for Friday prayer and then asserted
that it was in fact available. To support that view he
claimed that
had so stated In his Khiif, that the
Tradition of Zurra (in which Jacfar al-diq urged his
39. Ibn Idrls, Sar'ir, 63- . 65; cf. 161.
O. Muhaqqiq, SharhIc, I, 98; cf. Mukhtaar, 35-36.
141. CAllma, TaITlrIr, 158; of. Qaw'id, 14.
142. Shahld, I, in Rawda, 34.

163

followers to Friday prayer) indicated permission and finally

that the fugah' during the Ghayba dealt with matters greater
than Friday prayer such as bukm and ift': so much the
more was permission available for alt. He described Ibm
Idris, Salr and 1urtac as extremists (blagha): they had
interpreted the Tradition (of Zurra) as relevant only to
the time of the imm's presence and considered the permission
attested for lulan and ift' as irrelevant to alt. ShahTd I
thus expressed himself as perceiving in the judicial delegation
an even wider significance than had previously been claimed
for
Karaki took up the problem. The opinion that
Friday prayer was forbidden during the Ghayba was according
to him based on the writings of various fugah'. Amongst
these was T 1 in the Khilf. (It Is interesting to note
that Karakl interpreted TSI'S gnomic furt differently from
Shahid I.) These fuqah' had denied that permission was
available from the imm and had relied on arguments based on
the concept of bar'at al-dhimma. In a complex argument
Karakl refuted these claims and asserted In favour of his
own view - that Friday prayer might and ought to be held
during the Ghayba - the following arguments: a. Q62.9;
b. Traditions, Including the aiI1i of Zurra; c. istIs1b
al-hal; i.e. what Is incumbent during the time of the imm
continues incumbent afterwards. The Traditions he cited
were open to the objection that they referred only to the
time of the irnm's presence. To this objection he replied
as follows:

'

o5Ltr
L' j'

(2.4 -

J' -J

L-4 cJ?
L,L ()

d'

-?L
3. Shahld I, Dhikr, n.p., ad alt al_jumca

cJ

__

l6L

Permission from the im gm or compulsion by

him directed to the people of his own age
is not limited only to those people; for
their command [the imms', 1ukrriahum] to one
is a command to all. There is no need that
the imm be required [specifically] to
appoint a n'ib in advance because in that
case the n'ib would only be

But the
general [appointee] is not dependent on
mukallafs being [specifically] appointed for
they, the irnms, have appointed a n'ib cal
wajh al-umim based on the words of Ja c far aldiq in the maqbl of 'tJrnar ibn IIanala: I
have made him a Judge over you etc.
The opacity of that passage renders translation difficult
but that provided conveys the substance of Karaki's thought.
It is an important passage in so far as it is the earliest
indication by an immI writer that the maqbl of Umar ibn
kian z ala could be made the basis for a distinction between
n'ib cg
and n'ib khs. It is true that the 'general'
implications of that Tradition had begun to be perceived as
early as fluIaqqiq but a distinct terminological reflection
of that perception had not previously emerged. It is indeed
unclear to what extent Karaki himself understood the 'general'
delegation as applicable to spheres other than those of gad'
and Friday prayer. Such was the discrete nature of any work
of figh that, though he had devised and applied this new
concept to the problem of Friday prayer, Karaki thereafter
in discussing khums, zakt, jihd and liudd did not resort
at all to the term n'ib Cmm He seems to have been conscious
of the dangers of the broad interpretation of the concept
niyba; so much so that when he came to the problem of
(offensive) Jihd he had to specify that the n'ib in
question was mans bi-khusihi Il zuhr al-imm: he did
LLj

Karakl, Jmic, 130-131.

165

not desire to burden the fuqah' with responsibility for

military action.
Shahld II had at his disposal the concepts of
n'ib Cjfl and n'ib khss:
The Friday prayer is not validly held except
with the imm Cdjl or his n'ib, khuan or
Cumurnafl
This is the case even if the ri'ib is a faqlh
possessed of the qualities of ift
during
the Ghayba because he is manub mm al-imam
Cuman on the basis of Jacfar al-Sdiq's
words: Unur i1 rajul etc. (i.e. the maqbl
-
45
f Cumar ibn hanzala)
He stressed that while the imm was present Friday prayer
depended on him or his n'ib khs: that concept signified
an appointee of the imm during the time of his presence
(and might also signify the four sufar').46

The value of clerical biographies as record of fact

may, as I have indicated above, be doubted. Concerning
Shahld II however it may be noted that he emerges as a
dedicated scholar who studied not only with ImmI teachers
but also during his travels to Cairo and the Hijz with
Sunni teachers, of, it is said, the Hanball, MlikI and
Shfi C schools. He was concerned with political affairs
at least to the extent of a trip to Constantinople to

procure an official post within the Ottoman hierarchy. The

5. Shahld II, Rawda, 34.
46. It may be noted here that the alternative definition of
nib khs provided by Lambton (Jihd, 181, footnote) is
undoubtedly a later development. The use of the term r'ib
khs in Kohlberg's Ris g ia Jihdiyya (Jihd, 82-86) is
certainly to be understood as a reference to the preGhayba period, or to the sufar'. See above, p.158, n.28,
for a list of the most important references to nib
mm and
kh
in this thesis.

166

QI al-'Askar K. ibn Qutb al-DIn procured for him a post

in the Madrasa Nriyya in Ba'albak. The sultan Sulayrnn
issued the necessary documents and provided for him a stipend.
Shahid II is depicted in Ba talbak as being muftl kull firga
bi-rn yuwfig madhhabahu and as popular and respected by
all parties. The size of the Shi d i community in Syria makes
it at least possible that the Ottoman government turned a
blind eye to the continued teaching and partial implementation
of the ShI'I code.
In the event however Shahid II was
eventually denounced as a herebic, a mubdic who Iiad transgressed the rules of the four schools, and he was, not it is
said at the instigation of the sultan, but as the result
of an error, executed. His biography is interesting but
his greatest achievements were undoubtedly academic: that
too is freely recognised in the biographical literature - and
indeed in the syllabus of modern madrasas in Qum. 8 He is
accredited with a significant stylistic innovation in so far
as in his commentaries he mixed the matn and the sharh in
such a way as to produce continuous prose. That he apparently
adopted from the SunnIs.
The r'Iaslik al-Ifhm on Mu1aqqiq's
Shar t iC is an early version of the style, but rough and
uncertain. The Raw.at al-Bahiyya on the other hand with its
fusion of sharl2 and matn, and its combination of ul, furti'
and polemic is an impressively written and well-ordered work
which, while sensitive to the developments of previous
centuries, transcended a great deal of the muddle and
obscurity of earlier writers. One of the most significant
7. It would appear, admittedly for a later century, that leading
ShI'I Jurists might be recognised by the Ottoman government
as gis for members of their own community; Gibb and
Boweri, Islamic Society, I, 2, 123.
48. The Raw.at al-Bahiyya is the earliest work of flgh that is
still studied as part of the regular syllabus in Qum.
49. Nawawl/Shirblni, Nughnr a1-Mu1tj is a Sunnl example.

167

signs of his concern for juridical economy and elegance

was the way in which he utilised the concept of n'ib 'mm
to account for all the sharC functions of the fag3h. That
was a major artistic resolution to an essentially intellectual,
conceptual problem. It may be taken to represent the fact
that amongst the various causes of juridical development was
the resolution of logical and systematic problems that
emerged, unrelated to external events, in the elaboration of
the sharlca.
The term nib Cjpj became a pivotal concept in
Shahld II's expression of the sharl C a.

For him all the

executive functions of the imm, which had to a great extent

in the early Buyid period been declared sgi, devolved upon
the fagTh as general representative. The only significant
exception was offensive jihd which remained sqi.

The
initial contraction of the sharl'a and its separation from
the political entfty had been an essential preliminary to its
re-formulation as potentially capable of fulfillment under
the executive control of the fugah'. The process whereby
that re-formulation, that filling-out, was achieved was one
of gradual exegetical and termiriological innovation. The
single most important development was the re-interpretation
of a Tradition, originally perceived as relevant only to
judicial authority, as in fact relevant 'generally' to shard
activities. The resultant definition of the sacred (sharI)
community was markedly different from its Sunril counterpart.
The sacred community (dir al-imn) was composed of believers
whose acts were rendered legitimate by their submission to
and acceptance of the authority of the fugah. That
authority extended over all acts and activities which were
prescribed, proscribed or otherwise commended in the sharlca.
Largely co-extensive with, but distinct from, the sacred
community was the profane community, that is the world of
political acts, structures, hierarchies and duties which were
irrelevant or inimical to shar'I laws. Most ImmIs lived
thus at the same time in two moral systems. The Sunnls had
avoided the disjuncture by at all times recognising the

168

political entity as part of and essential to the sacred.

The tension inherent in the ImmI world-view was resolved
either by ensuring that political acts were brought into
alignment with legal requirements or by recourse to taqiyya.
The most significant aspects of that resolution have already
been discussed. 5 It may be noted however that while there
was general clerical agreement that something had to be done
to mitigate the disjuncture of the sharcT and the actual
there was in moments of crisis no agreement as to the mode
of mitigation. A certain polarity may be distinguished
between those who advocated an aloof distance, a dedicated
pursuance of shard affairs in despite of the constant propinquity of profane events and the compromises they elicit;
and those who urged an active effort to eliminate the profane,
to so control and constrain the actual that it becomes the
shard.
Those of the latter type are those who would set
up a state ruled by the mujtahids.
That there was in juridical expression of the
shari c a through the centuries a distinct pattern of change
and developments is not surprising; it remains true that
the fuqah presented the law as essentially immutable,
each new work of fur being for them a re-expression of the
one system ordained by God and reason, and mediated to man
through the Prophet and the imms.
The existence of permissible variation within the system was acknowledged but did not
prevent polemic effort to ensure that one view rather than
another shouldprevail. The basic structure of the law was
much older than the ImmI expression, older perhaps than Islam
itself; its elaboration reflects the general human need for
justification and legitimacy. That need may be thought cause
of the production of the delegation Traditions attested in
Kulayni. They provided for the existence,during the Ghayba
of a class of magistrates whose function was to resolve
disputes within the community (dayn, mTrth) and to provide
sanction and legitimacy for such social activities as
marriage, divorce etc. The authority which the clerical class
50. Above, Chap. IV, 85-97.

169

thus acquired within the community (or which was thus

sanctioned) tended to grow with the passing of time. It
has been suggested in the preceding chapters that juridical
developments may best be understood as response to and
justification of social and political change. However the
type of change that most found expression in legal works
was only such as was favourable to and consolidatory of
clerical authority. To a lesser extent juridical works
expressed the aspirations as well as. the achievements of
the clerical class. That was particularly evident in the
context of defensive jihd: it is unlikely that Shand II's
postulation of a role for the faqh there was actually a
result of their activities in that direction. The negative
aspirations or practical impotence of the fuqah' also
affected legal development. Formally, if the term 'appointee
in the context of khums, zakt and Friday prayer could be
resolved into amm and kha, there was no reason why the
same term in the context of offensive jihd should not be so
resolved. The failure of the fuqah' to assert a role for
themselves there reflects their reluctance or inability to
participate in that activity. The general nature of juridical
development as discussed in the present work betrays the
narrow vision and self-interest of the clerical class. They
were, perhaps inevitably, prisoners of their own ideology
and on the whole failed to distinguish between the interests
of their class and those of the community. There might be
exceptions to that, perhaps for example in the context of
defensive jihd; there, they seem to have transcended the
narrow vision and asserted their identity and common cause
with the whole Islamic community. That too however might
ultimately be related to their own self-interest. A final
factor in the development of juridical theory was found in
the exigencies of purely formal, logical and systematic
expression, and juridIcal elegance. There is a great deal of
conscious artistry evident in the works of especially the
later writers, Allma, Shand I and Shah3d II; that is a
quality painfully absent however from, for example, the Mabs

170

1 and the Jmic a1-Magid of KarakT. The most

positive effect of the clerical elaboration was probably
of T

this, that it provided for the whole community a set of

values and a principle of organisatiori through which life
and society gained a greater degree of significance and
meaning, was rendered In fact shard, real, sacred. The
legal speculation of the fugah' set forth, however narrowly,
a social ideal and a concept of justice which mitigated the
rigours and hardships of daily life. There is a certain
truth In the perception that men on the whole do not like
freedom or uncertainty or shifting values: what the fuqah'
provIded was a system which both ordered and sanctified daily
activities and promoted thereby both social and individual
stability.

PART III

171
CHAPTER VII
EXEGESIS AND CONSENSUS
The elaboration of the shari'a was more than the
imposition of order and structure upon the contingent
events of daily life, it was a manifestation of divine
interest in those events. The detailed shard prescriptions
were thought to embody the practical part of God's commission (taklif) and represented His chosen mode of worship (tac_
abbud). Adherence to them rendered sacred otherwise inconsequential acts and promised personal fulfillment and social
Justice as well as eternal salvation. The transmission of
God's command to His creation was achieved by the sending
of a prophet whose life and deeds confirmed or revealed the
law. The concept of law incarnate was not restricted to
Islam nor even to Semitic religious thought: the ascription
of (idealised) contemporary behavioural patterns to a mythical ancestor or divine or semi-divine founder is a widely
attested phenomenon of traditional societies. 1 Non-Islamic
(Jhi1i) Arab tribal groups may have assessed their traditional patterns of behaviour (sunna) as those of their forefathers, originating with an eponymous ancestor or other
hero (imm). In religious literature the term sunna was
related to a prophet and referred to religious law. 2 The
term din in such literature covered a semantic range similar
to that of the term sunna: it too signified a pattern of
behaviour or a body of prescription attributed to or preserved by pious ancestors. It is frequently found in
cinjunction with terms indicating forefathers, folk etc.
din b'ika, dIn g b'i, din qawmihi. 3 Just as God or a
prophet were thought to be the sources of sunna, so too,
in the SIra, God or Abraham were identified as the sources
of din. 4 Din in early Islamic legal literature continued
to signify prescription/law rather than "religion". 5 The
1. See Eliade, Myth, Ch.I.
2. See above, 1-2.
235, 265, 2144.
3. Ibn Hishm, SIra i
14. Ibid., 244, 250, 265.
5. See above, 5, 14-15.

172
manner in which the Muslim community of the late eighth
century depicted details of prescription or doctrine as
elements of prophetic biography or communal history has
been amply demonstrated by Schacht and Wansbrough. 6 The
aim was within the parameters of sectarian (Judaeo-Christian)
themes to provide for the Muslim community a normative
history which would illustrate the intervention of the
divine and the life of the prophet in such a way as to
provide sanction for contemporary events. The production
of a canonical scripture, the Qur'n, was, according to
Wansbrough, primarily a result of the need for tangible
evidence of Mul?ammad's existence and claim to prophethood.
It served thereby to support the concept of prophetic law
and confirmed the juridical value of prophetic biography
ftadith) while becoming itself a potential source of
prescription.7
The claim that normative codes, frequently of only
local validity, were a result of prophetic practice was a
part of the justificatory polemic of the early schools of
law. The idea that the relationship between the legal code
and prophetic biography should be straitly manifest or
logically demonstrated found persuasive and sustained
expression in the Treatises and Risla of Shf1 C : law for
him must be explicitly based on events in the prophet's
life. The force of his and similar arguments prompted not
only the canonisatlon of the Qur'n, as suggested by Wansbrough, but also the Tradition-collecting ardour of the
ninth century. The argument for the juridical importance
of Qur'n and Traditions (both designated revelation8)
precedes and accounts for their canonisation. The success
of ShficPs polemic may be linked to the symbolic value of
scripture and Traditions: they represented in a permanent
form and in a forceful way the confrontation between prophet
6. Schacht, Origins, esp. part 1; Wansbrough, Milieu, Chap. 1.
7. See Warisbrough, Quranic Studies, L3.52, 77-78; Milieu,
57-58.
8. Id., Studies, 176.

173
and divinity which had led to the foundation of the Islamic
community. The contents of those revealed texts took the
form of history or biography and served for as long as the
basic elements of life and revelation were roughly similar
as a model and mirror for Islamic society. The achievement
of ShfidI was to lay the foundations of a methodology
whereby the finite (and by the end of the ninth century)
fixed content of revelation could be rendered flexible and
if not infinite then open-ended.
tJ1 al-figh was a methodology whereby the fugah'
related revelation to rescript ion. The temporal priority of
the latter is nearly always manifest. That the content of
revelation compelled the community to adopt - as it were
unwillingly - this or that stance seems unlikely. Schacht
it is true suspected that Shfi'i in so far as he was bound
by the rules he formulated "cut himself off from the natural
and continuous development of doctrine in the ancient schools."9
It is nonetheless the case that Shfi'T's system did not
Involve any revolutionary changes in the structure of
traditional codes: his law was recognisably the same in form
and content as that of M1ik and the Iraqis while differing
certainly In details. Islamic schools of law In general
both before and after ShfI'I seem to typify that desire for
variations on a theme which Levi-Strauss discussed with
regard to certain Australian and French communities. These
communities, he thought, reflected "the emergence of a sort
of common philosophical and sociological style along with
methodically studied variations on it." Distinct but
related social groups were concerned with "constantly
elaborating themes only the general outlines of which were
fixed by tradition and custom." 1 Islamic legal systems
likewise exhibit a common structure together with an elaborate and fussy set of variations. That seems to have been a
result of intense cultural competition within a large extensively segmented but closed communal system: the various
9. Schacht, Origins, 13.
10. Levi-Strauss, Savage Mind, 90.

l7L
groups defined themselves by recourse to and interpretation
of a finite number of legal and doctrinal themes common to
them all. 11 ShfiC in spite of his novel methodology produced Just another variation on a structure: one doubts
whether he had really cut himself off from the ancient
schools. Indeed since the system he developed was made up
of elements which emerged in the course of rivalry between
those schools it could hardly be distinct from them. The
elements which went into his synthesis were the constituents
of revelation, normative precepts and a manner of dealing
with both; one might suppose since both revelation and
prescription were products of the same milieu that there
could hardly be insurmountable problems in relating them.
In spite of this the priority of legal rules over intractable revelation is evident from Shfi'I's frequent disregard
of his own methodology. 12 This is only one early illustration
of the fact that on the whole the fuqah' controlled or
disregarded revelation, they were not constrained by it.
The Imrni Shi c a by defining the imm as absent,
producing a canonical body of Traditions, accepting the
text of the Qur'n 13 and developing a system of ul signalled
their acceptance of the otherwise Sunni principle that the
moment of God's intervention was past and that law must now
be based on revealed texts - the residue and witness of
that ln t ervention . 14 Earlier or extremist Sh C I formulations
11. Cf. Wansbrough, Sectarian Milieu, e.g. 39_149 but passim.
12. Schacht, Origins, 15, 25 4 , 276-277, 323ff.
13. See Kohlberg, Some notes; and Eliash, The ShI'lte Qur'aTh.
14. The essentially divine origins of tiadth, and their
equivalence, as revelation, with scripture, are for the
Shi c a best exemplified in the Tradition quoted by G.
Lecomte, La Littrature du IIadlth, 96. From Ja'far
a1-Sdiq:
1

).,

'. >,,

J ))

) ,

C? '-- )'9

ci.

jyi ::.:- >'

cL&

175
asserted the continued manifestation of divine will in
human affairs through the person of the imm. ImmI Traditions and ImmI furti' were again new variations on old themes.
Linant de Bellefonds has noted that of the five major schools
of law it is the Hanafl not the Shidi which displays the
greatest deviation from the common norm of the other four.15
The Traditional literature of Sunnis and ImmIs also displayed a general conformity marked only by occasional variations, and even these were more significant for theological
Thus the methodology of reconthan for legal matters.16
ciling law and revelation which had developed in Sunn!
circles since the time of Shfidi was fully applicable to
the similar problems which faced the ImmIs in the tenth
century: they had only to adopt and refine from a wellestablished pool of exegetical techniques and terminology.
TsI's C uddat a1-Ul is probably the most significant work on u1 written by an ImmI writer in the Buyid
period. He acknowledged in that work the existence of only
two prior ImmI writers on the same subject, namely Mufid
and Murta. Mufld's analysis of ul is made available in
icarjikl's Kanz and is both considerably shorter and les.s
impressive than TsI's writing on that subject. Of Murta
T1 declared that he had written nothing in the field worth
turning to and nothing to be relied upon: that however
reflected professional antipathy rather than objective
T5I produced initially six exegetical
assessment.
principles relevant to the understanding of revelation, each
consisting of two elements: laqIqa and majz, awmir and
nawh, 'umm and khu, mu1aq and muqayyad, mujmal and
mubayyan, nsikh and manskh. Shfi'I had not displayed
such sophistication in his description of exegetical
apparatus: he had however provided the basic categories
15. Linant de Bellefonds, Le Drot Imamite, l8; see also
Coulsori, A History, 109-119.
16. See e.g. Lecomte, La Littrature, 9LL
17. For Murtad and flufid on usl, see Brunschwig, Les Ul
al-Figh Irnmites, 201-212.

176
which had been refined and subdivided by succeeding
scholars.18 In the event TsPs own discussion was not
free from some confusion and repetition but those headings
remained up to early Safavid times and beyond the essential
bases of exegetical analysis)-9 They were common to both
Sunni and ShII writers. Other principles which might be
related to the exegesis of revealed texts included the
application of rational argument (qiys, isti1b al-Il,
bar'at al-dhimma) and isn g d criticism (abaqt, jarl3 WatadIi). The whole series reflects the exegetical premise
that the 'sciences' of linguistics, rhetoric, history,
biography and logic represent independent and authoritative
disciplines; appeal to them in order to assess the meaning
of revelation was thought not only permissible but necessary.
That some or all of these sciences were in fact subordinate
to "the exigencies of scriptural [or Traditional] interpretation" has been suggested and may be demonstrated.2
IIaqiqa and Majz. Majz was in its earliest usage
an interpretative device applicable to revealed texts for
any of a number of reasons: to resolve grammatical problems,
to reconcile revelation and dogma, to obviate anthropomorphic
expressions or to remove or otherwise deal with legal
anomalies. 21 The transition from majz as interpretative
device to maJz as aesthetic category specifically applicable
to figurative usage is demonstrated by Wansbrough, 22 though
in fact rnajz never quite lost its broader application and
reappeared in Suytl, "a vague and general designation of
all phenomena requiring to be understood other than literally"?3
TsI adduced the antithesis 1aqIqa/majz and defined the
18. See Shfi'I, Risla, for
mm and khs, 179ff.; for
nsikh and manskh, 312ff.
19. TsI, cUddat a1-U1, 3, 11. For a Safavid example see
Hasan ibn Shahld II, Ma1im; the basic headings may be
found on pp. 25, 39, 10 14, 1514, 217 et al.
20. The quote is adapted from Wansbrough, Arabic Rhetoric, 1469.
Further reference to orientalist comment is given in
what follows.
21. See Wansbrough, Najz a1-Qur'n, esp. 258-259. Cf. Goldziher,
hirIs, 63-65/60-63. Goldziher found, perhaps a little
unfairly that the legal schools 'rise to the utmost levels
of distortion" when required to reconcile revealed texts and
accepted usage.
22. Wansbrough, Studies, 219-238.
23. Ibid., 237.

177
former as a word or phrase used so as to signify only its
appropriate (given) linguistic meaning; majz on the other
hand was present where a word or phrase had a signification
or significations which went beyond its given linguistic
meaning. Similar, that is as vague, definitions were given
later by CA11ma and Shand 11.24 Shahld II expressed the
antithesis as follows:

/ [ L

,j

c)

sj L
J-
, L jti

-q

jJ

In effect any word or phrase that might have more than one
possible meaning or that implied something other than its
specific formal content was thought an example of majz.
There was, it is true, with regard to all texts, a presupposition in favour of iaqTqa but it was never difficult
to find an indicator (dalli/qarina) suggesting majz.
Indicators adduced by T1 included reference to context and
inappropriate conjunction: they have as heterogeneous and
subjective a potential application as might have been
expected. Mo examples are provided but it may be noted
that it was only prior conviction that might detect in wa-j'a
rabbuka (Q89.22) inappropriate conjunction or discover in a
specific prohibition the meaning of not wjib or in the term
wjib the meaning "as if wjib".25
Under the heading majz TsI also discussed the
phenomenon whereby a word might have up to three layers of
meaning: the formal, the conventional and the legal (lughaw,
c urfi, shar q ). Thus formally dbba means simply animal or
beast, but conventionally it means a riding animal in
but
particular. Formally alt is the equivalent of du'
legally it is a reference to ritual prayer.26 it is not
214 . Ts, 'Uddat a1-Ul, 11; CAllma, Mabdi' al-Wul, 73_7)4;
Shahid II, Tamhid al-Qaw'id, 23; cf. Goldziher, Zhirs,
12 14125/117118. For the dominant Islamic belief in a God
given (as opposed to conventional) language, see H. Loucel,
L'origirie du Language.
h&rIs,
25. See Wansbrough, flajz al-Qur'n, 259; Goldziher,
64/60-61; above, 93-94.
26. This perception of layers of meaning might be compared
with Wansbrough, Studies, 2142-243.

178
quite logical but not surprising either that
si provided
for the preference of the sharI reading over the others:
"because a revealed text must be interpreted in accord
with the requirements of the sharCa.11 27 But the ostensible
purpose of ul was the reverse of that, namely to discover
the shari ca from a 'scientific' reading of the text.
's's
phrase however betrays the reality of ulT practice: texts
were interpreted in accord with the requirements of the
shari c a, not vice-versa. The law was more in need of
justification than discovery.
Shahid II discussed under majz, amongst other
problems, that of the unexpressed implications of a given
text. For example was the presence of a conditional or an
adjective in a revealed text restrictive? "I make this a
waqf for my poor children." Does that exclude the rich
children?
Some ulIs say that the implication (rnafhm)
of the adjective or conditional constitutes a
liujja; that is they indicate the denial of
the relevant juridical value in the absence
of the quality or condition specified. Some
say they do not constitute a Iiujja. Some
distinguish: they make the implication of a
conditional a buija but not that of an adjective
The implications of a conditional and an adjective
are only a liujja If there is not manifest In
the qualification a usefulness (f'ida) other
than the denial of the relevant juridical value.
If there is a usefulness then the qualifier
does not Indicate the denial of the juridical
value.....
It is "useful' [that an adjective be used] in
reply to a question; e.g. someone asked about
free-grazing sheep; Is there zakt on them?
The prophet said: On free-grazing sheep there
is zakt. The adjective there is [useful and
therefore] not restrictive. [I.e. the adjective
27. TsI, d lJddat a1-Ui1, 16-17.

179
"free-grazing" serves a purpose in so far
as it replies to a question so it may not
be construed as restrictive: there is also
zakt on sheep other than the free-grazing.] 28
That brief passage illustrates some of the characteristic

ulI

rules. There is ikhti1f in u1 as

qualities of
well as in furiic. The particular problem posed here has
at least three solutions and the ShI'a never came to any
conclusion on which was correct: all were possible.29
In giving expression to his own preferred view Shahid II
in effect stated that the adjective is restrictive except
when it may be shown to be useful. The assertion of
compelling rules which are mitigated by one or several
distinctions or exceptions is a characteristic feature of
ul literature. It preserves the illusion of rigidity
while providing for an extensive flexibility. In this
particular case the mitigating rule involves appeal to the
purely subjective notion of usefulness: even the example
given hardly inspires confidence. Again the intrusion of
subjective notions into ostensibly objective analysis is

ulI

a typical feature of
'Amm and Khi.

literature.

Amongst Muslim scholars opinion

on 'mm and kh expressions varied from that attributed
to Ash c arl - who assumed specificity unless there was evidence to the contrary - to that of Ibn IIazm who asserted
the 'general' significance of all statements unless there
was evidence of takhTs. 3 Goldziher noted that application
of the principle even when an extreme rule was adopted, as
by Ibri Hazm, was frequently a matter of exegetical taste
only. Between the two extremes there was naturally a
variety of possible positions including the negative one
that grammatical form did not indicate either cmm or
Iasan ibn Shahid II (d. 1011/1602-3) in a late summary of
Shi c i usl acknowledged that this view was favoured by some
Shi t i fuqah'. 31 Most however accepted that grammatical
28. Shahld II, Tamhld, 25 and 26.
29. Cf. cA11ma, Mabdi' al-Wul, 98-102.
30. Goldziher,

hirIs, l2O-l24/ll3-l17.

31. Hasan ibn Shahld II, Ma'lim, l04-106.

180
form was significant. There was no evident agreement on
what the significance was. }lasan ibn Shahld II noted for
example, though he had his own personal views in each
case, that defined plurals, defined sngulars and undefined
plurals were, according to some, of general, and according
to others, of specific application. 32 Further reading in
his work indicates both here and elsewhere a complete
absence of final decision. That was characteristic of a
synoptic writer In a mature tradition: he was concerned
to include and preserve as many options as possible. Such
generosity Is less evident in TsT's work: he belonged to
a more formative period and was more concerned with certainty
and precise definition in the face of an Intense polemic
encounter with other groups. In spite of this he too had
to provide for the phenomenon of ikhtilf: sometimes, he
admitted, the rules did not permit of final decision in
which case suspension of decision (tawagguf) and ultimately
choice became necessary. When for example there was no
clarification of a khss expression then all possible
aspects/applications (wujh) must be taken into consideration.33
Likewise if a mushtarak expression was devoid of any dalil
indicating which of its various possible meanings was intended
then tawagguf was necessary and choice. 3 This operation of
choice TsI was careful to distinguish from ijtihd, though
the distinction was in fact exclusively one of terminology.35
Again to Illustrate the characteristic features
of

uUlI prescription

a short quotation may be useful. The

following is from Hasan ibn Shahid II:
Some feel that if a general term is followed
by a pronoun referring back to part of what
is contained in the general term, that
oonstitutes takhI. 'Allma chose this
view [in one work]. NuIaqqiq relates that

T1

denied it. Muliaqqiq himself chose to

suspend decision (tawaqguf) s did 'A1lma

[in a different work] and Murtad.
32. Ibid., 106-111.
'Uddat al-Ul, 19-20.
33.
314. Ibid., 21.
35. See below, 192-193, 212-213.

181
Amongst the examples of this phenomenon is
God's word, [Q2:228J:
"Divorced women shall wait in seclusion for
three months etc."
Then God says:
"... and their husbands would do better to
take them back."
I'Iow, the pronoun in "take them back' refers
only to ru 1yyt rdivorced women who may
still return to their husbands] hence according
to the first view above only those are required
to wait apart for three months; and according
to the second view that applies both to them
and to the irrevocably divorced; and according
to the third view there is suspended decision
and that is best.36
This is a good illustration of ikhtilf in u1: four major
fuqah' embraced three opinions on this problem, one of them
two different opinions. They were all correct although
there was according to Iiasan ibn Shahid II a best view,
namely to avoid deciding. The motive behind the tendency
to eschew final decision is evident: a late writer in a
mature tradition could not afford to exclude any of the
great fuqah' who had been instrumental in moulding the
tradition and passing it on to him. That generosity of
ikhti1f and absence of decision was present in earlier works
but less markedly so. It may further be noted that sincerity
and piety were themselves not inconsiderable motives for
suspending judgement: meditation on the possible meanings
of revealed texts inspired the humble conclusion that they
were not susceptible to precise definition or limited
significance. The obscurity and infinite potential of
divine law reflected the qualities of its creator, as well
as the more accessible qualities of a literary style
(deutungsbedUrftigkeit) and an exegetical technique.
The general application of' mm and kh

was as
A11ma found

varied and as easy as that of majz/IaqIqa.

an Cmm term with a khs meaning in Q13.16, a11hu kh1iu
kulli shay'in, for obvious doctrinal reasons; in Q 9.5,
36. Hasan ibn Shahld II, Mat1im, 1L431.

182
fa-qtul

l-mushriklna, for well-established pragmatic

reasons. 37 The delegation Tradition from Ja'far al-Sdiq

was discovered to have an Cmm significance in order to
promote group interests and justify clerical authority.
A variety of legal rules were qualified by application, as
mukhai, of arra or Ija, terms which hide a variety
of extraneous and subjective factors.38
Awmir and Nawhi.
Isolation of this principle
as a separate element of exegetical apparatus reflects
the tendency of ul literature to proliferate distinctions
and sub-divisions. The problem was one of implication:
what was the juridical force of an imperative? Goldziher
noted that the orthodox schools adopted a flexible interpretation whereby an imperative might indicate isti1ibb,
nadb, ib1a or wujb. 39 Shand II in a general survey of
Muslim opinion listed fourteen different views on the matter,
amongst them the supposition that an imperative might
indicate any one of the five juridical values, including
tabrim!

That in effect was an abdication of the idea that

language provided authoritative rules, but indicates the

nature of the problem: the known law was such that various
imperatives had to be interpreted in divergent ways. The
principle of interpretation had to be such as would render
revelation consonant with the law. The Sh3'3 scholars
maintained in general that imperatives indicated wujb but
the rigour of that principle was mitigated by consideration
of such features as quantity, timing, the assumption of
kifya etcJ Shahid II having asserted his alignment with
the general ShI C T position carefully qualified its force:
If a command arrives with respect to a thing
which is related to the one commanded, and
if the one commanded suffers an impediment

(wziC) which prevents him from carrying it

out, then that command does not entail wujb;
for the intention of the jb is merely incitement to strive to do something and endeavour
not to neglect it in spite of the impediment .
37. CAllma, Mabdi' 130.
38. See also Wansbrough on C mm and kh,Studies,
39. Goldziher, Zh1rIs, 70-80/66-76.
0. See CAl1ma , Nabdi', 91-93.

169,191.

183
Shahid II then related that, by way of example, to the
propheYs injunction to youth to marry. That admonition
he indicated in opposition to D.wd al-h1rT does not
require Tjb for the reasons given. The reason was expressed
in the term wzic, impediment, which might mean as much or
as little as other factors dictated.141
Further analysis of exegetical technique would be
merely repetitive. Earlier assessment of developments in
furC provided ample evidence of arbitrary and subjective
readings, of tendentious intellectual arguments, of ambiguous
phraseology and of the unremarked interpolation of new ideas.142
Wansbrough's study of the manner in which a historical pattern
was imposed upon the details of revelation (naskh) produced
an impression of TT arbitrary if not irresponsible ascription."
There too options were rarely closed: a single Quranic verse
might be assigned to either Meccan or Medinan provenance
or otherwise related to a variety of historical events.14
The appeal to logic or to other rational arguments were no
more decisive than the appeal to the sciences of linguistics
and history. The Shi c i fugah
accepted the rational
principles of isti1b a1-I1 and bar'at al-dhimma as valid

ulI

proofs but their application respectively for and

against the validity of Friday prayer did not inspire conviction. '14 Ul arguments in general impressed none but
the already committed.
The methodology of ul involved ostensibly the
appeal to various independent sciences to produce authoritative
rules for the interpretation of revelation and ils translation
hirIs, 74-75/7O,
where the ZhirI rule is related to a Quranic verse not
to a prophetic Tradition.

141.Shahld II, Tamhld, 33; of. Goldziher,

142.See above (Shand II on the wujb of taking up office),

93-914; (on the significance of giving zakt to children),
1l5-l17 (Muhaqqiq on the distributers of khums), 138-139 ;
(confusion of s'I and faqlh),
123
143.Wansbrough, Quranic Studies, 177-178; see also 192-202.
144 See above, chap. vi , 161 and 163.

1814
into prescription. In all cases the formulation and
expression of those rules betrayed a concern to reverse
the process: revelation was interpreted so as to accord
with the requirements of the otherwise known shara. Ul
in practice was a mode of justification not a mode of
discovery. The need to project the illusion of authority
and rigidity led to the adoption of a variety of techniques
a major characteristic of which was the proposal of a constraining rule subsequently mitigated by the subtlety of
endless distinctions. The intrusion of a mitigating qualifier into a general principle further involved frequent
recourse to a purely subjective notion such as need, impediment or usefulness. With the passage of time and an
increasing sophistication in technique and terminology
there was a distinct tendency for the number of available
subdivisions to grow towards the number of known cases.
That tendency however was accompanied by less and less
certainty about what the rules were: generosity of ikhtilf
was another feature of util. That this betrayed the absence
of objective constraining principles was never conceded.
All of these features may be detected in any work of u1,
Shi d i or Surini, but a calm equanimity in the face of manifest doubt and a deliberate avoidance of decisions are more
characteristic of a mature than a developing tradition.

The function of ul was to deny historical change

and to assert at any given time the synchronic identity of
revelation and prescription. All experience could be related
to a timeless model, and, thereby, Islamic society "steered
a course between the contingencies of history and the immutability of design." 14 Paradoxically the exegetical instrument
which asserted identity was also, by its underlying flexibility and open-endedness, the prime instrument of diachronic
change. Distaste for the idea of change and development
already evident in the elaboration of exegetical technique
Lr 5. Levi-Strauss, Savage Mind, 73.

185
found further expression in the concept of tawtur. The
unchanging law was given chronological extension but was
presumed to be preserved free from modification. Concern
to demonstrate the faithful and reliable preservation of
the shari c a from generation to generation, from the moment
of divine intervention to the present day, accounts for the
proliferation of abaqt literature. That literature
served partly it is true as a reflection of juridical dispute 46 but it was primarily an assertion of charismatic
continuity. The tendency of biographers was to praise or
at least to excuse and justify everybody and that tendency
became more marked with the passage of time. Ibn Idris's
attack on TsI created some disjuncture in the sense of
charismatic continuity and so prompted the restitutory
efforts of MuIaqqiq and 'Al1ma. Subsequent attempts to
reject Ibn Idris were themselves modified: CAl1ma and
other fuqah' are said to have mentioned and accepted his
opinions (as elements of ikhtilf) and al-iurr al.-milI
(110 4 /1692-3) was able to say that "recent u1am' have
praised him and relied upon his work." 4 'A1lma's adoption
and adaptation of the theory of ijtihd 48 set a premium
on ikhtilf and prompted him and others to re-assess the
value of the fuqah' prior to TsI. Thus a pattern of consistency was imposed upon the past: all significant members
of the community were good members of the community. The
legitimacy of the shari c a as expression of God's will was
demonstrated on two levels: synchronically, by exegetical
assertion of the identity of law and revelation and diachronically, by biographical demonstration of continuity
between contemporaries and pious ancestors culminating in
the imms and the prophet. Submission to the implications
of these demonstrations produced the result that the sphere
of ikhti1f grew ever wider and the law became less not
more certain.
46. Wansbrough, Quranic Studies, 140.
47. Cit., in KhwnsrI, Rawdjt, 598, 1.9.
48. See below, Chap. IX.

186
The Culamat were the principal elements participating in tawtur and held a monopoly of exegetical technique.
They preserved, interpreted and declared the law, and took,
accordingly, in a very real sense, the place of the prophet:
Ibn Qayyim al-Jawziyya
they were warathat al-anbiy'.
(d. 751/1350) a late exponent of the Sunni (Hanbair) tradition noted that the ahba, the Tbidln and every succeeding generation of ulam' were, like the prophet, imms.
They shared that status with him and shared too their status
as muft: the prophet though distinguished from succeeding
generations by his reception of waby, revelation, was a
mufti acting as God's representative and declaring the law
on His behalf. Every succeeding muftr was likewise a
representative of God. 5 The tendency to elevate the faqlh
to a near prophetic status is evident. Muaqqiq exhibited
amongst the ShIca a similar tendency:

CU)

LJ

L,

JL

J%

You are when you give a fatw g giving information

from your Lord and talking with the tongue of
His law.51
The fuqah' like the prophet gave expression to God's law.
Thus the original confrontation between God and prophet
was re-enacted from generation to generation, the muft
taking the place of the prophet as mediator between God and
the community. That being the case, not only was it unwise, as suggested above, to attempt to impugn the opinions
of past scholars, it was necessary to recognise that any
opinion nce expressed could not be excluded. Or, viceversa, the ascription of an opinion to a pious ancestor
ensured that that option could never be denied. Ascription
of two or more opinions to an eminent scholar of the past
is a common feature of Muslim legal works. 52 There was in
149 . That concept was known to and exploited by the Sha
as well as the Sunnis: see e.g. Muhaqqiq, Mu'tahar, 3, ).
1-2; also Shahld II, Tamhld, Introductory passage.
50. Ibn Qayyim al-Jawziyya, Ae1m a1Muwaqqi C In, 8,9,11,114,21.
51. Mu1aqqiq, Mu c tabar, 14, 1. 1-2.
52. See e.g. Itin 1-lanbal's views as preserved in Ibri AbI Ya'1,
Tabaqt, passim.

187
fact a tendency for aqwI1 al-fuqaha' to be themselves
raised close to the level of revelation. cAllma declared
that a gT should have knowledge not only of the Qur'n
and Iiadlth but also of aqwIl a1-fuqah'. 53 Ibn Idris
dealing with khums was as concerned with the exegesis of
scholarly opinions as with the meaning of revealed texts.4
The proliferation of commentaries, especially in the mature
tradition, was a product of similar concerns: contemporary
law (the commentary) was presented as being identical with
the legal code of a major figure of the past. Contemporary
re-expression of the law thus asserted that what was now
the case was the same as always had been the case. The
lawyers imposed a pattern on experience which embraced all
t ime.

Appeal to the authority of a static law is contrasted by Wansbrough with appeal to the authority of a
(developing) charismatic community under the typological
headings of nomos and ecclesia. 55 He suggests that the
dominant cognitive category in SIra/Naghzi literature was
ecclesia but that was interrupted and replaced in the
classical period by the category nomos. The dominance of
the Idea of submission to an unchanging law is evident in
all Islamic legal literature and was scarcely vitiated by
appeal to the concept of tawtur which was presented as a
mode of preservation, not modification, by the community.
Appeal to the Qur'n, liadth and taw g tur as sources of law
certainly permitted change but disguised It as continuity.

53.

CAllma, Qawid a1-A1ikm, 145.

514. Above, 133-135.

55. Wansbrough, Milieu, 130-132 and all of chap. 14. Wansbrough

is concerned strictly with modes of salvation, whether

by submission to law, or membership of the community;
the relevance of his typology to authority is however
evident.

188
The Islamic community, both Sunn and Shi d i, presented itself as generally subject to an authority whose origins
were external to itself (revelation). Recognition of ijm'
as a further source of law may however mitigate that
presentation. It is possible that the concept of ijm' was
used to assert overtly the authority of the community over
part at least of the making and changing of the law.
Demonstration of the probative value of ijm' was
found for Sunns and Mudtazills in revelation: it was God's
or the prophet's word that 'proved' that ijm c was a source
of knowledge. 6 That being the case ijmc as authoiity
was strictly inapplicable to ideas or concepts logically
anterior to it. Demonstration of the existence of God or
Mu1arnmad or of the prophetic origins of liadlth or of the
modes of interpretation of revelation could not logically
be based on ijmc for it was prior acceptance of these
principles that proved ijm. The probative value of
ijma c rested in particular on tawtur. In order to avoid
a logically vicious circle it was necessary to demonstrate
that the concept of tawtur as probative did not depend on
ijm', nor on further tawtur. Concern amongst Sunni writers
to make this point, to distinguish ijm

frrm its logical

antecedents, suggests that there was a popular and unsophisticated use of that term which did in fact confuse ijm
arid tawtur. 57 The accepted proof for tawtur was 'objective'
'scientific' observation of the nature of things - which for
Sunnis reflected God's habit (da). Isolated attempts to
infer rationally from the 'fact' that tawtur was probative
a still stronger 'fact' that ijm was probative - the Cag1
argument - foundered on a distinction between the types of
knowledge to which each concept was applicable. Tawtur
depended upon an initial acquisition of cilm based on
56. See e.g. Hourani, The Basis of Authority, 19-31;
Marisour, L'Autorit, 60ff.; Bernand, LtIm t chez
'Abd a1-abbr, 37-38 and id., Nouvelles Remarques.
57. See Hourani, The Basis,

31-36.

189
sensory perception and then faithfully transmitted.
Ijm was applied to naarr problems of legal or theological
judgement. 8 The Sunnis, by their own lights arid in the
relatively sophisticated arguments manifest in their
literature, had a perfectly defensible theory of ijm'
but it was of limited application.
The ShI'i fuqah rejected the
ijmC.

59

samtl

argument for

Nor did they initially accept the probative value

of tawtur: that became a valid path to knowledge only once

it was conceded that the infallible imm existed and was
the mutawtirin. 6 Hence they rejected too the
standard CaglT argument for ijm.61 For them essentially
only the imm's word constituted a Ixujja. The imm however
though invisible or indistinguishable was anonymously present
within the Shidi community. If it were possible then to
ascertain the opinion of every competent person within the
community who might be the imm and if every such person
expressed the same view on a particular problem, then that
would constitute evidence as to the opinion of the imm.
On this basis, in so far as ijm uncovered the opinion of
the imm, it was a valid bujja.

62

That argument appears

more like submission to the tyranny of an accepted dogma ijmaC is a source of law - than a natural or inevitable conclusion. One suspects that the Sh'a were pushed in polemic
encounter to assert that they too had access to this source;
and devised thereafter a means which rendered it effective
only within their community. The argument justifying
acceptance of ijm as source was for the ShcTs as for the
Sunnis logically posterior to a chain of beliefs including
for the ShI c a, that God and the prophet existed, that the
imms were the successors of the prophet, that they were
58. Ibid., 31-32; BqillnI, Tamhld, l6l-l64; Tsi, 'Uddat
al-Util, 233-234.
Cljddat al-U1, 23 4 ff.; see also, above, Chap. III, 56.
59.
60. See above, Chap. III, 56-58.
'Uddat a1-Ul, 234.
61.
62. Ibid.,

232-233.

190

mam and that the last imm was anonymously present within
the community. Ijm
could not strictly be applied to these
beliefs. For the Shi c a as well as the Sunnis it was a
concept of limited application.

Is,

The opinion of non_ShI c

those who opposed the
Shi'i ul al-din, were according to Tsi not relevant to
- 63
the assessment of ijma.
The negative implications of
that belief were as important as the positive ones: Sunni
ijm' was not valid and by implication real knowledge (cilm)
was restricted to the Shi n sect. Furthermore, of Shi'is
it was only those who were learned in ul and furc whose
opinions were to be considered. The

and the muqallid

were excluded on the grounds that the imm, wherever he was,
was bound to be an C1jm. There was some willingness to
consider those whose status in f ilm was unknown, but generally,
for T 1 and for succeeding Shi d i writers, the elaboration
of the theory of ijmC' reflected and confirmed the authority
of the clerical class.6
Canvassing or recording the opinions of all Shi'i
scholars was hardly practicable so assessment of 1jm depended
upon analysis of various indicators, one of which was the
presence or absence of dissent. The opinion of acknowledged
dissenters to an otherwise general consensus could be disregarded if the forefathers and lineage of the dissenters were
known; for in that case they obviously could not be the ____
Now, the concept of anonymous consensus valid in the face of
any number of dissenters, provided they could be identified,
is not very satisfactory. How, for example, should the assent
of the anonymous be established? The matter seems more
likely to provoke debate than certainty. T 1 acknowledged
the problem and raised the question whether, if an opinion was
widespread and without apparent dissent, it could in fact be
accepted as iJm' 66 . His response to that question revealed

63. Ibid., 245-246.

61. Ibid., 248-250; cf.., CA11ma, Mabdi', 195-196.
cuddat al-Ul, 245.
65.
66. Ibid., 253.

191
a characteristic Islamic impulse to subordinate iJm

to

revelation:
We may know [that a widespread opinion is
(1)
in accord with the imm's] if there is an
indicator (dalil) giving rise to knowledge
which Indicates the validity of that view
But If there is an indicator opposing that

(ii)

view then we know that it is opposed to the

imam' s.
If we lack any indicator either way ... then
it Is necessary to accept the validity of the
[unchallenged] view [as i J mC ] .. because, if
the Imm's view were different, he would have
to make It known otherwise the commission
[takllf, from God] would be evil.6

It is evident from the first part of that quotation

that IjmC may be a result as well as a source of knowledge.68
It follows then that claims to Ijm may be either confirmed
or denied by appeal to revelation. In view of the highly
developed exegetical system at the disposal of the fugah'
their inability to render some part of revelation relevant
to some detail of iJmt may be doubted. Ijm
in this light
is both formally subordinate to revelation and may become
merely an ingredient In exegetical dispute. Submission to
the concept of ecciesia (community as authority) is transcended by the insistence on the appearance at least of submission to the concept of nomos (unchanging law). The second
part of that quotation established that widespread opinion
without dissent might be accepted as ijm; but this was
only the case if, in the heat of debate, it was impossible
to bring any element of revelation to bear on that opinion.
Ijm
thus defined was, as an instrument for the expression
of communal authority, of little value. It emerges rather
67. See above, Chap. 1114 53,n.1LI.
constitutes a root
68. Cf. Ghazll's comment that ijm
tT provlded it leads to the sunnah"; cit., Goldziher,

hirIs,

182-183/167. See also Bernand, L'Im', 35-36, for Narn's

view: "L'im' n'est gu'une consguence du vrai, il n'en
est pas la source."

192

as an element in polemic dispute. Claims to ijm' are likely

to be a result rather than a cause of conviction: this was
partly indeed conceded.
Absence of ijrn', which on the whole would be much
easier to prove than its presence, 6 implied the existence
of ikhti1f. That too was rendered formally subordinate to
the content of revelation (and to reason in so far as that
was a dalli giving rise to turn70): ikht1if might be
resolved by reference to any dalil giving rise to 'jim.71
If, of the various groups which embraced different opinions
on one matter, one consisted entirely of people whose
lineage and paternity were known then truth necessarily lay
with the more anonymous group(s). 72 But If there were no
external criteria, whether of revelation, or reason, or
biography, for deciding between the two (or more) groups
then the faqTh was at liberty to choose whichever opinion
he preferred.

Tsr, though

he wrote in a formative period,

and was more concerned than later writers with precise

definition and the discovery of certainty, was careful to
leave ample scope for the process of choice. That was both
a justification of ikhtIif and an Insurance against rupture
In the community whether historical or contemporary. He
explained the emergence of choice as an option in ikhtiif
as follows:
We express that opinion only because if the
truth lay with one of the two views [to the
exclusion of the other] then it would be
necessary that there be some way of arriving
at It; since there is no such way, that
Indicates that it is a question of choice.
If we postulate that the truth lies with one
of various views, when there is nothing to
distinguish it from other views, then in
such a case it would not be permissible for
the imm to continue in hiding; rather It
would be incumbent on him to appear and
-j Cj])
llL.11S.
69. See, e.g. above, ( cAllma
70. For the ImmI attitude to reason see below, Chap. VIII.
71. si, 'tfddat al-Uil, 2L6.

72. Ibid.

193
clarify the truth, or [he would have to ]
inform some of his followers ... and should
accompany the expression of his opinion
with a miracle to demonstrate its truth. If
it were not so takllf would be unsound. Since
we know that takllf continues valid, and that
the imm has not appeared, nor caused to appear
anyone to take his place, this constitutes an
Indication that (ikhtilf) has not occurred
[such that any one of the contending views is
actually wrong.]73
Thus the same Cagli premises that guaranteed ijmC were
used to guarantee ikhti1f. All variant views that arose
within the community (i.e. within the clerical class) were
permissible so long as they were not subject to the stronger
authority of a dalil giving rise to dim. The authority of
the community was carefully presented as inferior to that of
revelation. On the other hand the argument that the imm
would have to appear in order to rectify error guaranteed
the infallibility of the community. As long as the irnm
failed to appear the community could be sure that it had
produced so far no ruling which was actually wrong. Whether
it achieved ijm on a thousand details or none, all was
well: if anything went wrong the imm would have to manifest
himself somehow. It was not the doctrine of ijm which
rendered the community matim it was the doctrine of ikhtilf.
The same is true of Sunni Islam. What the ShI'a
achieved by intellectual principles the Sunnis had already
asserted on the basis of Traditions: "My community shall
not agree on error" and "Variance within the community Is a
blessing ". Of those,the. most important for justification of
the broad tolerance of the Sunnl community was the second.
The community's essential right-guidedness remains, even if
it achieved ijrnC, as in practice it did, on very little.
The articulation of charisma lay in the acceptance of
ikhtiif.
Wansbrough has noted the emergence of the idea
73. Ibid.; see also pp. 247 and 250-251.
74. See Goldziher,
hirTs, 94-102/39-96.

194
of a metaphysical community. He suggests that the quality
f

applied in sectarian theology to the imm was,

by the SurinTs, applied to the umma. That is, as he himself

says, a reasonable if unprovable assumption. The relationship of the concept of infallible community to Shfi'I's
is however, as he again admits, more complex
than he depicts. Shfidi finally perceived that ijmt (f
the community at large) had been achieved only on the basic
structure of the law, the 'ilm 'mma. On the details of the
theory of 1jm

law there was no 1jm', only ikhtilf. His objective, in

defining a sphere of f ilm khaa, which was also the sphere
of ikhtilf, was not uniformity, but rather to justify the
absence of uniformity. That done, the umma remained infallible in spite of manifest ikhtilf - but not therefore
because of ijm'. The argument is one of stress: it is true
that one of the features of the infallibility of the community
was its ability to achieve ijm; but a much more significant
feature was its ability to remain ma C tim in spite of ikhtilf.
Recognition and justification of ikhti1f were, I would
suggest, more necessary for the continued infallibility of
the community than recognition of ijm - though that too
was necessary. Shfi d l ? s terminological shift from jm' =
consensus doctorum to ijm

= communis opinio dimimished

the significance of the masses (al-mrna - whose knowledge

was restricted to general points) and enhanced the authority
of the clerical class (whose knowledge was of details which
were not required to be uniform).75
One of the results of T 1t assessment of ijma'/
ikhtilf, related to the comparatively modest sphere he
allotted to Caql as a source of iaw, 6 was to stress the
unique custody of Cilm by the clerical class. They were the
preservers of revelation, had a monopoly of exegetical
technique and variant views were valid if they arose within
that group. Murta, whose alignment with Mu'tazill beliefs
was straiter than TsT's, assessed much more generously
75. See Wansbrough, Milieu, 86; and above, Chap. I , 16-19.
76. For which, see below, Chap. VIII.

195
than he the application of Cagl to law, and much more
narrowly the application of ijm'. He asserted that the
imm was not bound to appear if the community agreed on
error: it was possible that the whole community could
achieve ijm in opposition to the imm without provoking
him to appear. 77 Ascertainment then of the imm's opinion
depended entirely on external factors (e.g. revelation)
and ijmC became definitively a result not a source of
knowledge. For Murta, as for al-Nazzm, (both TtMuttazilr):
"l'ima n'est gu'une consquence du vrai, ii n'en est pas
la source." 75 In the absence of any external evidence
supporting a claim to consensus Murta's group would disregard the claim and resort to action in accord with 'aql.79
Likewise in the case of established ikhti1f all attested
views must be rejected and recourse be had to caqi.80 The
significance of both these views is that they represent a
characteristic Mu t tazill effort to limit the definition of
divine law; they assert that the knowledge of the fuqah'
is both limited and finite. Where claims to ijm are not
supported by indicators, or where there is established
ikhtilf, the limits have been reached of clerical knowledge
and recourse must be had to Caql. But Caql was not a
prerogative of the clerical class. The tendency amongst
Muctazills to limit the authority of the clerical class
and to provide for a vast residual law based on reason may
account for their frequent popularity with political powers.81
The different and opposed views of TsI and Murtad
reflect already the impingement of ikhtilf in the sphere
of ShII uiil. The tendency of ikhtilf, in both ul and
furti t , to manifest itself more not less with the passage
of time and the need to justify tradition led to a transfer
of attention amongst u1i writers from the justification
of certainty to the justification of doubt. That reflected
77. TsI, cUddat al-Ul, 2L17.
78. See above, 191, n.68.
79. TsI, TJddat a1-Ustl, 253.
80. Ibid., 250-251.
81. See further, below, Chap. VIII, 206-214, 221-222.

196
the different concerns of a youthful and a mature system.
Ibn Idrls's outspoken attack on

Ts

brought to the surface

a nexus of problems associated with continuity, catholicity

and ikhtilf. That in turn provoked both Muhaqqiq and
A11ma to efforts of reconciliation and harmonisation. A
major precipitate of that effort was the introduction by
C Al1ma of a Shidi theory of ijtihd: that both reflected
and justified ShI'I acceptance that most of the law was
doubtful 82
The impingement of doubt is evident in Mu1aqqiq's
assessment of ijm:
ijm

is according to us a iiujja because it

includes the opinion of the imm; If a hundred

fugah' reached agreement but did not include
him that would be no lujja; conversely if two
did include him their word would be a lujja,
not because of their agreement but because of
the imgm's opinion
Therefore we do not accept the claim of one who
gives judgement and claims ijm t on the basis of
five or ten companions when we remain Ignorant
about the opinions of the rest. Unless, indeed,
there is definite knowledge of the presence of
the imm within the group.8
Appeal to definite knowledge (al-ilm ai-qa1T) meant
revelation since there was no other way of knowing the Imm's
opinion: Ijm t was result not source of knowledge. Absence
of known dissent, which for TsI constituted a principle of
certainty, for Muhaqqiq meant nothing:
for, just as we do not know [for certain] that
there is a dissenter so we do not know the
absence of one. As long as there is the
possibility [of an unknown mukhlif] there
is no certainty of the presence of the imm
amongst the rnuftIs.8
82. See further, below, Chap. IX. For the Sunnis, see Coulson,
Doctrine and Practice, 219.
83. MuIaqqiq, Mu t tabar, 6, lines 21ff.
8L . Ibid., 1. 2LL.

197
In MuIaqqiq's time things seemed much more doubtful and less
open to resolution. His views it is true are close to
Murtad's but his concerns were different: he was accounting
for the fact that ijm c , which superficially promised
certainty, had failed in the past to produce it (because
claims to ijm' were invalidly made) and promised for the
future little (because there cannot be certainty about the
absence of dissent ). t A11ma too accepted that ijm' based
on the absence of dissent was not permissible, but could
- 85
only be claimed on the basis of a dalil.
These two
writers reflect a complete abrogation of the principle of
communal authority: ijm c as a factor of diachronlc change
within the law was simply rejected in favour of revelation
and the unchanging law. Ijmt as a principle promised
perhaps too much certainty and rigidity at a time when
uncertainty was manifest and development necessary. In
practice too it was a cumbersome and inflexible tool not
so easy to manipulate as the various principles of exegesis.
Muliaqqiq had introduced the theme of accounting
for claims to ijm' where it manifestly did not exist. That
theme was taken up and dealt with by Shand I. The arbitrary
usage of the term might be, he suggested, excused on a number
of grounds. The claimants may have meant by ijm absence
of known dissent and that is not ijm c . They may have
applied the term ijmL to what is simply mashhr. They may
have failed to gain victory when they claimed ijm over
a dissenter. They may have subjected the khilf to a ta'wTl
so as to reconcile the elements of difference and so claim
1jm. Or they may have claimed ijm on the basis of a
Tradition when they meant simply that it was current in their
books and linked to the imm. 86 In this way Shahid I
"excused" various usages of the term ijm none of which
were the real thing and none of which constituted certainty.
This devaluation of the concept of ijm' was part
of the larger distrust of certainty which increasingly
85. CA1lma, Mabdi', 1914_195.
86. Shahld I, Dhikr, I.

198
manifest itself in ShI C I uii1. The usili writers were far
less concerned with factors conducive to knowledge (turn)
than with factors conducive to opinion (ann). That was
as re-defined by
the sufficient basis of ijtihd. Ijm
Shahld I could be used as only one of several elements
which justify decisions based on ann/ijtihd. Like earlier
writers Shahld I insisted that absence of ikhtilf was not
sufficient to establish the 'real' ijm: positive evidence
of the imm's participation was required and that must
depend on something other than ijm. He raised the question
of whether the imperfect ijrn' based on absence of dissent
could be used as a iujja in the absence of an independent
He replied that it could. That
bujia t aqliyya or naqliyya.
was less contradictory than it seemed because by this time
the meaning of bujja too had been devalued: it no longer
meant something productive of certainty (Shahrd I had already
conceded that absence of dissent did not produce that) it
meant justificatory support for an effort of ijtihd. He
added finally that shuhra - being widespread, well-known thus was reformulated
could also be used as a iiujja. 8 Ijrn
as a factor in an argument, a polemic prop, no longer productive of certainty but simply on a par with shuhra. In
any case the fuqah' were no longer concerned with certainty
but with ann and ikhtilf and the greater freedom and
authority offered by acceptance of the theory of ijtihd.
Hasan ibn Shahid II carried to extremes the
devaluation of ijm as a source of knowledge.
In truth it is usually impossible to ascertain
in our own time
the establishment of ijm
except by way of transmission (nagi). For,
there is no way to know the opinion of the
1mm since such knowledge depends upon the
existence of unknown [anonymous] mujtahids
within the consensus ... and knowledge of that
is certainly impossible.
claimed in the writings
Every case of ijm
of our companions from the time of TsI to

87.

Ibid., 4-5.

199
the present time which is not based on
nagi mutawtir or [under the appropriate
conditions]

or which is not accompanied

by gar'in giving rise to dim ... in every

such case, that which is meant by ijmC is
that which was claimed by Shahid I namely
shuhra not ijmc.88
That perception ensured that all claims to ijm
from the
time of T 1 onward could be disregarded as invalid. It
afforded retrospective justification for the continuance
of ikhti1f in spite of claims that it had been resolved
The argument it is true implied some criticism
by ijm'.
of early writers who, it appeared, had carelessly misused
a term which Hasan ibn Shahid II thought could be appropriately
defined and specifically applied. What he perceived as misuse however remained as peripheral elements of ikhti1f:
later us1Ts could define ijmc in various ways and justify
their definition by appeal to various writers. Hasan ibn
Shahid II himself disapproved of Shahid I's attempt to
'excuse' the arbitrary usage of earlier scholars: it was
simply wrong usage. 8 He further denied that shuhra could
be used as a tiujJa unless it dated to the period before TisI:
1T juridical
influence was so great that in the period after him most

that was justified on the grounds that

fugah' followed his opinions with the result that they

became mashhr, but that was simply the domination of one
man's opinion. 90 The motive of that argument was again to
break the rigid mould that might form as a result of acceptance of juridical opinions that had become prevalent after
the time of T1 Hasan ibn Shahld II denied both ijmC
and shuhra as juridical arguments except in so far as they
could be dated back to the time before T1
In sum:
88. Hasan lbn Shahld II, Ma1im,
89. Ibid.,
90. Ibid.,

176-177.
178-179.

177.

200
The claimant to ljmac has no alternative
but that his claim be based on one of the
urug giving rise to t ilm, the least of
which is a khabar [wid] qualified by
appropriate qar 'in...
The status of ijm once it enters the
sphere of transmission (I?ayyiz al-naqi)
is the same as the status of a khabar
and is subject to the same conditions of
acceptance.91
Claims to ijmC, with Iiasan ibn ShahTd II entered te
"sphere of transmission" and were subject to the whole
paraphernalia of grammatical, rhetorical, historical, biographical and logical rules which might be brought to bear
on a khabar. It was admitted that there might be conflict
between two transmitted ijmCs or a khabar and a transmitted
assertion of ijm C : the conflict would require reconciliation by tarjTi (reasoned preference) or tcgdul (choice),92
i.e. ijtihd. This represents in effect a firm denial of
the principle of communal authority: authority is contained
in the limited, defined, static, unchanging 'events' of
transmission (naql), is a product of past history not of the
developing community. Iasan ibn Shahid II's redefinition
of ijm as having to take place amongst those who knew
the imms is a concept which may be compared to that
al-s.al)ba.
espoused by Sunni groups who accepted only ijm
The potential of ijm as an instrument of diachronic
A reason for this may be
change remained unexploited.
detected in Hasan ibn Shahid II's assertion that claims to
were subject to exegetical interpretation: exegesis
ijm
with its stress on the unchanging law was, paradoxically,
a much more flexible basis of authority than ijm'.

91. Ibid., 182.

92. Ibid.

201
It would however be an oversimplification to
rest on the assumption that the fuqah' failed to exploit
iJm' because it was a rigid or unmalleable concept. It
was that, because they failed to exploit it. They preferred
to depict iJma' as not a real source of knowledge but a
result of knowledge gained from other in particular revealed
sources. IjmC in practice continued to be used in polemic
debate as long as it was accompanied by (a result of)
gar'in giving rise to ti1m/ann. This constant subordination
of communal authority to unchanging law, ecciesia to nomos,
' reflects an irrational but obstinate "fidelity to a past
conceived as a timeless model, rather than a stage in the
historical process "
Depiction of the law as a constant
restatement of the prophet's or the imams' biographies was
simply more congenial to the Islamic world view than depiction of the law as submission to the authority of the diachronic community.
For the Islamic world "antiquity and
continuance are the foundations of iegitimacy."

Sunni attitudes to ijrnt, like ShiI ones, were

not univalent. There were some who doubted whether it
could be achieved or be known to have
retricted its scope geographically to
the
hirTs, chronologically, to the
the end of the eighth century ShfidI

been achieved; some

Medina, and some,
Sahba. Already by

had perceived that

claims to ijm were largely polemic and did not reflect
real certainty or absence of dispute. He himself was inclined to restrict its application to 'urn al-mma, that
is to generally accepted structural elements in the law
such as fasting in Ramadn or zakt; the details of the
law he conceived to depend on 'urn al-khssa, the preserve
of the clerical class, wherein precisely there was no
certainty but only ikhti1f. 95 Appeal to ijm' in Sunnr

93. Levi-Strauss, Savage Mind, 236; cf. Wansbrough, Milieu,

81, 91 et al.
9L. Levi-Strauss, Savage Mind, 236.
95. See Schacht, Origins, 88-9'4; Shfi'I, Risla, 961-967;
above, Chap. I, 16-19.

202
Islam after ShfiI was constantly offset by qualifications
as to its scope or doubt as to its achievement. The legal
stability which ijmac seemed to provide for was balanced
by the need for flexibility which ijm' seemed to pre-empt.
The evidence of ikhtilf whether within or between the
schools was overwhelming and led to a generally cautious
estimate of what was covered by ijm, epitomised in the
following from Ibn Taymiyya:
))'

LL

LY

L
There are many problems on which it is thought
there is ijm' but the matter is not so: in
fact the alternative view may be preferable
based on Qur'n and sunna.6
The Sunnis in fact shared the tendency, evident amongst the
ShI Ca, to subordinate iJm' to exegesis and revelation.
Evident from that remark by Ibn Taymiyya, it is manifest
also in GhazlI's comment that consensus constituted a root
"provided it leads to the sunnah; it is consequently a root
of the third degree." 97 That reflects not only the formal
subordination of ijmC to revelation but suggests also that
it was for GhazlI as for the Shica a result rather than a
cause of knowledge: it constitutes a root only if it leads
to the (otherwise known?) sunna. It is in other words
probable that ijm as a source of authority was not for
the Sunnis much more effective than it was for the ShiCa.
In spite of this some western scholars have
assessed iJm' as the foundation of foundations of Islamic
law, as the ultimate mainstay of legal theory, even as the
conscience collective of the Islamic community. 8 That
conclusion may reflect a popular and unsophisticated Islamic
usage but the scholarly class on the whole were aware that
some such charge was possible and defended themselves
96. Ibn Taymiyya, Majmt, Vol. 20, 10.
hirIs, 182-183/167.
97. Cit., Goldziher,
98. Hourani, The Basis, L9-54; Schacht, Origins, 2;
Coulson, A History, 77-80; Berriand, L'Im', 30.

203
accordingly against it. The concept of ijm c as defined
and used by them had a distinct and limited application and
may be distinguished from the "foundation of foundations"
which orientalists have detected and termed iJm. The
basic elements of the Sunnl defence have been set out by
Hourani. Mansour in a recent work noted Hourani's analysis
of the Sunni position and suggested a distinction between
ijm
"comme abstraction, comrne concept des ul . . ." and
ijm ITcomme falt, come rsultat (ijm
sur x, sur y, sur z)

That distinction however was elaborated without
any reference to original texts and may not be legitimate:
it is after all only by application of ijm
as abstraction
that iJmC can be achieved as result, on x or y or z. He
suggested that there was at any given moment "un ensemble
d'ijm's de fait (sur x,y,z,) gui dfinissent une orthodoxie."10
In fact Muslim scholars In general were well aware that
there were very few ' T ijm C s de fait" 101 ; and In any case
orthodoxy depended far less on details of furi c al-f igh than
on details of usl al-dIn. The source of those details was
usually something other than ijmC.
Of the points cited by Hurgronje 102 as depending
ultimately on IjrnC none were conventionally acknowledged
to be so by Muslim scholars. Thus knowledge that there
was in Arabia someone called Mul?axnmad or that the Qur'n
was really a collection of Muhammad's sayings depended,
according to them, on tawtur.
That the Qur'n was ultimately from God, on the other hand, could be proved by
rational mearis) 3 That the contents of the Qur'n or
hadith had been well-understood or appropriately explained
was a proposition proved by appeal to the 'independent'
exegetical sciences: linguistics, history, biography,
logic etc. There was in all these matters no overt appeal
99.
100.
101.
102.
103.

Mansour, L'autorit, 30-31, 166-167.

Ibid., 167.
Cf. Coulson, Doctrine and Practice, 219.
See Hourani, The Basis, 50-51, and ref's. given there.
Cl. Bqi11nI, Tamhld, 114-116.

20
to ijmC. There were, however, underlying such Muslim
beliefs, certain epistemological presuppositions which
may be isolated and analysed. That God exists, for example,
that what He says is necessarily true, or that what a
prophet says is necessarily true, were axioms accepted by
Muslims on rational grounds (not ijm') though not in fact
susceptible to rational proof. Equally that what a tawtur
number of persons say under appropriate conditions must be
true, or that grammar is a science which might objectively
lay bare the meaning of a text are propositions conventionally
accepted by Islamic society but hardly convincingly a priori.
To describe these or similar axioms as based on ijmC would
be a misuse of a term whose application was restricted to
naarT matters of jurisprudence or theology. Such matters
were for a Muslim not naarT, they were rational and irreducible statements of objective scientific fact.
Isolation and analysis of the essentially irrational
but conventionally irreducible bases of Muslim thought and
belief is one achievement of Arkoun's article entitled,
Logocentrisme et Vrit Religieuse dans la Pense Islamique.
In that article he is concerned to demonstrate that Islamic
religious literature (as epitomised in the IClm of al-Amiri)
builds its coherent structure not on the basis of objectively
perceived reality but In response to the pressures of an
emotionally charged vocabulary, a traditionally argued and
slightly mystic technique of exposition and finally on the
irreducible but rationally indemonstrable principles of a
criteriology and a credo.1
As with all cultures coherence
is Imposed upon, not extracted from, objective reality.
The literary world thus created Arkoun describes as unreal
(rv) but coherent and true. 105 The basis of that truth
might possibly be described by the English word consensus
but probably should not be described by the technical term
ijm C . It might better be assessed, in Arkoun's terms, as
logocentricity, as cultural idealism or simply as convention.
lO4. Arkoun, Logocentrisme, 19-25, 26-27, 28_ 142, '42-45.
105. Cf. Wansbrough, Milieu, l'41-142.

205
The foundation of foundations of Islamic culture, as of
other cultures, rested on nothing more than the need to
build coherence by first assuming certain 'facts'.

206
CHAPTER VIII
REASON AND REVELATION
Appeal to the authority of a fixed and finite
body of revealed texts was the dominant principle of juridical speculation in Islam. Submission to that principle,
evident in the limited exploitation of ijm', was reflected
further in the generally negative or highly qualified assessment of reason (cagi) as an independent source of ethical/
legal values. The pragmatism of the more ancient schools
of law had ensured for reason in the assessment of prescription a considerable if undefined role and had prompted
a corresponding disparagement of the khabar wid - marginal
revelation.

ShfidIs work established revelation as a

major if not unique source for law, justified recourse to

khabar wiid and assigned to reason the merely subordinate
roles of Interpretation and analysis. It was at best a
handmaid to revelation. Later extremists such as the Zhirls
proclaimed themselves willing to adopt the weakest Traditions
rather than resort to ra'y. 2 At the opposite extreme the
Mu'tazlla and the Philosophers conceived divine law as circumscribed by the antecedent (rational) principles of good
and evil, or as a symbolic and inferior manifestation of
intellectual truth.
Hourani termed the chief opposing ethical theories
in Islam as on the one hand rational objectivism and on the
other theistic subjectivism. The former is applied to the
belief that what is forbidden Is so because it is objectively
evil; the latter to the belief that what is evil is known
to be so or rendered so because it is forbidden. 3 Theistic
subjectivism is characteristic of Sunnl and more particularly
Ash c arl belief. For its adherents every detail of the law
is conceived to be merely contingent: all values might have
1.

Schacht, Or1ins, 50-51.

2.

Goldzlher,
hirIs, 6-7/6-7.
Hourani, Rationalism, 8-12; see also, id., Two Theories

3.

of Value.

207
been otherwise. Actions become or are known to be good or
evil only as a result of God's command. The alternative
theory found its earliest known expression in the document
quoted by Wansbrough which presents a dispute between a
Jew and the Muttazill theologian al-Nazm (c.23O/845).
Al-Nazzm there proposes a distinction between non-contingent
ethical values such as justice, honesty and charity and
those that are contingent upon a (divine) command, such as
prayer and fasting. Godts law relates only to the latter
category: it is there that God's command renders certain
actions contingently good or evil. Those things which God
has made good "In their very nature" are not subject to His
prohibition and may be rationally known without recourse to
revelation. Revelation in other words was accorded a role
but only a limited role in the assessment of ethical values:
it could not contradict what was rationally perceived to be
good or evil, and was effective only in the sphere of the
contingent. Initial insistence by the Muctazills on restricting the corpus of revelation to the Qur.n alone may
reflect a desire to limit the realm of the contingent law
and correspondingly to limit the authority of the clerical
class. 5 (Though it should be recalled that their insistence
on the Qur'n as sole source of divine law was an early and
decisive assertion of clerical prerogative: it prompted
(and was prior to) much of Shfi''s more extreme theorising.)
The incorporation of the Muttazill theory of
contingent and non-contingent ethical values into Shici
legal speculation Is evident from the following passage,
taken from Ts's C uddat ai-Usih.
(I)

The deeds of a mukallaf may be either hasan

or qablli [good or evil]. The good may be
either obligatory or recommended or permissible.
Every deed whose evil is rationally known
l-Iar) ; for example
is forbidden (tal

1. Wansbrough, Milieu, 109-112.

5. For further details of Sunnl and Muttazill attitudes to
reason and revelation see Wensinck, Creed, 260-263;
Hourani, Rationalism, 129-139.

208
injustice, lying, frivolity, ignorance and
such like things. What is known [rationally]
to be obligatory ... is obligatory; for
example returning a deposit, thanking a
benefactor, justice and such like things.
What is known to be recommended ... is
recommended; for example charity and generosity. These things are as we specify because
it is not valid (l yaili1i) that they change
[in status] from good to evil or from evil
(ii)

to good.
There is ikhtilf concerning those things
which are useful to man on whether their status
is that of prohibition, permission or suspension
(wagf). Many BaghddIs and some ImmIs have
embraced the opinion that they are in a state
of prohibition ... Most Basrr theologians and
many fugah consider that they are in a state
of permission. That is the view of Murtad.
Many have embraced the option of suspension [and
say] that both permission and prohibition are
possible; it is necessary to wait on revelation
indicating one or the other. Our shaykh Ab
c Abdullh (al-Mufid) adopted this position
which is in my opinion the strongest.
The evidence for this view is that it is
rationally certain that to proceed on a path
concerning which there is no certainty that it
is not evil is like proceeding on a path which

is known to be evil.6
had
In the passage immediately preceding that,
indicated that the categories of malir (forbidden) and mub

T1

(permitted) went beyond the sphere of rational good and evil:

things, otherwise neutral, could be rendered into those
categories as a result of God's command. He thus accepted
a sphere of rationally known, non-contingent, ethical values
whose status could not change. He accepted further a sphere
of irrational contingent ethical values whose status was
6.

'tfddat al-Ul, 296.

209
rationally neutral but which were given legal value as
a result of God's command. His description of the noncontingent values was characterised by examples of' general
and open-ended significance (justice, charity) and by
recourse to the phrase "and such like things". There was
in practice no final demarcation of the contingent and noncontingent. He divided the category good into three subdivisions corresponding to the juridical values, wjib,
mandb and mubli. Like his older contemporary, the Muctazill
Abd al-Jabbr, he was unable to provide (rationally) for
the (legal) category makrh and so produced, like CAbd alJabbr, a four part system of values. This both reflects
the close connection of the Shi C r theory with its Muctazill
source and suggests perhaps an early stage of intellectual
assimilation.7 T 1 did not in fact in this passage provide
a rationally proved definition of the category mubI. He
stated that such a category could be assumed to exist if
one accepted the (rational) presumption that things which
are useful to men are in a natural state of iba. This was
a part of Mu'tazllI belief which had been accepted by
Murta. 8 Some Shi C is however, according to
had
adopted the opposite view, that is a presumption of pro-.
hibition. Others, like Mufid, had suspended judgement and
relied, for a solution to such problems, on revelation.
This was TsI's view. 9 The whole is an example of ikhtilf
In ShII ul and casts doubt on the supposition that
reason is an adequate judge of ethical values: reason
after all here provided for all possible conclusions.
TisI's rejection of the principle of "ibIat al" (the presumption of permission with regard to things
useful to man 10 )reflects his moderation in the application
7.
8.

See Hourani, Rationalism 39 and 9-5O.

See, for 'Abd al-Jabbr, Hourani, Rationalism, 71-72,

110-115; Peters, God's Created Speech, 88-89.

9. Cf. foil Mufld's view, KarjikI, Kanz, 192.
10. The term ibat al-asi implies that things are in a natural
state of iba and continue so in the absence of any relevant
detail of revelation. This presumption of continuity, pending revelation, is also termed bara'atal-asl or islat albar 'a.

210

of reason to law and ethics and his concern to assert a

sufficient role for revelation. Murtads acceptance of
this principle on the other hand procured for hm a vast
area of residual ethical knowledge available without
recourse to revelation and so reduced his dependence on
that source. The different attitudes of these scholars
to the problems of reason and revelation in law may be
further elucidated by consideration of their responses
to the problem of the khabar wid.
Murtad rejected appeal to the khabar wid In
order to ascertain details of the shar3a on the grounds
that it occasioned neither Clim nor carnal; for 'urn was,
according to him, a necessary precondition for Carnal. He
acknowledged that reason alone did not preclude the possibility that worship might be based on the khabar wiid but:
We reject qiys and the khabar wid
simply because God has not in fact imposed
worship by them, nor provided any dalil
suggesting recourse to them. Hence we reject
action in accord with them and we deny that
they are paths to tarTm or talilfi.
In this way our companions, ancient and modern,
have denied practice in accord with the
khabar waiId and qiys and have criticised so
severely those who cling to them [as source]
that the [immI] madhhab has become famous
for its display and manifestation of this
rejection and it is never doubted as one of
their characteristic opinions.11

Tsi, while

agreeing that the khabar wid did

not give rise to Gum, asserted nonetheless that not only
was it rationally possible for acts of worship to be based
on it but specific permission was available in the law
(fT 1_sharc). That permission related only to those akhbr
which were transmitted by a member or members of the True
Sect and who possessed the qualities of Cadla and others
11. Murta, cit., Ibn Idrls, Sarir, 14, 11. 12-15; cf.
Brunschwig, Les U1 al-Fiqh, 210; for Mufid, see
KarjikI, Kanz, 193.

211
(wa-ghayruh). In the event Ts produced no revealed
sources for his opinion. Having rejected the arguments
against khabar wliid based on Q.2.169 and 17.36, he adduced
as his only prop 1jm' al-firga al-mul4q q a. This he inferred
from what he considered to be the constant practice of the
ShIa from the time of the prophet oriwards) 2 There is rio
doubt that TsI's advocacy of the khabar wiid was completely novel in the context of ShIfl ul and his arguments
spurious. A sufficient indication of this is provided in
the first two objections which he advanced for the purpose
of refutation:
Ci)
How can you claim that the True Sect has
achieved ijm on permission to act in
accord with a khabar wi1d when it is wellknown concerning them that they do not approve
of actions based on the khabar wuid just as
they do not approve of actions based on qiys?
(ii)
Is it not the case that your shaykhs constantly
dispute with their opponents on the grounds
that the khabar waluid is not to be acted upon...?
We have seen not one of them who mentions that
such is permissible, nor one who wrote a book
on such, nor one who dictated a view on this
matter. How then can you claim the opposite?13
TsPs innovation, best understood as an affirmation and
extension of the revealed basis of the law, In view of the
radical departure from accepted opinion that it represented,
requires to be explained.
He himself produced an extensive defence of his own
view. The first element of that defence was an assertion
that the traditional Shi c i rejection of the khabar wid
referred only to those transmitted by their sectarian opponents. It was permissible to use an Immi khabar wiid only
if the transmitter possessed the quality of C adla; if the
transmitter did not possess that quality, as was Ipso facto
12.
Cljddat a1-U1, L1_57.
13. Ibid., 51 and 52.

212

the case with all non-ShI'1 transmitters, then it was not

permissible) That assessment characteristically denied
the validity of all alien sources while defending and upholding ShII sources.
A much more significant problem was raised when

T sI ' s

interlocutor posed the question whether his accept-.

ance of the khabar whid did not lead to the conclusion
that truth lay in two different directions (jihatayn mukhtalifatayn).
Denial of that conclusion, the interlocutor
claimed, was well-known, on the part of ShI'i scholars and
elders. TST replied:
What is well-known is that the truth may not
lie both with [the Shi ca] and those who
oppose their faith. As to its being well-known
that the truth cannot lie in two directions
if it emerges from two variant khabar, we have
demonstrated that this is known.15

What such a situation implied was the necessity and the permission to choose. TsIts concern to establish areas of
choice has already been demonstrated several times: it
reflects a desire to incorporate as harmoniously as possible
the divergent characters and views which had been gathered
into the nascent Shi c i tradition. Acknowledgement of choice
as an option in the assessment of legal value was in effect
a confession of doubt and brought the ShIa very close to
the Sunni theory of ijtihd, which nonetheless tradition
required that T 1 reject. He was himself aware that the
distinction between choice and ijtihd was at best a
subtle one:
Some say (in a case requiring choice) that
each mukallaf should act in accord with his
own ijtihd but that would lead to the
supposition that kull mujtahid mub,
which we do not accept. Only choice is
possible. 16
l Ll. Ibid., 51-52.
15. Ibid., 53.
16. Ibid., 21.

213
It is patent however that every chooser was muIb.
1 ' consciousness that ikhti1f was a serious
problem may further be inferred from the fact that he made

that phenomenon itself a justification for accepting the

khabar wid. The evidence of ikhti1f, arising out of the
use of khabar wiid, proved that the latter was a valid
source. He pointed out that ikhti1f amongst the ShIa was
greater than that between Ab IanIfa, Shfi'r and M1ik.
In spite of this great divergence of opinion
I have found [the ShI C a] such that not one
has cut the bond of association nor accused
his opponent of going astray or being corrupt;
nor has anyone severed himself from the
dissenting opinion.17
This claim to have detected in the past history of the ShI'a
a broad tolerance in legal affairs reflects a contemporary
concern both to justify the past and to establish for the
present and the future a similar broad tolerance. TsI's
concern to introduce in a moderate way some provision for
doubt and flexibility may be contrasted with Murtad's more
rigorous concern with certainty. For the latter, the
existence of ikhtilf proved the limits had been reached of
shar'I knowledge, and indicated the need to turn to. reason.18
He argued that the conditions of taklTf required that God
provide a dalil qa C T, a definitive sign, for every Iukm
shar'I: if there was concerning a given problem no
definitive sign then again it was necessary to turn to
reason. 19
Murtad's generous assessment of the capacity of
reason to deduce legal and ethnical values, evident from
his acceptance of the principle of ibat a1-a1, was
accompanied on the whole by a correspondingly limited
assessment both of what constituted revelation (not the
khabar wiid) and of how useful it was (no ikhti1f). He
proposed in effect that the contingent law, based on
17. Ibid., 56-57.
18. See above, 194-195,.
19. Murtad, cit. Ibn Idrls, Sar'ir, 14, 11. 28-32.

2l4
revelation, should be both limited in extent and certain
as to content. The opposing view represented by

TS

insisted on a more restricted role for reason in ethics,

evident in his rejection of the principle of ibat al-al.
Associated with that rejection was TSI'S generous assessment of the elements of revelation and his acceptance that
ikhtilf within the clerical class was itself a source of
knowledge/authority. Murtad's beliefs, probably closely
allied to NuCtazill beliefs, implied for the clerical class
a limited sphere of authority.
though he accepted
that reason had some role to play in the establishment of
legal and ethical values, acknowledged, by comparison with
Murta, fewer limits to the contingent law (though still
it could not contradict the non-contingent values) and
correspondingly fewer limits to the authority of the clerical
class. The contrasting attitudes of these two scholars
provides evidence for the existence of a spectrum of possible
juridical positions varying from extreme ItMuctazill?' on the
one hand to "Zhirr" on the other; the former characterised
by insistence on reason as the major source of legal/
ethical knowledge, the latter by insistence on revelation.
That spectrum may be detected amongst SunnT as well as ShI'1
writers. That the r .Iu t tazill position should be more congenial to de facto powers is hardly surprising. The censorship exercised by Ibn t Abbd in Rayy was directed against
all Traditionists, quite irrespective of their Sunn/ShI'I
affiliations. 20
TsI's comparative devaluation of reason together
with his stress on revelation might be related then to a
concern for the authority of the clerical class as well as
to a concern for the incorporation of divergent elements
that had emerged during the development of the ShI'I community.
That there was a strong Traditioriist element within that
community was already evident from the production by the
ShI c a of four major collections of iadrth, two of which were
by

Tsi

himself. Indeed a personal interest in his own

20. Madelung, Imamism and Mu t tazilite Theology, 20.

215
collections may have influenced his u1I positions since
his two books were notorious for containing a great number
of khabar wid. 21 That he was aware of the pressures
exerted by a Traditionist bloc within ShI'ism, and that he
conceded the need to placate that bloc is made evident in
the Introduction to the r .'!abs. There he complained that
the Sunnis spoke slightingly of the ImmIs because of
their paucity of fur

and mas'i1. He conceded that

paucity but denied that it was intrinsic: the ImmI corpus

of revelation could be interpreted exegetically to provide
as many masil as the Sunnis produced. Where the Sunnls
had recourse to giys and 4jt1hd - which led only to opinion
(ann) - the Shi c a referred to the principles of rational
al and bar'at al-dhimma - which produced knowledge and
required action (dim and Carnal).
I have been desirous both recently and in the
past of producing a work which would contain
[a profusion of fur and mas'i1] but have
been prevented for various reasons. My
intention of doing that has further been
weakened by a lack of desire amongst the
Imrnls for this and a lack of concern on
their part for this problem. For, they have
collected and related Traditions clearly
expressed in such a way that if the wording
of a problem were changed or its meaning
expressed in a manner different from the one
they were accustomed to they were astounded
and incapable of understanding it.
T 1 indeed when he wrote the Nihya had deliberately clung
to the wording of Traditions in order not to provoke astonishment and reaction. 22 The need to make concessions to this
Traditionist bloc may consciously or unconsciously have
prompted T 1 to stress revelation as the major source of
Shi C i law.
The existence within ShI'ism of a strong Traditionalist element side by side with a strong MuctazilT element is
21. Brunschwig, Les Ul al-Figh, 211-212.

22. Tsi, Mabs,

Introduction.

216
confirmed in Madelung's study of Imamism and Nuttazilite
Theology.

He is inclined to detect a movement from (i)

an early stage of almost complete incompatibility, to (ii)

a drawing closer in some of the basic concepts, and finally
to (iii) an outright adoption of the method and content of
Muctazill kalm. 23 His study does not suggest a strict
chronological development: the NutazilI impingement began
as early as the end of the ninth century with the BanI
Nawbakht but the pure Traditionist stance was still represented a century later by Ibn Bbya Cd. 381/991-2). It
is probable that a spectrum of positions from "MuCtazill"
to "ZhirI' subsisted within ShPism throughout the tenth
century and beyond.
work, in the legal field at
least, amounted to a moderate and tolerant synthesis of
Traditionist and Rationalist views. That synthesis however
further bore witness to the extraordinary force within the
Islamic community at large of normative Sunni legal methodology. Both ImmIs and Mu'tazi1s were faced with the problem of adapting a law based art revelation to a system amply
and generously provided for by reason and both in the end
submitted to the assumption that revealed law (sam'iyyt)
can be justified only by reference to the four long-accepted
canonical sources.2
The actual dominance of the principle of revealed
law in spite of the, to them apparent, competence of reason
was noted by the Islamic philosophers who accounted for it,
most plausibly, on the grounds of practical efficiency. For
them a prophet was one who had gained insight into the highest level of intellectual reality but who gave expression to
that reality, by virtue of a strong imaginative faculty,
symbolically. 25 Intellectual truths found expression in the
form of vivid and potent symbols capable both of convincing
and compelling. This figurization of truth Faziur Rahman
refers to as the Psychological Law of Syrnbolisation: both
23. Madelung, Imamism, 28.
2L4 . For the Mutazi1Is, see Hourani, Rationalism, 137-139.
25. Rahman, Prophecy, 30-36.

217

the prophet and the non-prophet have respectively a

psychological need to express or grasp the truth in symbolic form. That the symbolic expression was both more
comprehensible and more compelling than the intellectual
may be conceded and it provides precisely a motive for
relation of law to revelation: revelation was symbolic,
persuasive, compelling. As far as the philosophers were
concerned the symbols had been produced by one man, divinely
Inspired. In modern terms it would be more satisfactory to
consider that the community itself fabricated its own symbols, its prophet, its canon and its "mythology", as a
means of definition and self-expression; to demonstrate
the sanctity and salvation of the Muslim as opposed to other
communities. Some of the stages in that fabrication have
been demonstrated by Wansbrough.26 The purpose was partly
utilitarian: as Mary Douglas notes: 'the whole universe
is harnessed to men's attempts to force one another into
good citizenship." 27 To concede only that however would be
to under-estimate the extent to which psychological or
cultural demands for order and identity transcend the merely
functional.28 The functional nonetheless was the second
element isolated in Fazlur Rahman's analysis of the Islamic
philosophers as distinguishing a prophet from a mystic or a
philosopher. 29 Positive religions could be interpreted as
pragmatic movements designed to satisfy the masses and bring
them to order by eliciting their adherence to the symbols
and allegories of truth. Only a prophet could institute
such a religion and establish a divine law for he alone would
have behind him the authority of God, made manifest in
miracles. For the Sunni clerical class the divine law was
contingent upon God's command, which was the sole source
of ethical values: the shari ta was a direct expression of
divine will. For the philosophers the shari C a was "only"
a symbol or allegory of rational truth which existed quite
26. Wansbrough, Sectarian Milieu, 99-103.
27. Douglas, Purity and Danger, 3.
28. Cf. Levi-Strauss, Savage Mind, chap. 1.
29. Rahman, Prophecy, L O_ L18, 52-64.

218

independent of God's command: the allegory was useful

because its form and content were such as to subdue the
masses to order, and good because it brought the people
nearer to God. But the knowledge of God acquired rationally
was superior to that acquired through symbols. Hence
implicit in the pragmatic assessment of positive religion
was an esoteric and elitist view which recognised that some
at least of the community could go beyond the symbols to
the ultimate intellectual truth.
Recognition of the psychological and pragmatic
value of revelation helps to account for the Shi c i elaboration of a shari'a based on Qurt.n and Traditions. The
problem however remained of how to reconcile a Muctazili
theology - such as the Shi c a had on the whole adopted - with
a revelation-based law. The Mu'tazills and the philosophers
had each proposed some kind of solution and the Shii thinkers
of the Buyid period had begun the task of reconciliation
largely by Incorporating and modifying Muctazill ideas.
A11ma,as eminent a theologian as he was a jurist,brought
together the various strands of legal, philosophical and
Mu c tazili Ideas to produce a synthesis of outstanding genius
and a corpus of work of revolutionary import. His writing,
more than that of any other single scholar, laid the foundations for the combined jurisprudential and philosophical
achievements of the Shi Ca in later centuries and earned for
him the appellation, most learned, 'Allma.
In his Kashf al-Murd, a commentary on Iiair al-Din
T 1 ' Tajrid al_I c tiqd, 'Allma reaffirmed Shidi adherence
to the Muctazili principle of a rational ethics. Good and
evil (basan and qabii) are intellectual principles (amrn
c aqliyyn) defined as meriting or not meriting blame, a
definition shared with the Mutazili 'Abd al-Jabbr.3
C Al1ma, probably following Nasir al-Din si or earlier
Shi'i/Mutazili thinkers, provided for the subdivision of
liasan on purely rational grounds into four a1km, so that
the ethical values of the intellect corresponded precisely
30. Cf. Hourani, Rationalism,

39,

4-47.

219
with those of revelation. That had not been achieved at
the time of CAbd a1-Jabbr and TsI. 31 A mubi. act was
defined as one the perpetration or neglect of which did
not merit praise or blame. A wjib action was one
perpetration of which deserved praise while neglect deserved
blame; a mandb action, one perpetration of which deserved
praise while neglect did not deserve blame; and a makrh
action, which 'Abd al-Jabbr and TsI had failed to define,
was one neglect of which deserved praise while perpetration
did not deserve blame. 32 The objective content of those
categories was typically open-ended: charity, generosity,
filial piety, tyranny, lying etc.33
If the Intellect provided so much what precisely
was the scope and function of prophetic law? CAl1mats
answer is here translated in a slightly abbreviated and
schematised form. The question Is: What is the good of
prophecy?
la.

It provides support for the Intellect by means

of revelation In those matters wherein the
intellect alone is sufficient; e.g. the Unity
of God etc.

lb.

It permits the deduction of an ethical/legal

value (bukm) in those matters wherein the
intellect alone provides nothing; e.g. ritual
(al_sg?jt) etc.

lc.

Some actions are good and some are evil. Of

the good, there are some concerning which
the intellect alone can assess their goodness
and some concerning which the Intellect alone
is Insufficient. This is true also of the
evil. By means of revelation knowledge is
acquired of those good and evil things which
the intellect alone cannot know.

2a.

The human species differs from other animals

by being T political' by nature, madanl bI'l-tjab;
he requires many things In his life which can

31. Ibid., 3-; of. bove, 207-209.

32. Al1ma, Kashf, 2314-235.
33. IbId., 23 14235, 252-253.

220

only be achieved by co-operation and coordination. But anarchy is present in human

natures so as to bring about conflict which
is opposed to the wisdom of social cohesion
(liikmat al-ijtim).

Therefore there must

be a principle of cohesion subduing men to

social cohesion (fa-1 budd miri jmi yaghuruhum
C a1 1_ijtim c ). This is the sunria and the
shar c . Every sunna requires a law-giver to
establish it and its curbs. That law-giver
must be distinguished from his fellow-men by
a miracle wrought by God which leads men to
believe his claim, and inspires in them fear
of opposing him, and threatens them so that
they follow him. In this way order is
established and men achieve their highest
perfection.
2b.

Men are varied in their grasp of perfections,

in their acquisition of knowledge, and in
their possession of virtues. Some are free
from the need of any helper (mustaghnin Can
mu c win) by virtue of their strength of
spirit, the perfection of the comprehension
and the fineness of their talent to achieve
higher things, (al-ittil bi'l-umr al-aliyya).
Some are totally Incapable of this. Some take
up a medial position, ranging nearer or farther
from the two extremes. The usefulness of the
prophet lies In his making perfect the
deficiencies [in comprehension] in accord with
the various abIlities.

Paragraphs la to ic there relate to the MuCtazi1 theory

of the contingent and non-contingent law; paragraphs 2a and
2b to the pragmatic, psychological and elitist views of the
philosophers. The significance of the contingent law was
further elaborated in discussion of the "Brahmins'" puzzle:

314. Ibid., 271-273.

221
if the prophet brings a message in accord with the intellect is it not superfluous, and if in disaccord, must it
not be rejected? 35 Neither ShIs nor Nuctazills conceded
that revelation might contradict reason. As to the problem
of superfluity, revelation was useful in that it confirmed
what was known to the intellect or provided details which
while not required by were yet not opposed to the intellect,
details that is of ritual. Revealed details however were
definitely Inferior to rational ethics and served only as
aids to the achievement of rational good and the avoidance
of rational evil.
!an, if he is assiduous In perpetrating
revealed duties (wJibt samciyya) or avoiding
divine prohibitions (manhI sharCiyya) comes
closer to the perpetration of rational duties
and avoidance of rational prohibitions.
Those duties, In other words, which depend entirely on
revelation, are instrumental. Exactly the same system of
non-contingent and contingent (primary and secondary) law
was found by Hourani in the works of the Mu'tazilr cAbd alJabbr.6
Revelation in so far as it confirmed rational
ethics was also useful because it served to subdue the
masses and incline them to social order and co-operation:
that idea (2a above) was borrowed from philosophical sources.
The clerical class as custodians of revelation were naturally
therefore the political leaders of the community: philosophical theory and juridical furti t thus bolstered each other
in upholding the political/executive authority of the cu1am.
The distinction between the clerical class and the learned
class which in the ninth century had made the MuCtazill
theory seem to restrict the authority of the clerical class
had in any case by the thirteenth century largely vanished:
the clerics monopolised all branches of knowledge. The class
of non-clerical but learned men was in the later centuries
of Medieval Islam both small and relative to the clerics
35. Cf. Hourani, Rationalism, l32-134; Wansbrough, Milieu,
Allma, Kashf, 273.
135;
36. Hourani, Rationalism, 132-134.

222

uninfluential. cAllmats synthesis of NuCtazill and philosophic beliefs both accounted for and justified the exiSterice of a revealed law side by side with a rational ethics,
while bolstering the political, social and religious authority of the clerical class. At the same time his elitist
views (2b above) and his acceptance of the 'secondary t or
instrumental nature of the samciyyt paved the way for the
continuation and development of Islamic philosophical
thought In Shidi Islam.

223
CHAPTER IX
DOUBT AND PREROGATIVE
Abdication of certainty and assertion of prerogative
are both implicit in the term ijtihd. The concern for
certain knowledge which characterised ShIfl polemic literature in its formative period was displaced with the passage
of time by the recognition of doubt as an inalienable feature of divine law. Precise definition In the mature tradition was seen to transcend the abilities of the scholars
whose statements of the law were recognised as merely opin-.
ions (uniin), product of doubt and subject to variation.
While the general movement from knowledge (cilm) to opinion
(ann) was a development that took place over centuries,
the critical moment of transition for the ShI'a may be
situated in the lifetime of cAllma who elaborated for the
first time and definit vely a ShI C I theory of ijtihd.
That the desire for Cilm dominated the juridical
writings of Murta has already been suggested. The same
desire persisted in T 1t t Uddat al-Ul: that work began
with a discussion of what constituted t ilrn and did not
readily admit that choice, even when it was recognised as an
option, indicated doubt. It indicated rather two correct
assessments both of them object of knowledge not opinion.'
The only serious abdication of the search for knowledge in
T SI 'S work was common to him and Sunn writers, namely his
granting that although the khabar wliid did not produce
knowledge it yet required action. In polemic works directed
to Sunni readers as well as ShI ones however even that
mitigation was forgotten. Sunni furti' were depicted in the
introduction to the Mabs as the product of subtle speculation and recourse to such doubtful sources as giys and
ijtihd, whereas Shi'I furi' depended on Traditions from
the imm and other methods giving rise to knowledge not
mere opinion. Such polemic about sources however was not
reflected in any concomitant major differences in furu:
1. See above, (ikhtilf), 212-213; cf. 180, 192-193.

224
TusI's Mabst for all the fuss was similar in content to
and borrowed extensively from the Sunni tradition.
Underlying every work of ul and usually explicit
was the question of how and to what extent one had access
to knowledge of God's will, and, if knowledge did not exist,
then how far opinion (ann) could be a basis for taklif.
Ul works were naturally works about the nature of knowledge
and as such generated dispute on precisely that level without
more than marginal concern for the effect of such dispute
They were concerned not simply with
(never much) on furii c .
justification of the sharica as it existed, but also with
justification of the very existence of the sharl C a: it
was in ul literature that the fugah' asserted or 'proved'
that there was in fact a manner of gaining knowledge of
God's takllf. Logically prior to the arguments of the usl!s
were the arguments of the theologians who dealt with God's
existence, Mul2ammad's prophecy, the imm's infallibility
and so on. The distinction however was merely a literary
convention: theological and juridical hats were worn at
lim. Associated with the
different times by the same
assertion that knowledge of the shari ca could be obtained
was inevitably an assessment of who possessed that knowledge and who therefore possessed authority, (for, that
knowledge gave authority was one of the unspoken assumptions
of all Islamic clerical writing). It was that question,
not worries about the substantive content of fur', that
provoked the intensity of uli dispute. It was freely
conceded that correct conclusions might be reached by the
wrong paths: but it was precisely the paths which mattered.
As TsT pointed out: "he who cannot judge the ul of the
shari ca is only a 1kI or a muqallid not an Clim, and
that is a position which the people of excellence avoid."2
That "it is more often the derivation than the substance
of halakha that distinguishes the historical schism" is
2.

Ts, CtJddat al-tlstil, 2. Cf. the distinction between

true belief and 'ilm noted by Hourani, Rationalism, 17
and n. 11.

225

not in fact surprising: it was by the nature of its ul

that a community defined itself and its magistrates.

The

issue was rarely what to do but rather what (remote)

authority to acknowledge: correct acknowledgement was the
basis of salvation and the source of immediate and tangible
authority within the community. T11 assessment of what
gave rise to knowledge and what required action was in the
long run such as to exclude both Sunns and Mu'tazills from
attainment or appreciation of either: only ImrnI tawtur
gave knowledge and only Imami khabar wid required action.
Only the ImmI fuqah' possessed knowledge and authority.
The ignorance and delusion of the heterodox persisted In
spite of subsantive agreement in fur, because of their
wrong sources: they were Ik and muqallid at best, not
lIm. Intense and lengthy arguments about the necessary
existence of the Imm and his universal authority, even
centuries after his absence, retain significance and vitality as the first stage in the definition of a community and
a system of authority and salvation.
The polemic confrontation between Sunni, Shi t i and
Mu c tazili declined towards the end of the Buyid period and
became relatively unimportant thereafter. During that
period however the exigencies of dispute imposed a rigidity
of definition which made the law seem more certain than It
actually was. T 1 displayed some concern to mitigate that
rigour by defending the use of the khabar w]iid and emphasising areas of choice and Ikhtilf. That however had to
be reconciled with the established slogans of Shr

uii1:
no qiys, no ijtihd, no khabar whid. All of these had
been rejected by the ShI c a on the grounds that they represented an abdication of certainty.
si had to balance the
need for the introduction of doubt and the provision of
flexibility against the demands of continuity. It was
mostly done by sleight of hand. In the very work wherein
he had argued extensively on behalf of the khabar wld as

3.

The quotation is from Warisbrough, BSOAS, XLII, I, 1979,

14l_1L2.

226
source he stated:
Action based on the khabar wiid is not
valid as we have already explained.4
That was not hypocrisy or contradiction but a
product of his own interpretation of the khabar wiid and
the safeguards he introduced against its misuse. What he
desired to achieve was the preservation of a slogan (continuity) with a change in its practical application (flexibility).
A similar process is evident in T 1t attitude to ijtihd.
The established slogan was preserved: no ijtihd. 5 But
something remarkably similar designated choice was introduced.6
Qiys was a method of rendering a fixed text open-ended in
application.
following established Sh1 c 3: practice,
denied all forms of qiys which involved extrapolation of
an Cilla, even where the cilia was specified in the text
(man llayh); 7 though he accepted the arguments a maiore

TsI,

ad minus and a minore ad malus which had been classified by

Shfi'I as 'qiys. 8 In practice the ShI'a provided themselves
with other exegetical devices for dealing with revealed texts
which enabled them to do without giys. Generally the preservation of slogans while modifying to a greater or lesser
degree their content was a characteristic means of preserving
the appearance of continuity while achieving some development.
The slogans, no ijtihd, no qiys, no khabar whid, remained
as fundamental elements of ShI
polemic, (an increasingly
desultory effort as creative rivalry between sects decreased)
14

Uddat al-Uil, 230.

5. Ibid., 59, 290ff.
6. See above, 180, 192-193, 212-213.
TsI, tuddat al-tJi, 254ff.
7.
8. See Schacht, Origins, 124; C Allma later accepted qiys as
long as it was rnan
al_Cilia: Mabdi', 218-219. Wansbrough's
doubts about "explicit ratio legis" (= al'i1ia al-man
ilayh)(Milieu, 195,n.2) arise from the assumption that giys
is a term Indicating precisely analogy such that the common
c ilia of the
and the farc is deduced not given. That
reasoning on the basis of an explicit cilia was subsumed under
the term giys is clear from TsI's discussion (9Jddat, 258).
A rather more stringent view might possibly be detected in
C A11ma: "The best view in my opinion is that a liukrn based
on an explicit i1ia may be applied to all matters wherein
not by qiyas."
that Cilia is known to subsist, this, by na
(Mabdl', 218). For kiasan ibn Shahid II qiys depends on a
common Cilia which may be either mustanbia or mansa
(?4atllm, 223).
sT,

227
until the time of Muliaqqiq and tAllma.9
With the advent of the Saljuqs, the removal of
T T to Najaf and the cessation of meaningful sectarian
debate, there came to ShI c ism a period of intellectual
stagnation designated by Shidi writers, a period of taglid,
by which they meant taqild of TsI. In effect problems of
continuity, flexibility and ikhtilf were obscured by a
general uncritical submission in legal affairs to the views
propounded by him. The intellectual revival of Shlcism
began with the work of Ibn Idris al-Hilir and may be related
to the increasing political liveliness of the ShIa in the
late Saljuq period, especially during the time of the caliph
al-Nsir (d. 622/1225) during whose reign Ibn Idris died.
In the introduction to his Sar'ir al-Islrn Ibn Idris
presented the apathy and ignorance of the ShT'a as the
motive for his producing that work, which contained in the
event a broad and frequently petty re-interpretation of
details of the sharl t a and an intense attack on
The
attack was characterised by an unremitting critique of
use of the khabar w1id and an appeal from Ts's interpretation of u1 to Murta's. His detailed justifications
of sharti precepts involved frequent dismissal of Traditions
as khabar wiid and a repeated and idiosyncratic application
of the principle of bar'at al-dhimma. His work, introduced
with an attack on taq1d, was concluded with a peroration
on the same theme, culminating in a striking plea for
flexibility and development:
It is not fitting for one who has understood
those who have preceded him in certain affairs
that he consider that superiority belongs to
them... for they err where they err because
they have exhausted their minds and used up
their time on other things, then they have
turned to those things wherein they erred with
minds worn out and spirits ravaged and little time.
Those who come after them benefit without effort
9. See e.g. Qazwrni, Naq, 1414, 50; RzI, Tabirat a1-'Awmm,
191-192.

228

from the achievements of earlier thinkers . -.

and so being in full possession of their
faculties and with plentiful time and unburdened
by weariness may notice what their predecessors
did not notice and consider what they did not
consider. Hence it is that the muta'akhkhirin
are superior to the mutacsddimIn and knowledge
Increases with the increase of men and the
progress of time...
That plea for progress and development found general acceptance amongst later writers 10 but was initially perceived,
In conjunction with the attack on Ts, as disruptive of
continuity.
Disruption prompted mediation and harmonisation,
tasks undertaken and successfully carried through by Muliaqqiq
and 'A1lma. Muhaqqiq's efforts in that direction are
exemplified in his Nukat al-Nihya 11 , a commentary on
work in which he justified against the critique of Ibn Idrls
the anomalies, contradictions and Isolated opinions which
the latter had found there. His general concern to heal

TsT's reputation

and to establish a compromise position

Is further evident in his attitude to the khabar wiid. He

asserted with regard to this problem that there were extreme
Tradltionists (Iiashawiyya) who accepted all khabar wid
irrespective of their contradictions, and there were the
opposite extremists who rejected all khabar whid. But
tawassut was best. Tawassut meant acknowledgement of those
khabar wid which had been accepted by the companions (ImmI
fuqah') or confirmed by indicators. 12 That was a modified
restatement of TsI's view. But perhaps the most striking
feature of Muaqqiq's discussion of usl is the evidence it
provides of the impingement of doubt as to the content of
the divine law. The Muctabar is addressed to a certain M.
Ibn M. al-Juwaynl, apparently both scholar (muqarrir qaw'id
10. E.g. It is quoted In al-Iurr al-mili, Amal, I,

7-8.

11. Contained in KhwnsrI, ed., a1-Jawrni al-Fiqhiyya.

12. NuIaqqiq, Muctabar, 6, ii. 3-8.

229
al-Imn) and official (liib diwn al-mamlik etc.) and
begins with a section on advice (way).
I warn you against the Iashawiyya and the
mugallida of the scholars, for they may cheat
you in order to attract you into their ignorance... Let it not worry you that such a one
should say: The truth is visible, there is
no doubt. , the way is clear... For it is not
difficult to answer him: All that may be
known may validly be described as clear even
if the path is hard and the [ascertainment of]
truth difficult... for if you considered the
variation [of opinions] amongst the excellent
in legal matters, that would indicate to you
the difficulty of achieving [an answer] without study and reflectlon...*3
Hence if a matter seems to you distinct and
certain then express It, otherwise cling to
suspension (tawaqguf) for that is the shore
of perdition.

j L

1
Clearly, for Muhaqqiq, the shari t a was not a structure,
easily known, of order and stability; it was uncertain
shifting and doubtful. The awesome nature of the muftPs
responsibility as articulator of divine law, coupled with
the warnings discovered in Q2.169 and 10.59 were brought
forward as Incentives to caution on the part of the mufti.
To declare as certain that which has not been established
to be so Is "to Invent a lie concerning God." 15 That the
law was doubtful and that scholars should be cautious in
expressing their views or should suspend decision (tawag)
13. IbId., 3, 11. 25ff.
11. Ibid., Li. , 11. 1-2.
15. Q. 10.59; for this argument see Muhaqqiq, T4u'tabar, 'i., 11.
1_Li..

230
were opinions that reflected Muhaqqiq's careful mediation
of past and present, his justification of ikhtilf, his
balancing of continuity and flexibility. The disadvantage
of his position was perhaps that suspension of decision,
the shore of perdition, represented an abdication of
authority and a withholding of guidance.
If Muaqqiq 1 s work betrays the impingement of
doubt, Allma?s represents its establishment and exploitation. His achievements in juridical theory may be summed up
by reference to two spheres of innovation. First he proposed a new (within Shi d ism) terminology for the organisation
and evaluation of Traditions. Second, he adopted the
technical terminology of ShI C I u1 in such a way as to
render its central principle that of ijtihd. The essence
of his conclusions there was that the actions of ordinary
mukallafs (muqallids) could and should be based on the mere
opinion (ann) of mujtahids. Both innovations were essentially terminological and both owed much if not everything
to Sunnl tradition.
According to M. Sdiq Bahr al-'Ulum cA1lma was a
distinguished writer in the science of iadth but his writings on that subject have been lost. It is nonetheless
known that he was the first Shi c t to adopt a technical
terminology for the division of Shi c i Traditions into the
categories of sahTli, asan, muwaththaq and ca'If, all of
them well-known from Sunnl sources) 6 It is clear that
'Allma was trying to deal with the vexed question of the
khabar w1i1d and that he felt the introduction of more
subtle distinctions would help. From the point of view of
certainty the results of his analysis were apparently depressing: it Is possible that his low estimate of Shici
Traditions may account for the disappearance of his works
in this field. MIrz Mu1ammad al-AstarbdI (d. 1036/1626-27)
whose work entitled al-Faw'id al- .-Madaniyya became an early
expression of the AkhbrI position accused cAllma of
claiming that most of the Shi d I Traditions were ghayr aIiI1.17
16. Balir al_ c Ulim, Faw'id, II, 260, (commentary).
17. Astarbd, Faw'id, f3a.

231
CA1lma himself in a passage made available in Iasan ibn
Shand II's Macaiim stated:
As to the imamiyya, the akhbaris of them
have relied In u1 al-dIn and fur' al-din
only on akhbar abad; and the ulis such as
and others have agreed in accepting the

T1

khabar wid. None have denied it except Murtand those who followed hIm.18
a
It appears that 'Allama understood most of the ShI d I Traditions to be aIa. Some clarification of what he meant by
this may be had from his definitions of tawatur and aId.
In defining the former he rejected the idea that it depended
on a specific number of transmitters, and claimed that the
only significant factor was whether It did or did not produce
certainty.' 9 Likewise a khabar wiid was one which gave rise
merely to ann, irrespective of the number of transmitters.2
18. Flasan Ibn Shahld II, Na'1im, 191.
On the vexed question of the origins of the Akhbari and
uii controversy there remains much study to be done.
It is my belief that the terms used in this passage do
not have a technical and "party" significance but a much
vaguer sense of "Traditlonist" and "jurisprudens". I
think the Akhbri movement as such Is best understood as
beginning where the Shi d i historians see its beginning,
with the attack of Astarabadi on cAllamats innovations.
The Akhbrls defined themselves against primarily Allrna
and did not begin to do so until the time of Astarabadi.
Buyid, Saljuq, Mongol and Mamluk texts occasionally
refer to ulIs and akhbarrs but to what extent these
can be aligned with UslIs and Akhbaris is not clear.
But see also Madelung, Imamism, 20-21.
19. t Allma, Mabadi', 201-202.
20. Ibid., 203. Cf. for a similar Sunn view, Aghnides,
Finance, 39....144.

232

In this way c Allma made the experience of the community

the major criterion for the assessment of the value of revelation. It was largely then the evidence of the community's experience, the cumulative evidence of ikhtilf and
non-certainty, that led cAllma to his pessimistic assessment of ShT'I Traditions.
Most of the law, for cA1lma, was in a state of
doubt. He found no difficulty in providing an elaborate
argument for the use of the khabar wiid as a 1ujja, based
this time on Quranic sources, as well as ijm, and much
more sophisticated than either Tsi's or Muliaqqiq's arguments. 21 His system as a whole freely acknowledged a
complete lack of certainty in the law. This feature of his
thought may best be illustrated by quotation. At the
beginning of his Tandhib al-Wul 1l 'urn al-U1 he defined
the word ugh.
JJ J'

t4

1VJt

LJ

;L)1i

LL

j,

f,

cj-i'

);UL

r
JI LL,
.W\

Figh is a word meaning understanding. By

conventional usage It means knowledge of
detailed sharci values the status of thich
is inferred In so far as their being part
of din is not known for certain (darratan).
Hence [the following] are excluded: knowledge
of essences, of rational [as opposed to
revealed] values, or that ijmC or khabar wid
are a Iujja, and the knowledge of the muqallid
and the certain usili (al-ul a1-darrIyya)
such as sa1t and zakt.22
21. 'A1lma, Mabdi', 203-205.
22. Id., Tandhlb, (1501) f. 2b-3a; (1502) f. 2b.

233
A similar definition was given at the beginning of the
Ta1rIr al-AIikm. There too one or the exceptions, not
to be included in fiqh, because it was part of tarrI
knowledge, was knowledge of ul al-sharIa. 23 By that
term he meant things "such as alt and zakt." tinder the
heading, fi l-mujtahad fhI, that is, the sphere of iJtihd,
he said:
It is every Iukm shar d , on which there is
no daili qaI; so it excludes rational
values; and [the existence of] a dalil
qEtic excludes [also] those elements of the
sharc which are known arratan such as the
incumbency of ai gt and zakt.24
What emerges from this is that, for 'A11ma, knowledge of
the law was of two types. There was an immediate necessary
knowledge (arrI) which related to what I would call
structural elements such as the incumbency of a1t and
zakt. This knowledge was provided with definitive
indicators, dalli gaI or dalil gijL The other type of
knowledge, devoid of definitive indicators, concerned
detailed shard values, (al-aIikm al-shar'lyya al_fardiyya)
and was the prerogative of the fuqah. It is to be noted
that since he defined fiqh as a word relevant only to the
uncertain type of knowledge, which was also the sphere, of
iJtihd, the terms faqTh and mujtahid would be synonymous.
It is clear that this division is precisely the
same as was enunciated by Shfi'I, who used, to express that
division, the terms dim Cjp
and Cilm khsa. The latter
type of knowledge for ShfiCI was the preserve of the fugah'
and was the sphere wherein there could be no certainty.25
CA11ma t s debt to Sunni models was acknowledged by later and
frequently hostile (Akhbri influenced) Shi'I sources, but
he himself made no secret of it: his discussion of itihd
contained at least one reference to ShfiI's argumeritatiori.26
23. Id., TarIr, 2.
24. Id., Tandhlb, (1501) f. 205a-205b; (1502) f. 309a.
25. Above, 16-19, 194, 201.
26.

Allma, Tandhlb, (1501) 1. 203b.

23L
The overall conclusion of tA1lmaIs observations was that
whereas there was a basic structure to the law, known
bi'l-arira, the details were varying uncertaln and matters
merely of opinion.
It remains to clarify what was meant in this
context by artrI. That term normally means direct or
immediate knowledge gained either by sensual observation
or intellectual ratiorialisation. According to tAllma a
khabar mutawtir gives rise to dim arirI. This is based
on the fact that:
Our certainty as to the happening of great
events, such as the existence of MuJ:iarnmad,
is no less than our knowledge that the whole
is greater than a part and other basic
principles [of arrI knowledge]. Further,
this knowledge is present amongst the masses
(al-. Cawmm) and those who do not practice
inference (istidll) so it is not subject
to doubt.27
What CAllma meant then by a darrI knowledge in the context
of the revealed law was a knowledge based on tawtur and
found amongst the masses. It is to be noted that though
there would be ijm c also on such knowledge, the ijmc
would be construed as posterior to and a result of the
tawtur. This tawatur/qarrr knowledge was found on such
things as the incumbency of prayer and zakt. The other
type of knowledge was an inferred (mustadall) and uncertain
(arinI) type, the exclusive preserve of the tulam.
Ijtihd was defined as a word indicating expenditure of effort by a faglh in order to acquire an opinion
(zann) on a shard value (1ukrn).2B Neither the prophet nor
the imms were permitted to act on the basis of ijtihd
for they were capable of gaining 'jim, in which case action
on the basis of ann was not permissible. The culamv
however were permitted to deduce annI values by ijtihd.29
27. Id., Mabdi', 199-200.
28. Id., Tandhib (1501), C. 2Oflh-202a; Mabdi', 2LO.
29. Id., Tandhib (1501) C. 202b-203a; Mabdi', 214O24l.

235

The statement that kull mujtahid muib which had been a

particular object of Tsi's denunciation was still rejected
as a characteristic Ashcarl belief: in fact, God had only
one answer for each problem. But the erring mujtahid was
understood to be forgiven. Ijtihd in effect was a reasoned
effort to state the law, and wrong conclusions were not
subject to penalties. 3
The process however was undertaken seriously and piously. In arriving at his conclusion
it was necessary that the mujtahid be able to demonstrate
a justificatory methodology, based on revelation or exegetical technique. If he reached only one view free of
opposing views (muCric) then action upon it was incumbent;
If there was an opposing view and if one of the two was
clearly rjili then action was incumbent on the basis of
the rjib; if there was no visible criterion of preference
(tarjlb) then it was a matter of choice (as for Ts) or
tsqut (as for Murtad: that Is, both lapse and recourse
to reason.) In any case a juridical value (1ukm) was
established and any one who neglected it was a sinner:
$%

1?

-' '''
31
Insistence on the establishment of a juridical rule which
could and should be acted upon was C A11ma ? s answer to
problems of abdication of authority Inherent In Mu1aqqiq's
advocation of suspended decision (tawaqquf). CAl1ma?s
perception that the law was based on unn was combined with
a perception that guidance was necessary: although the
decision of the mujtahld was based on unn action in accord
with it was necessary and he who abandoned action was a
sinner. The essence of the system was expressed In a pithy
axiomatic form probably coined by CAllma and adduced by
him at the beginning of his Tandhlb, and again at the
beginning of his TaIirIr al-Alikgm.
1
d}''

30. Id., Tandhlb, (1501) f. 205b-206a; MabdI' , 2414_2145.

31. Id., Tandhlb (1501) f. 206a; Mabdi', 25-246.

236
A method based on opinion is not inconsistent
with a juridical value eased on knowledge.32
That paradox conceded that the mujtahid's sources were
matters of opinion and his conclusions therefore susceptible to variation but the conclusion once given was "known"
and action in accord with it was necessary.33
The way in which Shii acceptance of ijtihd was
accompanied by a general acknowledgement of doubt, if not
a complete abandonment of certainty, may be illustrated
from }Iasan lbn Shahld II's Malim al-Din. He presented a
series of arguments of by then well-established Import
explaining why tawtur Traditions were a source of knowledge.3
He then raised the question whether the khabar wiid could
be used as a basis for defining the law. 35 A positive
answer was provided by exegesis of the two Quranic passages
formerly adduced by
(9.122 and 9.6), by appeal to
iibq qudam' al-a1b, a prudent terrninological substitute
for jjmgc3G, and by the following argument:
The sphere of definitive knowledge of those
aIikm shariyya which are not known for
certain in din and the Shi'i madhhab is in
our time cut off. For, that which is found
32. Id., Tandhib, (1501) f. 3b; Talirir, 2.
33. The rather abrupt appearance of the term ijtihd after
centuries of Shi'i rejection of this term requires some
explanation. It seems probable that the term had come
into use in a non-technical and undeveloped way after
the time of Ibn Idris as a retroactive justification of
decisions. Muhaqqiq uses the term casually (e.g.
; and
above, ad annulment of judicial decisions, 101
Mutabar, 7, line 10) but not as part of his exposition
of the technical apparatus of u1. The way was thus
eased for C Allma t s development of the term; and his
writings on the subject were presumably preceded by
years of teaching and polishing.
34. Iasan ibn Shahid II, rTa1im, 183-185.
35. Ibid., 188-192.
36. See his stringent remarks on ijm, above, 198-200.

237
in the indicators of juridical value
(adillat al-a1ikm) produces only ann
because of the lack of mutawtir proof,
and the absence of any method of ascertaining ijm other than transmission of khabar
wiid; and because it is evident that ilat
al-bar'a produces [also] only ann and the
Book is anni a1-da1la. If the cutting off
of knowledge (ilm) be granted with regard
to the Iukm shar'I then takllf must be by
ann. 37
That passage concedes that there can be no certain knowledge
about matters not already established and accepted, and
that all available sources of knowledge give rise only to
opinion. Even intellectual principles (ilat al-bar'a)
produce only ann and the Book (though mutawtir of transmission) was only annI al- . dalla.
This establishment and
confession of doubt as part of the essential character of
divine law reflected the need to provide for the harmonious
integration or at least peaceful juxtaposition of past and
contemporary attitudes, while providing also for development
and flexibility. But it reflected something beyond that.
For Islamic scholars the shari c a was not so much a law to
be followed - though it was certainly to be aimed at - as a
meditation upon an unattainable ideal. The ideal could
hardly be defined let alone put into practice.

8 The minute

and detailed elaboration of doubts was an expression of

piety and worship. The intangible and indefinable law was
a reflection of the transcendent and ineffable deity.

Ijtihd was a terininological innovation which,

by naming, justified doubt and uncertainty, but it justified
37. Hasan ibn Shahld II, Malim, 192.
38. Cf. Becker's remarks, cit. Wansbrough, On Recomposing,
l63-l6; also Coulson, Doctrine and Practice, 219.

238
also a distinction between the clerical class and the rest
of the community. That distinction with all that it implied
of authority and submission had been elaborated by the
Sunnls as early as Shfi'I's Risla. Shfi
was re-expressed by

CA1lma as

CIt s

Cjlm a1-khsa

uncertain (not qaI), Inferred (mustadall), doubtful (annr) and ijtihdT. It was
this knowledge which was the subject of figh and the preserve of the fuqah'. The acknowledgement of ikht1lf
within the sphere of that knowledge rendered this theory
one of the corner-stones of the charismatic community; it
was inerrant because individuals, appropriately qualified,
might differ. Opinions, once offered and justified, could
not be rejected but were preserved and recorded assiduously
against the day they might be useful.
The right to express
an opinion depended on knowledge which was the essential
quality separating the e lim from the mmr, and accounting

39

for the authority of the former over the latter.

Though the ShI'a initially avoided the term
ijtihd, the distinction between the
mmT and the C1im
was already implicit in one of Kulayni's Traditions.
Ab 'Abdullh was sitting in the circle (of
students) of RabIa, when a beduin came and
asked Rabl e a a question. Rabi a answered.
When he finished speaking the beduin asked him:

is it your responsibility (a-huwa fT Cunqika?).

Rabl e a did not reply. The beduin repeated the
question but received the same response.4..
[Finally] Ab
said: it is his

CAbdu1lh

responsibility.
The chapter heading for that was: bb inna al-mufti dmin.
The point was that the mustafti had only to obey; responsibility for right or wrong lay with the muftl. The
implications of that Tradition were confirmed in

C Uddat

sI's

al-tJtii. In that work he asserted that there were

two views on the question of a person incapable of istidll

39. Cf. Wansbrough, Sectarian Milieu, 87, 91.

al-KfT, II, 358.
4O. Kulayn, Fur

239
(that is, an C!pfl, someone incapable of deriving the law
from revelation.)
(1)

It is related from the BaghddIs that such a one

may not follow by taglid a muftl but must merely
refer to him so that he may direct him to the path
of dim concerning theob1em in question.

(ii) The BarIs and most fuqah' say that the 'mmi may
accept the statement of the mufti
That which we follow is that it is permissible
for the 'rnmI... to follow by taglid the cijm
The division of society into muftis and muqallids was there
acknowledged. TsI justified his opinion by a reference to
precedent: the generality of the sect (trnmat ai-'ifa)
had always sought and received fatws from the Cuiamt.
Though permissible In matters of fur t al-f igh taglid was
not generally thought permissible in matters of ul aldin, albeit there were various qualifications to that view.42
As the fuqah' consolidated their position within
the community and bolstered it by giving expression to it
in fur
and utl, or by establishing real control of
judicial and other community affairs, their perception
became sharper of their unique status. Mu1aqqiq noted that
the constant appearance of new problems (IawdIth and mawnIc)
required a class which would specialise in tafagguh: this
he related to Q9.22 (after ShfIdi43)and to various Traditions
urging the quest for knowledge. A paeon of praise for the
learned was summed up by reference to a Tradition from CAll:
Study figh for the fuqah' are the heirs of the prophet.44
This preceded and led to his assessment of the uncertainty
of the law and his assertion that the mufti in giving a
fatw is mukhbir Can rabbihi wa-nig bi-lisn sharcihi.45
The markedly charismatic view of the clerical class that
'Uddat al-Uii, 293.
l.
42. Ibid., 2914; for a later synoptic survey, see Hasan ibn
Shahld II, Malim, 237-239.
43. See ShEfidi, Risia, 988 and 997.
44. MuIaqqiq, Muctabar, 3, 11. 1-2.
145 . Above, 186.

24O
emerged in Muhaqqiq's work is apparent also in tAl1maIs
writing, 6 and was institutionalised by adoption of the
theory of ijtihd. The essence of that theory was that it
was both permissible and incumbent for the muqallid to act
on the basis of the opinions (unn) of the mujtahid.
It is permissible for the Cmni to act by taglTd
(1)
on the decision of the muitahid in fur' al-shar',
based on 1awl nafara (Q9.122) wherein God has
rendered learning incumbent on part of the group
and permitted to the rest tag1id.1
It is incumbent on the 'mmI to act by
(ii)
taglid if he is incapable of ijtihd.8
In the matter of choosing a muftl/mujtahid the
mustaft/muqallid was required, it is true, himself to
exercise ijtihd. The criteria of choice were summed up
In the phrase bI-mashhad mm al-khalq and correspond
precisely to the criteria affecting the choice of a judge.
A mujtahid was Ipso facto a shar'T judge and capable of
giving judicial decisions. His fame and Influence depended
largely upon his ability to find favour in the eyes of the
people and lent certain 'populist' elements to the quality
of his leadership. The rnugallid was required as far as
possible to choose the more pious and the more learned;
where there was conflict between these two, preference was
to be given to the more learned. Oral or written cornrnunlcation of a rnujtahid's opinion to a muqallid was permissible.
The ijtihd of a dead tlim was not binding. 50 These considerations, all of them adduced by CA11ma, became the
essential elements of all subsequent articulations of ijtihd.
Though elements of ikhti1f grew up there were from CA1lmats
time to the early Safavid period few significant developments.51
L16. See e.g. t Allma, TalrTr a1-AIkm, 3_Li.
47. Id., Tandhlb (1501), f. 207b-208a.
148. Ibid., f. 208a-208b.
49. Above, 78-80, 82-85.
50. Al1rna, MabdI', 2147_2119.
51. For a later assessment of ijtihd see Iiasan ibn Shahld II,
MaClim, 235-242.

24l
In other words the parameters of things relevant to a
discussion of jtihd were so finely perceived by 'Allma
that later writers had little or nothing to add. There
can be little doubt that the perfection of his formulation
was due to extensive study in borrowing from and re-interpretation of Sunnl sources. His debt to Sunni sources was
later (in Safavid times) acknowledged and disapproved of.52
The qualities formally required in a rnujtahid did
not differ markedly from those adduced by

TsI for a mufti,

qI. 'Allma

and were the same as those required for a

thought they could be reduced to one consideration, namely

that he should be able to deduce a1km from dalil sharCiyya.53

Shahid I's longer assessment of those qualities may be given
in abbreviated form as being on the one hand typical but
also interesting in so far as it was purportedly deduced
from the famous maqbl l2adlth of 'Umar ibn Ianala. The
judicial delegation, the 'general' delegation and the
authority of ijtihd were all exegetically related to that
same badith.
Thirteen matters are to be taken into consideration with regard to the faqlh [muitahid].
They have been indicated in the maqbl of Cumar
ibn Ianzala from the imgm al-diq:
[given in full] .....
They are:
- Iman
- adla
- knowledge of scripture (kitb)
-
"
" sunna
-
ijma
-
kalm
H

VT

-
-
-

"
IV

VT

" ul [al-fiqh]
" lugha, arf wa-nahw etc.
nsikh and manskh, mulikam and
mutashbih,
hir and mu'awwal etc.

52. See al-Hurr al-AmilI, Amal, 89; Eairnr, Lu T lu'a, 31-32.

53. See CA1lma, Mabdi', 2Ll_2L2.

2112

- knowledge of jarti wa_taCdfi (biography)

-
"
" muqta
al-laf, lughatan,
t urfan, sharcan511
- that he understand from scriptural texts
the required end if there is no qarina and
the end intended if there is a qarina.
- that he have a good memory.55
The mujtahid in coming to his decision was required
to justify it as far as possible by explaining the grounds of
his preference (tarjlii).

The relevant factors were the

various exegetical devices mentioned in chapter VII above
and may be summed up as appeal to the 'independent' evidence
provided by the sciences of grammar, rhetoric, history, biography and iogic. 6 That these factors rarely if ever
represented any real methodological constraints on the
possible options of the mujtahid has been one of the conclusions of the present work.
What did represent a real constraint on their exegetical and legislative activities were those conventions
designated by Arkoun logocentricity. A particular example
may be found in the fact that all Islamic legal systems had
a common inalienable structure, corresponding to those parts
designated by CAl1ma as known bi'l-arira, by ShfidI as
'ilm tmrna.
That society consisted of a group of people
who performed purification and prayer, payed zakt and waged
religious war, fasted in Ramadn, bought, sold, married,
divorced, judged and carried out all such and other daily
activities consciously in the sight of God was an assumption
never questioned. This basic structure since it is presupposed both in works of figh and in the elements of revelation (largely in any case a later elaboration than substantive
law) must be conceptually and presumably chronologically prior
to both. That jihd is necessary "because of" Traditions

5)4 . ee above, 177-178.

55. Shahld I, Dhikr, 3.
56. For a full discussion, see, 'Allama, Mabdi', 230-238;
and Iasan ibn Shahid II, MaCaiim, 2142_2118.

243
would be the supposition of a Muslim but the concept of
jihd must have existed before the Traditions. Hence
the concept Jihd represents an ultimately irrational
but irreducible assumption about the structure of society
and may be assessed as part of the Islamic c1ture logocentrique, as can the whole of the basic structure of the
sharl C a. Beyond that structure, variations and options
were created or limited by a variety of social, political,
economic and personal factors, not easily, perhaps not at
all, quantifiable.
The clerical class, by exercise of ijtihd,
recorded their reactions to the experience of the community:
they created rather than uncovered God's law. What they
created was a literary expression of their aspirations,
their consensual interests and their achievements; what
they provided for Islamic society was an ideal, a symbol,
a conscience, and a principle of order and identity. To
assume that the process of defining the law was mechanical,
rather than profoundly creative, would be to underestimate
their achievement in defining a culture:
The life of the law has not been logic,
it has been experience.57

57. Justice 0. W. Holmes.

21414
APPENDIX
Dates and biographical details are given here of the Imml
writers whose works formed the basis of this study. Of the
many Imml works of clerical biography references are given
to only two: Ba1rnl's Lu'lu'at al-Balirayn (LB), a succinct,
handy, eighteenth century work, and Khwans grl's Rawt alJannt (RJ), a compendious, synoptic, late nineteenth
century work. The value arid function of clerical biography
have been briefly discussed in the previous pages, C 158-160,
165-167, 1814-187)
AL-KULAYN, MuIarnmad ibn Yacqib; d. 328/9140-9141 or 329/9 14 19142. A scholar of Rayy, he produced the first major standard
collection of Shidi Traditions, the Kitb al-KfT. (LB 38639 14 ; RJ 550-5514.) There were three other 'canonical'
collections of Traditions: one by Muliammad ibn tAll ibn
Bbya, al-Shaykh a1-adq, (d. 381/991; see LB 372-381;
RJ 550-5514) and two by

T1

AL-MUFID, Muliammad ibn Mulammad; d. Ramadn 14 13/1022. One

of the major Imml theologians of the Buyid period, he has
been much discussed by Western writers; e.g. Sourdel,
L'Imamisme, and McDermott, Theology. (LB 356-372; RJ 563570.)
AL-MURTAQ, 'All ibn Husayn al-Msawl; d. Rabl'I 1436/1014)4.
His father was influential at the Buyid court. He himself
took up official posts as nagib of the Alids, as amlr a1-Iajj
and as a judge In the Malim court. He combined the life
of a courtier and official with that of theologian, jurist,
poet and belle-lettrist. He studied theology and law with
Mufld and is generally portrayed as an active propagator of
the Imml faith. (LB 313-322; RJ 383-388.)
AL-TS, Muliarnmad ibn Ilasan, known as al-Shaykh; d. MuI?arram
1460/1067. He arrived in Iraq from
at the age of 23 and
studied first with Mufld, then with Murtad. He eventually
became the acknowledged leader of the Imrnl sect, a position

2145
which he retained even after his flight to Najaf, shortly
after the Saljuqs took Baghdad. He produced two collections
of Traditions and a number of works on fur al-f igh and
ul al-figh. The quantity and the subtlety of his juridical
work (though it contained a great deal of contradictory and
confusing elements) made him the dominant influence on Imml
legal studies for centuries after his death. The fact that
he survived the advent of the Saljuqs and established a
centre of learning in southern Iraq helps to account for
his influence. In the sophisticated, cosmopolitan, sectarian,
polemical Baghdad of Buyid times Mufld, Murtad and TSI
played an active part in public life: their debates with
the Muctazill, 'Abd al-Jabbr and the Sunni, BqillnI are
particularly well-known. (LB 293-30'4; RJ 580-591.)
Ab YaCl SALAR AL-DAYLAMI; d. Ramadn 1463/1071.
Muhammad ibn 'AlT al-KARJIKI; d. 1499/1057-8.
CAbd al-'AzIz IBN AL-BARRAJ; d. Shacbn 1481/1088.
These were significant minor figures of the late Buyld or
immediately post Buyid period. They were all pupils of
Murtad and TsT. Their juridical works reflect the influence of their teachers (especially sT) and include commentaries and small independent works. (RJ 201-203; 579-580;

3514-356.)
lEN IDRrS, Mul?ammad ibn Al2.mad a1-Ii11I; d. Shaww1 598/1202.
Of events in his life nothing is known, but the significance
of his intellectual revolt against the dominance of TsT's
influence in ImmT legal thought is much discussed. Hiz
criticism of T 1 made him in turn an object of some criticism
but he was finally accepted as a reliable thinker whose views
were to be considered. (LB 276-280; RJ 598-602.)
AL-MIJAQQIQ, Jacfar ibn Hasan al-Hull; d. abl'II 676/1277.
Of events in his life little is known other than a notable
meeting with Nasir al-DIn T1 This meeting suggests that
he was on good terms with the Mongol government which
established itself in Baghdad during his lifetime. On the
whole the Shi'Ts of southern Iraq probably benefited from

2)46

the advent of the Mongols. Muliaqqiq's fame rests primarily

on his work in furi al-fiqh, in particular on his abilities
as an organiser and a systematiser. (LB 227-235; RJ 1146
1)49.)

AL-'ALLAMA, Iasan lbn Ysuf ibn al-Mutahhar al-}illi;

d. Muharram 726/1325. He was an influential figure at the
court of Oljaytu, Khodbandeh and is reputed to have been
responsible for his conversion to the Immi faith. He was
frequently involved in public debate at the court and
composed a number of polemic works, one of which inspired
a refutation from Ibn Taymiyya. He studied philosophy and
theology with Nalr al-Din Tsi, and figh with Muliaqqiq.
He also studied Sunni juridical literature. He produced
works of considerable importance in all branches of religious
literature. (LB 210-227; RJ 171-176.)
AL-SHAHID AL-AWWAL, Shams al-Din ibn Makki; d. 786/138)4-5.
He lived in Syria during the reign of the Mamlk sultan,
Barqiq. He is reputed to have been put to death in Damascus,
after adjudication by two Sunni judges, as a result of his
writing a slanderous book about the Sunnls. His most famous
work of furii c , Al_Lum c atal_Dimashqiyya, was written at the
request of Shams al-DIn al-Awl, one of the chief ministers of
the Sarbedarid AlI ibn Muayyad. This request was made
when for a brief period it seemed possible that an ImmI
Shi d i state might be set up in Sarbedarid Khurasan. (LB 1)431148; RJ 517-522 - this should read 617-622 owing to pagination error beginning p. 605.)
AL-KARAKI, cAll ibn Husayn ibn C Abd a1cAll, al-Mu1aqqiq alThni; d. 937/1530-1 or 9)40/1533-14. He was one of the foremost ulam at the court of Tahmsp and became Shaykh alIslam. He is reputed to have enjoyed considerable power as
adviser to the Shah, propagator of the Shi'l faith and as
organiser of tax-collecting activities. Biographical works
stress and perhaps exaggerate the extent of his power and
authority under Tahrnasp. They preserve a Persian document
in which Tahmasp refers to Karakl as n'ib of the imms,

2147
and another in which he confers upon him a waqf (also,
soyurghl) on the lands of Iraq. KarakPs written works
include a great deal of ant-Sunnl polemic as well as
internal Shl:'1: polemic. (LB 151-159; RJ 14021407.)
AL-SHAH!D AL-THAN!, Zayn al-Din ibn cAli al_ c Amili; d.
Rajab 960/1559. Born in Syria, he travelled extensively
throughout geographical Syria and Egypt, to the Hijaz and
Iraq and, once, to the Ottoman capital in Constantinople.
While in Constantinople he gained an official post as teacher
in the Madrasa Nriyya In Ba'albak, where he taught for
several years. His execution is said to have resulted from
the complaint of a dissatisfied litigant, who complained
to the qi of Sidon. But the details of the events that
led up to his death are confusing. Like 'Al1ma and Shahid
I he was known, or perhaps notorious, for his study of Sunni
religious literature. His works on furti' al-figh, especially
his commentaries on earlier writers, were Innovative as to
their organisatlon, style and technique as well as in the
presentation of details. His works on util, notably the
Tarnhid a1-QawId, were also important and much admired.
(LB 28-36; RJ 288-299.)
}IASAN IBN SHAHID II, known as Shib a1-Malim; d. 1011

1602-3. Of the details of his life little Is known except

that some time after the death of his father he travelled
to Najaf. His only significant work was the Ma1im al-Din
on uti1 al-figh, but that displaced earlier works of the
same genre and became for some time the standard work on
Shi t i util.
(LB L 0, 45-50.)

2148
BIBLIOGRAPHY
ABEL, A., 'Le chapitre sur l'imrnat dans le Tamhtd d'alBqi11n', in Colloque de Strasbourg, Shltisme.
ABtJ YSUF, K. al-Kharj, Cairo, 1352/1933.
AGHNIDES, N.P., Mohammedan Theories of Finance, New York, 1916.
_ C.ALLAMA, Hasan b. Ysuf b a1-utahhar a1-}il1I, Kashf almurd fi shar tajrld a1-itiqd, Qum, n.d.
-ALLAMA, Mabdi' al-Wu1 i1
j1 a1-Ul, Najaf, 1970.
_ L ALLAMA, K. Minhj al-Karma fi ithbt a1-imma, Persia, 1296,
-'ALLMA, K. Mukhtalif al_Shr e a, (14 Vols. in 1)' , Persia, 1324/
1906.
_ C ALLMA, Qawa'id al-AIkm, in Karald, Jmic' a1-Magid (Vol.
II).
.-ALLMA, TandhTb a1-Wul ha 'urn a1-u1, MS. India
Office, Ar. II 1501 and 1502.
-'ALLrIA, Tal?rr al-Alikam (2 Vols.), Persia, 13114/1896.
ARISTOTLE, Ethics, (Tr. J.A.K. Thomson), Penguin Books, U.K.,
1955.
ARISTOTLE, Politics (Tr. B. Jowett), Oxford, 19142.
ARKOUN, M.,'Logocentrisme et vrit religieuse dans la
pense islamique' SI, XXXV 1972.
-ASHAR!, Al-ibana an ul al-diyana, in Maturidi, Shar1
al-Figh al-Akbar, Haidarabad, 1321.
ASTARBD!, Mirza Mul?ammad Amin, A1-Fawa'id al-Madaniyya,
MS. India Office, Ar. II 1507.
-BAGHDADI, CAbd al-Qahfribn Thir, U1 al-D3:n, Istanbul,
13146/1928.
BAHA'!, Muhammad ibn Husayn, Jami c i c Abbsl, Tehran, 1319.
BAHR AL-'ULM, MuIammad al-MandT, A1-Fawa'id al-Rujaliyya
(ed. Bahr al-Ulm)(4 Vols.) Najaf, 1385-6/1965-7.
BAHRNT, Ysuf ibn Ahmad, Lu'lu'at al-BaI)rayn, Najaf, 1386/
1966.
-BQILLNI, Mu1arnmad ibn al-Tayyib, al-Tamhid fi 1-Radd,
Cairo, 19147.
BARKER, E., Greek Political Theory, London, 1970 (5th Ed.).
BARKER, E., Principles of' Social and Political Theory,
Oxford, 1951.
BERNAND, M., 'L'ima chez tAbd al-Gabbar et l'objection

249
d'an- Nazzrn', SI, XXX, 1969.
'Nouvelles remarques sur 1'im' chez 'Abd alGabbar', Arabica, XIX, 1972.
BINDER, L., 'A1-Ghaz1I's theory of Islamic Government',

MW, XLV, 1955.

BINDER, L., 'The Proofs of Islam', in Makdisi, ed., Studies.
BRAVNANN, M., The Spiritual Background of Early Islam,
Leiden, 1972.
BRUNSCHWIG, R., 'Les us1 al-fiqh Immites a leur stade
ancien', in Colloque de Strasbourg, Sht'isme.
BURTON, J., The Collection of the Qur'n, Cambridge, 1977.
CHARDIN, J., Voyages en Perse et autres Lieux de 1'Orient,
(10 Vols.), Paris, 1811.
Colloque de Strasbourg (1968), Le Shi t isme Immite, Paris,

1970.
COULSON, N.J., 'Doctrine and Practice in Islamic Law',
BSOAS, XVIII, 1956.
COULSON, N.J., A History of Islamic Law, Edinburgh, 1964.
DOSTOEVSKY, F., The Brothers Karamazov, (Penguin, U.K., 2
Vols.)
DOUGLAS, Nary, Purity and Danger, London, 1978.
EBERHARD, E., Osmanische Polemik Gegen die Safaviden,
Freiburg, 1970.
ELIADE, M., The Myth of the Eternal Return or Cosmos and
History (Trans. W.R. Trask), Princeton, 197.
ELIADE, N., Rites and Symbols of Initiation (Trans. W.R.
Trask), New York, 1975.
ELIASH, J., 'The Ithntasharl_ShTCI juristic theory of
political and legal authority', SI, XXIX, 1969.
ELIASH, J., 'The Shicite Qurn', Arabica, XVI, 1969.
FRIEDMANN, W., Legal Theory, London, 1953, (Third Ed.).
GIEB, H.A.R., 'Government and Islam under the early
'Abbasids', in Colloque de Strasbourg, L'Elaboration de 1'Islam, Paris, 1961.
GIBB, H.A.R., 'Al-Mawardi's Theory of the Caliphate', in
Id., Studies.
GIBB, H.A.R., 'Some Considerations on the Sunni theory
of the caliphate', in Id. Studies.
GIBE, H.A.R., Studies on the Civilisation of Islam, (eds.

250
S.J. Shaw and W.R. Polk), London, 1962.
GLASSEN, E., 'Schah Isma'il und die Theologeri seiner Zeit',
Der Islam, L18, 1971-2.
GOITEIN, S., A Mediterranean Community, (Vol. II - The
Community), U.S.A., 1971.
GOITEIN, S., 'A turning point', in id., Studies in Islamic
History and Institutions, Leiden, 1968.
GOLZIHER, I., Muslim Studies, (2 Vols.)(Ed., S.N. Stern),
London and New York, 1967 and 1971 (refs. given
for German and English pagination).
hiris (Tr. W. Behn), Leiden, 1971,
GOLDZIHER, I., The
(Refs. given for German and English pagination).
GOODENOUGH, E.R., By Light, Light: The Mystic Gospel of
Hellenistic Judaism, Amsterdam, 1935 (1969 reprint).
HAIRI, A.H., Shi'ism and Constitutionalism in Iran, Leiden,

1977.
}ASAN IBN SHAHID II, MaA1im al-Din fi l-uiI1, Qum,n.d.
HAWTING, G.R., 'The Significance of the slogan l Iiukrna ill
11llh', BSOAS, XLI, 1978.
HOURANI, G.F., 'The Basis of Authority of Consensus in
Sunnite Islam', SI, XXI, l964.
HOURANI, 'Two Theories of Value in Medieval Islam', MW, 50,

1960.

HOURANI, Islamic Rationalism: The Ethics of 'Abd al-Jabbr,

Oxford, 1971.
-HURR AL-AMILI, MuI?arnmad b. Hasan, Amal al-mil (2 Vols.)
Baghdad, 1385.
lEN AB YACLA, AL-FARRA', M. b. M., abaqt a1-kIan.bila
(Abridged by M. b. CAbd a1-Qdir a1-N.bu1usi)
Damascus, 1931.
lEN AL-BARRAJ, Sa t d al-Din, al-Q41, Sharl2 Jumal al_Cilm
w'al-Amal, Mashad, 19714.
lEN IANBAL, AI-imad, Nusnad. (6 Vols.), Cairo, 1313/1895.
lEN HISHAM, Sirat al-Nabi (14 Vols.?), Cairo, 1356/1937 ff.
IBISH, Y., The Political Doctrine of Bgillni, Beirut, 1966.
lEN IDRTS, Muhammad ibn Ahmad, al-Hull, Sar'ir al-Islam,
Persia, 1270/18514 (n.p. - refs. are to British
Museum edition, where the beginning, hdh kitb
al-sara'ir, is page 2.)

251
IBM QAYYIM AL-JAWZIYYA, AC1m al_muwaggidin, Beirut, 1973.
IBM QUTAYBA, Ta'wIl Mukhtalif a1-IadIth, Cairo, 1326/1908.
FatwI (30 Vols.), Riyadh, 1381IBN TAYMIYYA, Alimad, Majm

83.
-KARAJIKI, Muiammad ibn cAll, Kanz a1-faw'id, [ashad, 1323.
-KARAK!, CAll ibn }usayn, al-Muliaqqiq a1-Thnl, Jmi' alMagid (fi sharl2 al- .Qawid), (2 Vols.), Tehran,

1395.
KEDDIE,The Roots of the ulama's power in Modern Iran',
Studia Islamica XXIX, 1969.
-KHWNSRI, Muhammad Bqir (ed.), A1-Jawrni' al-Fighiyya,
Tehran, 1276/1860.
-KHWANSART, Muhammad Bqir, Rawt a1-jannt, Tehran, 1306.
KOHLBERG, E., 'The Development of the Imml ShI E I Doctrine
of Jihd', ZDMG, 1976.
KOHLBERG, E.,'From Immiyya to Ithn-"ashariyya', ESOAS,

XXXIX, 1976.
KOHLBERG, E., 'Some miami Views on Taqiyya', JAOS, 95, 1975.
KOHLBERG, E., 'Some Notes on the Immite Attitude to the
Qurn', in Stern, Hourani and Brown, Eds.,
Islamic Philosophy.
-KULAYN!, Muhammad ibn Yaqb, Fur' al-kfl (2 Vols. in 1)
Persia, 1315/1898.
LAMBTON, A.K.S., 'Khi1fa', in El (ii).
LAMBTON, A.K.S., 'A Nineteenth Century view of Jihd', SI,

32, 1970.
LAtIBTON, A.K.S., 'The Persian tU1am' and Constitutional
Reform', In Colloque de Strasbourg, Sht'isme.
LAMBTON, A.K.S., 'Quis Custodiet Custodes: Some Reflections
on the Persian Theory of Government', I and II,
SI, V and VI, 1956-7.
LAMBTON, A.K.S., 'A Reconsideration of the Position of the
Marja c al-Taqlld and the ie1igious Institution',

SI, 20, 19614.

LAOUST, H., 'Les Agitations Religieuses . Baghdad aux
et e siecles', in Richards, ed., Civilisation.
LACUST, H., 'La Critique du Sunnisme dans la Doctrine d'alHull', REI, 311, 1966.
LAOUST, H., Essai sur les doctrines sociales et politiques
d'Ibri Taymiyya, Cairo, 1939.

252
LAOUST, H., 'La pense et l'action politique d'al-MwardP,
REI, XXXVI, 1968.
LECOMTE, G., 'Aspects de la Literature du Ijadtth chez les
Immites', in Colloque de Strasbourg, Shtisine.
LEVI-STRAUSS, C., The Savage Mind, London, 1972.
LEWIS, B., 'Islamic concepts of revolution', in P.J.
Vatikiotis, ed., Revolution in the Middle East,
London, 1972.
LINANT DE BELLEFONDS, Y., 'Le Droit Immite', in Colloque
de Strasbourg, Sht'isme.
LOUCEL, H., 'L'Origine du Language d'aprs les grarnmairiens
Arabes', Arabica, X and XI.
MADELTJNG, W., 'Imarnism and Mu'tazilite Theology', in
Colloque de Strasbourg, Shtisme.
PIADELUNG, W., 'Imma' in El (ii).
MAJLIST, Muhammad Baqir, cAyn al-Hayyt, Iran, 1373.
MAKDISI, G., (ed.), Arabic and Islamic Studies in Honour
of H.A.R. Gibb, Leiden, 1965.
MAKDISI, G., 'The Sunn Revival', In Richards, ed.,
Civilisat ion.
/
/
MANSOUR, Camille, L'Autorite dans la Pensee Musulmane,
Paris, 1975.
-NAWARDI, CA13: ibn Mulammad, Al-Alikm a1-Su1niyya, (ed.
Enger), Bonn, 1853; (trans. Fagnan), Alger, 1915.
MAllAOUI, M.M., The Origins of the Safavids, Wiesbaderi, 1972.
McDERMOTT, M.J., The Theology of al-Shaikh al-Mufld, Beirut,

1978.
-MUHAQQIQ, Jacfar b. Sa d id, al-Hull, A1-Mukhtaar a1-Nfi',
Cairo, 1376 (Qum reprint).
-MUHAQQIQ, K. al_Muctabar, Persia, 1318/1901.
-MUHAQQIQ, Shar'i a1-Is1m 11 mas'i1 a1-ha11 w'a1-1iarm,
(4 STols.), Najaf, 1389/1969.
-MURTADA, A1l ibn Iusayn, al-Sharlf, Jumal al- t ilm wa1_CAma1,
in Ibn al-Barraj, Shari.
-MURTADA, A1-Sh g fl fi 1-Imma, Tehran, 1301/1884.
-MAS'UDI, 'All ibn Husayn, Nurj al-Dhahab (14 Vols. in 2),
Cairo, 1384/1964.
-NAWAW, Yahy ibn Sharaf, Minhj al-TlIbln, in Shirbini,
Mughnl.

253
NEUSNER, J., Early Rabbinic Judaism, Leiden, 1975.
PATTON, W.M., AIrnad ibn klanbal and the Mi1na, Leiden, 1897.
PELLAT, C., 'L'Im.mat dans la doctrine de Ghit, SI, XV,

1962.
PETERS, J., God's Created Speech: a study in the speculative
theology of ... 'Abd al-Jabbr L Leiden, 1976.
PETERS, .R.S., 'Authority', in Quinton, ed., Political
Philosophy.
-QAZWTNI, CAbd ai-JalIl, K. al-Naq (ed. MuI?addith), Tehran,

1331/1952.
QUINTON, A., ed., Political Philosophy, Oxford, 1977.
FLAHMAN, Fazlur, Islamic Methodology in History, Karachi, 1965.
RAHMAtT, F., Prophecy in Islam, London, 1958.
R.zI, Sayyid Murtad, Tabirat al_ c awmm fi ma'rifat maqlt
al-anm (ed. Iqbal), Tehran, 1313/193k.
RICHARDS, D.S., (ed.) Islamic Civilisation, 950-1150,
Oxford, 1973.
ROSENTHAL, Political thought in Medieval Is1m, Cambridge,

1958.
SCHACHT, J., Introduction to Islamic Law, Oxford, 196k.
SCHACHT, J., The Origins of Muhammadan Jurisprudence,
Oxford, 1950.
...SHAFI'T, Mu1ammad ibn Idrls, Al-Ris1a, Cairo, 1388/1969.
(Refs. are to paragraph numbers not page numbers).
-SHAFI'I, Mu1ammad ibn Idris, K. al-Urnm, (7 Vols in k),
Cairo, 1388 (reprint of 1321 ed.).
-SHAHTD I, MuI?amma d ibn Makki, Dhikr al_ShT c a fT A1ikm
al-Sharia, Persia, n.p., n.d. (Also contains
Shahld II, Tamhid.)
-SHAH!D I, M. ibn Makki, a i _L umcaal_Dimashgiyya , (in
Shahid II, Rawca.)
a1-Arni1T, Masalik al-Ifham,
-SHAHID II, Zayn al-Din ibn
Sharb Shar?i C al-Islam, (2 Vols), Persia, 1310
and 131k.
-SHAHD II, Zayn al-Din ibn S AlT, al_ t Amili, A1-Rawqat alBahiyya F Sharl2 al-Lurn t aal-DimashgiYYa, Tabriz,
1271.
-SHAHD II, Tarnhid al_Qawa c id, in Shahid I, Dhikra.
SHIRB!NT, Muhammad ibn al-Khatib, Mughni al-Mul)taJ (FT shar

25)4
al-Minhj), (14 Vols.), Cairo, 1308.
SOURDEL, D., L'Imamisme vu par le Cheikh al-Nufid, Paris,

197'4; (also REI, XL, 1972).

SPERL, S., 'Islamic Kingship and Arabic Panegyric Poetry
in the early ninth century', JAL, VIII, 1977.
STERN, S.M., HOURANI, A., and BROWN, V., (Eds.), Islamic
Philosophy and the Classical Tradition, Oxford,

1972.
-TABART, M. ibn Jarlr, Ikhtilaf a1-Fuqah' (ed. Schacht)
Leiden, 1933.
-TABAR!, M. ibn Jarir, TafsTr al-Tabarr, Jmi al-Bayn,
eds., Shkir and Shkir, (16 Vols.), Cairo,

137)4-1388.
-TUST, Muhammad ibn Hasan, Kitab al-Ghayba, Najaf, 1385.
-TtSI, Muhammad ibn Hasan, K. al-Mabs, Persia, 1271, (n.p.)
Mabs, K. al-Zakt, 7 indicate
the seventh page of the K. a1-Zakt).
-TST, Muhammad ibn Hasan, K. a1-Nihya, Beirut, 1970.
-TSI, Muhammad ibn Hasan, K. Tandhb al-A1krn, (2 Vols.)
(Refs. to

Persia, 1317/18.
-TSI, MuIammad ibn Ilasan, Ta1kh

a1-Shfr, in Murtad,

ShafT.
- T S! , M. b. H., t uddat a1-U1, Tehran, 1314.
TYAN, E., Histoire de L'Organisation Judiciaire en Pays D'
Islam (2 Vols.), Paris, 1938/43.
VAN ESS, J., 'The Beginnings of Islamic Theology', in The
Cultural Context of Medieval Learning, eds., J.E.
Murdoch and E.D.S. Sylla, Dordrecht, 1975.
VAN ESS, J., 'Das Kitb a1-Ir' des kiasan b. Mulammad b.
a1-Ianafiyya', Arabica, XXI, 1974.
WANSBROUGHI.Quranic Studies, Oxford, 1977.
WANSBROUGH, J., The Sectarian Milieu, Oxford, 1978.
WANSBROUGH, J., 'Arabic Rhetoric and Qur'anic Exegesis',
BSOAS, XXXI, 1968.
WANSBROUGH, J., 'Majz al-Qur'an: Periphrastic Exegesis',
ESOAS, XXXIII, 1970.
WANSBROUGH, J., 'On Recomposing the Islamic History of
North Africa', JRAS, 1969.
WATT, W.M., 'The Reappraisal of 'Abbsid Shiism', in

255

Makdisi, ed., Studies.

WATT, W.M., 'Reflections on A1-Ghazll's Political Theory',
Glasgow University Oriental Society, Transactions,
21, 1965-6.

WEBER, Max, Max Weber on Law in Economy and Society, ed.

Max Rheinstein, tr. Edward Shils and Max Rheiristein,
Cambridge, 1assachusetts, l95.
WENSINCK, A.J., The Muslim Creed, London, 1965.
WINCH, P., 'Authority', in Quinton, ed., Political Philosophy.
-ZAWZN, Sharli al_Muca11agt al-Sab Beirut, 1966.