Bernard Weiss - The Theory of Tawatur According To Ghazali

Maisonneuve & Larose
Knowledge of the past: The Theory of "Tawtur" According to Ghazl

Author(s): Bernard Weiss
Source: Studia Islamica, No. 61 (1985), pp. 81-105
Published by: Maisonneuve & Larose
Stable URL: http://www.jstor.org/stable/1595409 .
Accessed: 27/12/2013 13:51
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Maisonneuve & Larose is collaborating with JSTOR to digitize, preserve and extend access to Studia Islamica.
http://www.jstor.org
This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

All use subject to JSTOR Terms and Conditions
KNOWLEDGE
OF THE PAST :
THE THEORYOF TAWATUR
TO GHAZALI
ACCORDING
Despite the great emphasiswhichmodernsciencehas placed
upon empiricalinvestigation,it still remainstrue that a great
part of what the average man would call his knowledgeof the
worldis not deriveddirectlyfromhis own experience. A native
Nebraskanreadilydeclaresthat he knowsthe earthto be round
even if he has nevertravelledoutsidehis home state. He also
readily declares that he knows that Chicago is located on a
lakeshore, that Russia has a communistgovernment,that
the Confederateswere defeated in the Civil War, or that
President Kennedy was assassinated. To suggest that he
ought to say "I believethat President Kennedy was assassinated" or "I knowthat I heardnews to thiseffect,but I cannot
say that I know whether President Kennedy was actually
assassinated" would make no sense to him. He would be
convincedof his rightto say, "I knowthat PresidentKennedy
was assassinated."
Historyis made up of events which,except forthose which
happen to constituteone's personal history,lie outside one's
individual experience. This means that if a knowledge of
past events is to be predicated as a possibility a type of
knowledgeof the world must be admittedwhich is essentially
non-empirical. (I am using the phrase "knowledge of the
world" in contrastto the knowledgeof mathematicalor logical
truths and to religious or mystical knowledge. "World"
will be taken to mean the totalityof spatio-temporalpheno-

82
BERNARDWEISS
mena, past and present.) If I insistdogmaticallythat I know

that President Kennedy was assassinated, I am claiming
a knowledgeofan eventwhichhas no basis in myown experience
(contraryto myknowledgeofwhat I heardor saw in the media).
A historianwho adopts a strict empiricistpoint of view
is constrainedto deny the possibilityof having a knowledge
of the actual eventswithwhichit is his businessto deal. The
only knowledgewhich he is entitledto claim is the knowledge
of what is containedwithinhis sources,the knowledgeof data.
This data consists mainly of verbal statements about past
events, although non-verbaldata (artifacts,etc.) will also be
useful. The data or statementsabout past events are what
the historiandirectlyexperiences,not the events themselves.
They are his real "facts" out of whichhe attemptsto construct
in his imagination a picture of past events. This picture
is reflectedin the writtenaccount which he gives of the past
event. A pictureor account of a past event is judged more
or less "appropriate"or "probable" withreferenceto the data.
The data must, of course,be siftedin accordancewith agreed
upon methodologicalrules, and the historianwill seek to rely
on the data which have the best methodologicalcredentials.
It is presupposed that some statements about past events
are more true than othersor that some are true and some are
false. However,thistruthand falsenesscannotbe determined
throughempirical verification,since the historiancan never
experience the event. Hence the need for methodological
criteriawith whichto evaluate the usefulnessof data in constructingappropriatepicturesor accountsof past events (or we
mightsay: with which to determinethe degree of presumed
truth-as opposed to empiricallyverifiedtruth-which may be
reasonablyattributedto statementsabout past events.)
This strict empiricistpoint of view is at odds with the
ordinary layman's view. The layman is aware that most
of what he considersto be his knowledgeof the world, past
and present, is derived, not from personal experience,but
frombooks,the media,generalreport,education,etc. Indeed,
in ordinaryEnglish "knowledge" has this broader layman's
meaning(ordinarylanguage being, in any case, by definition

KNOWLEDGEOF THE PAST
83
the language of laymen). When one attends a university

to acquire knowledge,one generallyhas this sense in mind.
The universitystudentwill ofcourseseek to masterthe methods
of empiricalresearch,partly by undertakingresearchprojects
on his own,so that he may in timeadd to the existingknowledge
of the world,but the fact remainsthat he will build up a store
of knowledgeof the world not throughempiricalresearchbut
throughtheresourcesofthe libraryand the classroom. Knowledge of the world, in the broader sense, is the possession of
society as a whole; the individual seeks to acquire as large
a part of it as possible and in so doing to furtherhis own
socialization.
Islam, like its kindredpredecessorsJudaismand Christianity,
is firmlyrooted in history. Any religion that is centered
upon revelation necessarilyhas a large stake is history,for
revelationis always an event in time,or seriesof events. The
revelatoryevent or events, togetherwith certain surrounding
events,constitutea special historywhichis vestedwitha sacred
character. In Islam the supreme revelatory event is the
"descent" of the Qur'An upon the Prophet Muhammad and
his deliverance of the Qur'an to his contemporaries. Each
verse,in facteach word,is, in a sense,an eventin its own right,
foreach representsan utteranceon the part of the Angel and
then subsequentlyon the part of the Prophet. Those words
and deeds of the Prophet himselfwhich are the product of
divine inspiration(wahy) are also to be counted as revelatory
events. Together with the revelatory events themselves,
the events involved in the establishment of the Muslim
community (umma) under the Prophet and its subsequent
expansion and developmentunderthe RightlyGuided Caliphs,
especially Abu Bakr and 'Umar, complete the sacred history
of Islam.
That the past should be excludedfromthe domain of human
knowledgeis unthinkableto any serious Muslim who adheres
to the traditionalvocabulary of Islam. "Knowledge" ('ilm),
in the Islamic tradition,is not exclusivelyempiricalbut has
the broader charactermentionedabove. Traditional Muslim
learned opinion takes the point of view of the average man:

84
BERNARDWEISS
'ilm has much the same sense as "knowledge" in ordinary

English. Accordingto this point of view, I have everyright
to claim to know that President Kennedy was assassinated,
just' as every Muslim (or, for that matter,any non-Muslim)
has the rightto claim to know that the Prophet
ammad
all
of
uttered
words
the
found
in
the
Muh.
actually
present-day
text ofthe Qur'an. It would no moremake senseforme to say
"I thinkPresidentKennedywas assassinated" than forsomeone
to say "I thinkthe words of the Qur'an were actually uttered
by the Prophet Muhammad." The events of sacred history,
like many events of public historygenerally,belong to that
world the knowledgeof which is available to individuals in
any age.
Muslim scholars and theoristshave, of course, always been
perfectlyaware of the fact that most of the past lies beyond
the reach of human knowledge,that most of it is, in fact,lost
foreverand will never be retrieved. They have also been
aware that of that part of the past whichlingersin the human
memoryor in writtenrecords much is speculative: the best
we can do is forman opinion (zann) as to what precisely
happened, knowledge of the event being beyond us. This
is true even fora good part of the sacred history. The bulk
of sayings and deeds attributedto the Prophet,for example,
have a merely probable character. Even when the degree
of probabilityis very high, as is the case with sayings and
deeds recordedin the canonical hadithcollections,it still falls
shortof becomingknowledge. However,that the past should
be regarded as beyond knowledge altogether,as limited by
definitionto the probable, was unthinkableto' the medieval
Muslim. To relativizethe past so completelywould have been
to relativize and ultimatelydo away with revelation itself.
There had to be a core of genuinehistoricalknowledgeif there
was to be revelationin the true sense of the word. Particular
sayings and deeds of the Prophetmightbe open to discussion,
but the edificeof Islamic revelationas a wholehad to restupon
a foundationof unassailable historicalcertainty. Chiefamong
the factsof historywhichwereincludedin thisfoundationwere
the existence of the Prophet Muhammad, his prophetic mission,

85
the words which he delivered to mankind as from God and

whichconstitutedthe proofof his prophethood,and the pivotal
events of his career and of the early development of his
community.
Any system of thoughtwhich allows for the possibilityof
a knowledgeof the worldnot based on directexperiencemust
deal with an obvious question: if direct experienceis not the
basis of this knowledge,then what is? The Muslimtradition
has a ready answerto this question in the concept of lawdlur.
It is this concept that I propose to explore and attempt to
explain in the followingpages. The concept is propounded
primarilyin the books on the principlesof jurisprudence
(us.l
the
al-fiqh). There are many books on this subject, and since
seems
of
to
be
more
or
less
uniform
concept lawdtur
throughout
the literatureon us'l al-fiqhone book would seem to serve our
purpose as well as any other. I have chosen AbU H;Iamid
al-fiqh entitled
al-GhazilI's (d. 1111 A.D.) work on
for
added
us.l
the
reason
min
'ilm
that, unlike
al-Mustasfd
al-ugsl
most other books in this field, this book provides us with
a generaltheoryof knowledgewithinwhichto place GhazalI's
discussionof lawdtur. The usual place to deal withthe theory
of knowledgein medieval Islam was withintheology(kaldm),
and GhazMlI'sepistemologicaldoctrineis the mainlineone found
in the great works of theology,such as al-Iji's Mawdqif.(1)
What is unique about GhazalI's treatment(and convenient
forus) is that withina singlework,i.e. the Muslasfd,he treats
both a topic fromjurisprudence(lawdlur) and a topic from
theology (theory of knowledge),even though he deals with
these topics in separate sections.(2)
(1) Al-Iji's epistemological doctrine has been studied in great detail by
Josef van Ess in his Die Erkenninislehredes
al-fct (Wiesbaden, 1966).
'Ad.udaddtn
(2) The subject of tawdturis dealt with in a special subdivision of the section
on the Sunna of the Prophet (vol. I, pp. 132-140). This is the normal placement
of this subject within usill al-fiqhbooks. The theoryof knowledge, on the other
hand, is to be found in the Introduction (muqaddima) to the Musitafd. Ghazdlt
nowhere formallyelaborates a theory of knowledge. The primary focus of the
Introductionis upon logic. However, the elements of a theory of knowledge are
discernible in the section of the Introduction in which GhazAlt deals with the
"material" (as opposed to the "formal") aspect of logical proof (burhdn), that is
to say, with the ultimate premises upon which all valid reasoning, i.e. reasoning

86
BERNARDWEISS
The term lawdiur means literally "recurrence." As a

technicaltermof traditionalMuslimscholarshipit has reference
to the recurrenceof statementsabout past events. According
to the theorywhich GhazalI and other
(writerson ugsdl
us.dlis
the
recurrence
of
such
statements
propound,
produces
al-fiqh)
in the minds of the hearersa knowledgethat such statements
are true. This theoryis expressed succinctlyin the phrase
al-lawdtur yufid al-'ilm ("recurrence imparts knowledge"),
which appears repeatedlyin workson usdl al-fiqh. And it is
importantto add that the knowledgethat a statementabout
a past event is true is taken to be tantamountto a knowledge
of the event itself. This makes perfectsense. If I am able
to say that I know that the statement"President Kennedy
was assassinated" is true, then I am able to say that I have
a knowledgeof the event itself. Since I understandEnglish
and knowwhat the wordsmean, I have some sortof knowledge
of what happened in Dallas on November22, 1963. It may
not be quite on the same orderas the knowledgeofeyewitnesses,
whichis empiricaland therefore
vivid and richin sensoryinput,
but it is knowledgenonetheless. This beingthe case, it is quite
legitimate,in discussingthe subject of the knowledgeof past
events, to place the focus on the knowledgeof the truth of
statementsabout past events, for when one has established
the latterone has ipso facto establishedthe former.
This is notto say thattheknowledgeofpast eventsis reducible
genericallyto a knowledgethat statementsabout past events
are true. This knowledgewhich the eye witnessof an event
has certainlyentails much more than a knowledgethat given
statementsabout that event are true,althoughthe eye witness
will of course possess the latter. What is affirmedhere (in
our attempt to explain the medieval Muslim theory) is that
which leads to true conclusions, is based (vol. I, pp. 43-49). Ghazdlt calls such
premises"the sources of knowledge" (maddrikal-'ilm) or "the sources of certainty"
(maddrik al-yaqtn). References in this article are to the Bulaq edition of the
Mustasfd (1322 hijrt). This edition has been reproduced in recent years in
H. Laoust provides a briefsynopsis of
Baghdad by Qasim Muhammad al-Rajab.
the content of the Mustasfd (including the section on tawdfur)in his La politiquede
Ghazdlt(Paris, 1970), pp. 152-182.

87
the richertype of knowledgeof a past event which the eye

witnesspossessesdoes not representthe only type ofknowledge
that is possible;rather,thereis anothervalid typeof knowledge
of the event, namely that which is representedby the sheer
knowledgethat given statementsabout the event are true.
This takes us to the crucial question,one which the reader
will no doubt have alreadyraised in his mind beforethis point.
How is it possibleto have a knowledgeofthe truthofstatements
about past events withouta priorempiricalknowledgeof the
event itself? Is not the formerderived from the lqtter?
Are not the two ultimatelyinseparable? Stated otherwise:
How is it possibleto knowthat a statementabout a past event
is true withoutdirectempiricalverificationof the statement?
The phrase al-tawdiuryafid al-'ilm is precisely the answer
to this question. The theory which it expresses attempts
to show how a knowledgeof the truth of statementsabout
past events can be possible without empiricalverification.
Quite obviously, the theory under considerationin these
pages stands in sharp contrastto modernempiricism. It is
instructiveto compare the usifli way of thinkingwith that
modern historians. Both the
of empirically-inclined
ussli
and the empiricisthistorianagree that what we directlyexperience is not past events as such but statementsabout past
events. 'It is the statementthat lingersafterthe event has
disappeared and intervenesbetween us and the event. It is
the statement that we have directly before us. However,
means of his conceptof recurrenceis able to gain
the
upsli by
from the statement a knowledgeof the event itself (which,
we have suggested,amounts to the knowledgethat the statement is true, with any imaginativereconstructionas a nonhistorian
essentialby-product),whereasthe empirically-inclined
must contenthimselfwith an imaginativereconstruction
of the
event which is at best appropriateor probable (or presumed
to be true, as 'opposed to being actually verifiedas true), and
this is because the latter is never in a position,lacking direct
to affirm
unequivocallythatthestatement
empiricalverification,
is true.
We said above that accordingto the theoryunderconsidera-

88
BERNARDWEISS
tion the recurrenceof statementsabout past events produces

in the minds of hearersthe knowledgethat the statementsare
true. This knowledgethusarisesin the mindwithoutempirical
verification. However, this general statementof the theory
should not be taken at face value. It is subject to a number
of veryimportantqualifications,and it is only afterconsidering
these qualificationsthat we will be in a positionto understand
what the theoryreally is tryingto say. These qualifications
will make it apparent how very technical the term lawdlur
and the related term mulawdlirare. It is because of the
highly technical character of these terms that we shall use
them in their Arabic formrather than using literal English
equivalents such as "recurrence"and "recurrent."
In order to be mulawdlir(literally"recurrent")a statement
about a past event must be more than simplyrecurrentin the
ordinarysense. It must meet a numberof importantconditions called "conditionsof lawdtur" (shurutlal-tawdtur). In
one passage (') Ghazali stipulates foursuch conditions(stated
fromhis):
here in an orderslightlydifferent
1. The statement must be based on knowledge, not on
opinion. (This seems to be saying that the opinions which
persons may have concerningpast events can produce in the
minds of others only their like; opinion can never produce
that which surpasses it, namely knowledge. Hence, statements expressing opinions about past events cannot be
mulawdtirin the technical sense.)
2. This knowledge must be necessary, that is, based on
what is perceived throughthe senses.
3. The numberof personsmakingthe statement(or number
of occurrences of the statement) must be kdmil (literally
"complete"). Kdmil is in this case a technicaltermmeaning
"sufficientto rule out the possibilityof collaborative fabrication." We will translate kdmil in this technical sense as
"adequate." In order to fulfillthis conditionof adequacy,
a statementmust occur again and again (that is, issue from
(1) Mustasfd, vol. I, p. 134.

89
one person after another) until it has occurred a sufficient

number of times to rule out all possibilityof collaborative
fabrication. This recurrenceseems to entail both a subjective
and objective aspect. If I am to have a knowledgeof the
truth of a statement about a past event, I must hear the
numberof time (that is, froma sufficient
statementa sufficient
number of differentpersons) to rule out collaborative fabrication. In other words, the recurrenceof the statement
must take place within my experience. On the other hand,
the numberof occurrencesof the statementwithinthe world
at large may far exceed those which take place within my
own experience. I as an individual may be exposed only
to a portion of this total number of occurrences. What is
importantis that the number of occurrencesI am exposed
to be adequate. It goes without saying that the number
of occurrencesin the world at large will be adequate; they
will, in fact, be in excess of the adequate number.
4. When a reportis transmittedfromthe originalwitnesses
to an event througha series of intermediaries,
the threeabove
mentioned conditions must apply to each successive point
or stage in the transmissionprocess.
In anotherpassage, Ghazilli speaks of two ratherthan four
conditions. He calls these "conditions of the occurrenceof
knowledge(of the truth of statementsabout past events),"
ratherthan "conditionsof lawdlur," although it is clear that
the two categories come down to the same thing. That is,
the "conditions of lawdlur" represent not only conditions
governingthe application of the termmulawdlirto statements
about past events but also conditions governingthe actual
occurrenceof knowledge,conditionsunder which a statement
about a past eventwill actuallyproduceknowledgein the mind
of the hearer.
It will be useful to quote the relevant passage in full:
If an adequate number (of persons) makes a statement
(about a past event) and the knowledgethat theirstatement
is true does not occur (in the mind of the hearer),then one
cannot but be certainthat the personsin questionare telling
a falsehood, for there are just two conditions governing

90
BERNARDWEISS
the occurrenceof knowledge:(1) that the number(of persons

makingthe statement)be adequate and (2) that thesepersons
be making the statementon the basis of certainty(yaqin)
and empiricalobservation (mushdhada). Therefore,if the
numberof persons making the statementis adequate, then
the fact that the knowledge(that their statementis true)
does not occuris due to a failureto meetthe secondcondition;
forif the personsmakingthe statementhad been tellingthe
truth,then, given the adequacy of theirnumber,the knowledge (that their statementwas true) would have occurred
(in the mind of the hearer)necessarily.(1)
Here, clearly,we have two principal"conditionsof lawdiur":
(1) the statement about the past event must be based on
certaintyand empiricalobservation;that is to say, it must be
truein the sense of beingempiricallybased; and (2) the number
of persons making the statement(or number of occurrences
of the statement)must be adequate. Clearly Ghazali is here
combiningthe firsttwo of the fourconditionsmentionedabove
as a single condition. Strictlyspeaking,the firsttwo of the
four conditionsare not separate conditionsin any case, since
the second simply narrowsdown the firstby specifyingthat
the knowledgerequiredin the firstbe a knowledgebased on
sense perception. That is, if the second conditionhas been
met,the firsthas also necessarilybeen met,since sense perception rendersknowledge. As forthe fourthcondition,it is in
fact not an independentcondition but a "super-condition,"
as it were. It states somethingabout the other conditions,
namely that in cases where a statementis transmittedfrom
person to personthese conditionsare applicable to each point
in the transmissionprocess.
The "super-condition"comes into play where the hearer
is not in directcontact with originalwitnessesto an event but
is rather dependent on intermediaries. Most of what we
considerto be our knowledgeof the world is acquired from
statements,not of witnesses,but of intermediaries. Knowledge is thus passed along a chain, or rather a multitudeof

91
werepresumed
chains. In GhazMlI'stime,these intermediaries
to be almost exclusively persons. Ghazall did not live in
the age of the media, as we do, and even the writtenword,
being the product not of the printingpress but of fallible
copyists, was considered inferiorto the spoken word. The
"super-condition",we have said, states that at every point
in the process of transmissionof a statementthe two primary
conditions must be met. Only if they are met can the
knowledge of the event flow through time, especially from
generationto generation. Thus there must always be, at
every point in time, an adequate numberof persons making
possible an adequate numberof occurrencesof a given statement within the experienceof any hearer. The number of
persons who originallywitnessed the event must necessarily
be adequate. As they pass the word along to intermediaries,
their statements will for a time be co-mingled with the
statementsof intermediariesto insure the flowof knowledge.
But eventuallythe originalwitnesseswill all die, and the flow
of knowledgewill thereafterdepend entirelyon intermediaries,
whose number, as the word is continuallypassed on from
personto person,must always remainat or above the adequate
level. Furthermore,the statements must, at every stage
in the transmission
process,always be based on senseperception.
not
it entirelyclearwhatthismeans. Presudoes
make
Ghazali
will always
mably it may eithermean that the intermediaries
be makingstatementson the basis of sense perceptionin that
they will state only what they have heardotherssay (just as
the originalwitnessessaw or experiencedin some other way
the event itself),or it may mean that the intermediarieswill
always go on statingwhat has been experiencedby the original
witnesses,so that what the originalwitnessesexperiencedis,
as it were, carriedon throughtime and the knowledgewhich
flowsthroughtime continuesto have its ultimatebasis in that
experience.
Any statement about a past event which meets GhazalI's
two primary conditions and which, if transmittedthrough
time, continuesto meet these two conditionsat every point
in the transmissionprocess can, according to the theory

92
BERNARDWEISS
expounded by Ghazalt, be termed mutawdtirin the technical

sense, and such a statementwill by virtueof havingmet these
conditions produce in the mind of the hearer a knowledge
of its truth,whichis tantamountto a knowledgeof the event
itself.
ofthe two primary
It mustbe emphasizedthat the fulfillment
conditionsin no way entersinto the thinkingof the hearerof
a mulawdlirstatementas a preludeto the occurrenceof knowledge. They are extraneousto the hearer altogether. The
theoryexpoundedby Ghazzall affirms
simplythat ifa statement
about a past eventis truein the sense of beingempiricallybased
and ifit is sufficiently
widelycirculatedto ruleout thepossibility
of collaborative fabrication there will occur spontaneously
in the mind of the hearer,i.e. withoutany logical antecedents,
a knowledgethat the statementis true.
Neitherof the two conditionsis sufficient
by itself. A statementwhichis true but does not recuron a scale thay may be
deemed adequate may be persuasive (particularlyif those
but it will not impart
makingit are knownto be trustworthy),
in the technical
knowledgeand will thereforenot be mutawdtir
sense. The same is true,interestingly
enough,of a statement
to be deemed adequate
whichis circulatedon a scale sufficient
but does not derive fromthe experienceof originalwitnesses
and is thereforenot true. It may seem peculiarthat Ghazall
in view of its
is unwillingto call such a statementmutawdtir
wide circulatioD,but it is at this point that the technical
characterof the Arabic term tawdturbecomes most apparent:
it is the recurrencenot just of any kind of statementsabout
past events but only of empirically-basedtrue statements.
One may ask how it is that a falsestatementcan gain circulation among a large group of people whose number precludes
collaborative fabrication. GhazalI's answer is simple: the
fabricationmay have been the work of a part of the group,
whose numberdid not rule out collaboration,and the resultant
falsehood spread to the rest of the group, who accepted it
without being aware of its falseness. These latter become
unwittingperpetuatorsof a falsehood. The essential point
hereis that a falsehood,thoughit can arise onlywithina group

93
small enough to make collaborationpossible, can easily gain

circulation among a large group of people whose number
precludescollaborativefabrication. There are ample examples
of this. GhazMlImentionsthe Prophet Muhammad's alleged
appointmentof 'All as the firstImam, which to the Sunni
Muslim is a clear falsehood. The numbers of Sht'is who
to rule out
perpetuatethis "falsehood" is more than sufficient
collaborativefabrication.(1) The same can be said of Jewish
and Christianheresies.
If it is possible for a falsehood to be circulated ampng a
numberof people sufficient
to rule out collaborativefabrication,
why bother with the second condition at all? The answer
would seem to be that some conditionmust be posited other
than the sheer truth of a statement,and this conditionmust
have somethingto do with number. Ghazali seems to feel
that thereis some sortof strengthin numbers. The statement
of the many seems to have in his thinkingsome sort of impact
which the statementof the few does not have. This impact,
so Ghazla's theoryseems to be affirming,
can be best explained
the
fact
that
the
are
by
many
incapable of collaborative
fabrication,in contrastto the few.
The positingof the two conditionsseemsto be based on some
such thinkingas the following. We are all aware of the
presencein our mindsof knowledgeof the truthof statements
about certain historicalevents and about certain phenomena
in the presentworldto whichwe have notbeendirectwitnesses.
Like all knowledge,this knowledgeis a subjective state perceived throughintrospection;it is simplythere. We cannot,
however, attribute this knowledge to sensory perceptionor
rational intuition. We must thereforeexplain it on some
other grounds. On attemptingto analyze the circumstances
that attend the occurrenceof such knowledgein our minds,
we note an importantpattern: such knowledge occurs only
when we have heard the statementfroma large number of
people. An awarenessthus arisesof numberas a conditioning
factor. This knowledgeis never presentin our minds when
(1) Mustasfd, vol. I, pp. 138-139.

94
BERNARDWEISS
we have heard the statementfrom a few persons but only

when we have heard it frommany. And yet numberin itself
cannot provide the whole explanation,since there are statements which we have heard from large numbers of people
whichwe do not knowto be true(or even knownot to be true).
To this category belong heresies, superstitionsand fables.
Thereforeit is necessary to add a furthercondition,namely
the truth of the statement.
The termkdamil
(which we have been translatingin its technical usage as "adequate") applies to a precisenumber,namely
theminimalnumberofpersonsrequiredto precludecollaborative
fabrication. Any number which is more than this number
is describedas "excessive," or "more than adequate" (zd'id).
Thus thereis a fixednumberof whichwe can say: withgroups
of this numberor more collaborativefabricationis out of the
question,whereas with groups whose numberis less it is not.
Can this minimalnumberbe determined? Ghazali rejects
the attemptsof other Muslim theoriststo set the number at
fortyor seventy. Rather,he holds that this numberis known
only to God. We certainlyhave no way of computingit.
If we were to attemptto do so by countingthe instances of
hearinga givenstatementup to that pointwheretheknowledge
that the statementwas true occurred in us, we would find
that it is impossibleto pinpointthe momentwhen this knowledge occurs. It would be like tryingto pinpointthe moment
when childhood ends and adulthood begins.(1)
It does not, in any case, matterthat we cannot know what
the minimal number is. The conditions which govern the
occurrence of knowledge of the truth of statements about
past events,we have said, in no way enter into the thinking
of the hearer. It is not necessarythat we know the number
of persons making a statementabout a given event is kdmil
in ordersubsequently
to knowthat the statementis true. One
does not argueone's way fromadequacy of numberto the truth
of the statement. The reverse,in fact,is the case. If I find
within myself a knowledge that the Prophet Muhammad

95
actuallyexistedI can arguefromthisthat the numberofpersons

fromwhom I have heard a statementto this effecthas been
adequate, since the knowledgecould not exist if the number
were not adequate.
(1)
Obviously, extremelylarge numbersmust necessarilybe in

the "more than adequate" category. One can know, for
example, that a millionpersonsare incapable of collaborative
fabrication,since this would require a common purpose and
common motivationswhich, in the existingscheme of things,
are impossiblein the case of such a large number. However,
this knowledgeis in itselfno guaranteethat knowledgeof the
truthof a statementwill occur. In the case of the statement
"the Prophet appointed 'All to be the firstImam," we know
that the numberof personsmakingthe statementis well above
the adequate number. If we findthat a knowledgeof its truth
does not exist in our minds,we may argue fromthis that the
other conditionhas not been met, that the statementis not
empiricallybased, is not true.
Thus in the end the term mulawdtlir
designatesa category
of statementsabout past events which we know to be true
by virtue of a knowledgewhich we "find" within ourselves.
This knowledge does not require or result from empirical
it is simplythere,apart fromany such verification.
verification;
Where the knowledgeof the truth of any statementresults
it is indistinguishable
fromempiricalverification,
fromempirical
In
as
such.
the
of
case
statements about past
knowledge
is possible. The theoryof lawdttur
events,no such verification
statementson
places the knowledgeof the truthof mutlawdtir
an independentfooting,renderingsuch verification
unnecessary.
a purely historicalknowledge,distinctfrom
Through tawdtiur
sensoryknowledge,is upheld. Like sensoryknowledge,this
historicalknowledgeis not the productof discursivereasoning;
it simply occurs withinus. However, in contrastto sensory
knowledge,it is subjectto thosespecial conditionswhichwe have
just considered.
(1) lakinnd
natabayyanu kamdla'l-'adadi li- anna
bi-hus.2li'l-'ilmi'l-dardrtyi
bi-kamdli'l-adadi
nastadillu 'ala
Muslasfd, vol. I, p. 135.
hus.zli'l-'ilm,

96
BERNARD WEISS
The theorybehindthe concept of lawdiurand its conditions

can perhaps be stated with greatestclarity and succinctness
in Englishas follows:thewidespreadrecurrence
of truestatements
about past eventsproducesin the minds of hearersa knowledge
that thesestatements
are true. "Widespread" must of course
to rule out the
be understoodto mean "on a scale sufficient
this statement
fabrication."
From
of
collaborative
possibility
of the theorytwo corollariesfollows:(1) a recurrenceof true
statementsabout past events which is not widespread does
not produce in the minds of hearers a knowledgethat these
statementsare true,and likewise(2) the widespreadrecurrence
of false statementsabout past events does not produce in the
mindsof hearersa knowledgethat theyare true. It is helpful
to bear in mind,especiallywith respectto the second corollary
which appears to be tautologous, that the knowledge that
a statementis true is to be distinguishedfromthe actual truth
ofthestatement. The knowledgeis a subjectivestateoccurring
within the mind; the truth of the statementis an objective
realityexistingoutside the mind, the objectof the knowledge
that takes place within the mind. Thus what the second
corollaryis meantto say is that knowledge,thougha subjective
state, cannot exist apart fromits properobject. Hence, the
knowledgeof the truthof a statementcannot exist unless the
statementis in facttrue. If the statementis false,one cannot
have the knowledgethat it is true.
Despite its sophistication,the theoryelaborated by Ghazalt
leaves certain difficultiesunresolved. It seems to posit a
knowledgewhich has no demonstrableobjective underpinning.
We are presentedwitha knowledgewhichis simplythere,which
we simply"find" withinourselves,withoutreally being shown
how the knowledgegot there. The "conditionsof lawdtur"
do not really explain this. Hence the theory seems to be
locked into an essentially subjectivist stance. Knowledge
is, of course,a subjectivestate; we do "find"it withinourselves;
but if some sort of objective underpinnings
cannot be pointed
out it ceases to be knowledge.
This subjectivisttendencybecomesevidentwhenwe consider
cases where conflicting claims to knowledge of the truth of

97
statementsabout past events arise. How are such disputes

we have noted earlier,
to be resolved? Empiricalverification,
is ruled out by definition. Rational intuition and logical
deductionare also ruled out. The truthof statementsabout
past events is not axiomatic, nor is it derived fromaxioms.
If it were possible to resolve the disputes on the basis of the
widespreadrecurrenceof statements,we would have an objective underpinning. But we cannot do this. One disputant
cannot say to another, "Wait until you have had a chance
to listen to more people and you will see that I am right,"
because falsestatementscan be just as widelycirculatedas true
statements;in fact,theycan be morewidelycirculatedin given
instances. Thus the conditionof adequacy of number does
not provide us with our objective underpinning. And the
other primarycondition of lawdiur,namely the actual truth
of statements,is hardlya groundforargument:it is ratherthe
desideratumof the dispute.
In the section of the Muslasfd on the definition(hadd)
of knowledge, Ghazilt notes that knowledge may easily be
confusedwithcredence(i'tiqdd). This is because both exhibit
a kind of confidence (jazm). Thus a person who claims
to have knowledgemay in fact be mistaken; his knowledge
may turn out to be mere credence. The differencebetween
knowledgeand credenceis that in the case of knowledgeone
has taken into considerationthe opposite of what he knows
and ruled it out, whereasin the case of credenceone attaches
oneselfblindlyto what one believes withouttakingreal cognizance of its opposite. Knowledge involves an open mind,
credence a closed mind, with the result that knowledgehas
a firm intellectual groundingand is secure against doubt,
whereas this cannot be said of credence.(1) In accordance
with this distinction,we could attemptto argue, with respect
to cases of conflictingclaims to knowledge of the truth of
statementsabout past events,that in such cases at least one
of the disputants must be a dogmatistwho has mistakenly
taken his credence to be knowledge. But this would take
(1) Muslasfd, vol. I, pp. 25-26.

98
BERNARD WEISS
us back to square one, forwe would be dealingwitha situation

where one disputant(the one who possesses true knowledge)
would be expected ultimatelyto prevail over his opponent
by revealingthe processwherebyhe had taken both that which
he now knows and its opposite into account and had settled
withabsoluteassuranceon theformer,
and thiswouldpresuppose
the objective underpinningwhich we have not been able to
discover.
In order to discover the place which the knowledgeof the
truthof statementsabout past eventsoccupieswithinGhazili's
theoryofknowledge,we mustturnto thesectionofthe Mustasfd
which deals with logical proof (burhdn). In examiningthis
section,we must bear in mind that the knowledgeof the truth
of statementsabout past eventsfallswithinthe largercategory
of knowledge of the truth of statements about phenomena
whichlie outsideour experience. Such phenomenaincludenot
only past events but also present phenomena, such as the
existence of cities we have never visited. As was intimated
at the beginningof thisarticle,what is at stake in the discussion
of knowledgeof the truth of statementsabout past events
is the possibilityof a non-empiricalknowledgeof the world
at large,both past and present. Such knowledgein its entirety
is the concern of all theorizingabout lawdtur,for present
phenomenaare just as much as past mediatedto us by widely
recurrenttruestatements.
In order to spare ourselves of the clumsinessthat comes
with frequentreferenceto "the knowledge of the truth of
statementsabout past events and present phenomena lying
outside our experience" or "knowledgeresultingfromwidely
recurrentstatements," we will in the followingparagraphs
take the libertyof referring
to this knowledgeby means of the
shorter phrase "second-hand knowledge." This designation
can be misleadingunless it is carefullykept in mind that the
term "second-hand" is intendedto convey nothingmore than
the notion that the knowledgein question is a knowledgeof
the truthof statementsmade to ,the hearerby othersand that
such statementsare possible only because someone was direct
witnessto the eventsor phenomenathemselvesand was prompt-

99
ed to make statementsabout these events or phenomena.

The knowledge in question thus presupposes an empirical
knowledge in othersapart from which the true statements
(which are its object) could not exist in the firstplace. It is
not a question of empirical experience being handed from
one person to another:that by definitionis impossible. It is
rather a question of statementsbeing made on the basis of
empirical experience and of their truth being subsequently
known to personsotherthan the originalwitnessesas a result
of theirwidespreadrecurrence(lawdlur).
According to Ghazal, the knowledge which we are here
designatingas "second-handknowledge" belongs to the class
of necessary knowledge (al-'ilm al-dari~ri). In his usage
necessary knowledge is knowledge which "imposes" itself
upon the intellectin the sense that it does not admit of doubt.
It is thereforenot the product of any conscious deductive
reasoning,for that which results fromdeduction may, when
severed fromits logical basis, be doubted; it requires logical
proof in order to be raised above doubt. Knowledge which
is dependentupon logical proofconstitutesa second category,
namely discursive knowledge (al-'ilm al-nazari). Necessary
knowledge is the source of discursiveknowledgein that all
discursive knowledgeis derived ultimatelyfromit by means
of logical deduction. The distinction between necessary
and discursiveknowledgethus amountsto a distinctionbetween
knowledgewhichis the sourceofotherknowledgeand knowledge
which is derived fromthe source. It would be in keeping
with Ghazl11'sway of thinkingto speak of necessaryknowledge
as "source knowledge" and discursiveknowledgeas "derived
knowledge." It must be emphasized, however, that the
terms"source" and "derived" are used here in a logical rather
than strictly epistemological sense. All logical deduction
must start somewhere. Necessary knowledgeis this starting
point. The conclusion of a syllogismrepresentsknowledge
which has been derived froma source, namely the premises.
If the premises are themselves derived, then they must be
carried back to prior premisesuntil finallywe reach the real
sourceof all logical deduction,namelythose propositionswhich

100
WEISS
BERNARD
are not arrived at throughdeductionbut are simply there as

principleswhich impose themselvesupon the intellect.
Ghazili divides necessary knowledge into five types or
sub-categories,which he calls the five "sources of (derived)
These
knowledge" (maddrikal-'ilm, or maddrikal-yaqin). (1)
are:
1. A prioritruths(awwalydlt),as representedby statements
such as "I exist," "A thingcannotbe both eternaland created,"
"If one of two contradictorystatementsis true the othermust
be false," "Two is greaterthan one," etc.
2. Introspective observations (al-mushdhaddl
al-bd.intya),
e.g. "I am hungry," "I am thirsty,""I am afraid," "I am
happy," etc.
3. External sensory perceptions (al-mahsstsdial-zdhiriya),
e.g. "The snow is white," "The moon is round," "The sun is
bright,"etc.
4. Inductions (al-tajribiydl),e.g. "Fire consumes," "Bread
satiates," "Rocks fall," "Wine intoxicates," "Scammony
relievesthe bowels," etc.
5. Widely recurrent data (al-mulawdlirdl), e.g. "Mecca
exists," "ShWfi'lexisted," "The number of (required daily)
prayersis five,"etc.
(1) Mustasfd, vol. I, pp. 44-46. GhazAll speaks of seven sources of certainty
(i.e. knowledge) and confidentcredence, but makes it clear in the course of his
discussion that only five of these can be considered sources of certainty (yaqtn,
It should be noted that knowledge and certaintywere, for GhazAltas for
qat').
the usulis in general,correlativeconcepts. To know a thingwas to be certain of it.
GhazAll in fact combines the two concepts in the phrase maddrik al-'ullm alyaqtntya(p. 46). In his discussion of logical proof the concept of certaintyis the
more prominent,primarilybecause of its special relevance to logic. The question
with which GhazAll is chieflyconcerned in this discussion is not "what are the
ultimate sources of all knowledge" (although his answer to this question emerges
clearly enough between the lines) but rather "how can we be certain that the
conclusions of our logical proofs are true?" The certainty of the conclusion
obviously depends on the certaintyof the premises,and the ultimate premisesfrom
which all logical deduction proceeds become the "sources of certainty." But, as
certaintyis a correlate of knowledgeand as the conclusions of logical proofsclearly
constitute knowledge, these ultimate premises may also be called "sources of
knowledge," it being understood that in the context of logic they are sources of
derived knowledge.

101
It may be noted that if we combine the second and third

ofthesesourcesofknowledge,namelyintrospective
observations
could
also
called
internal
and
external
be
(which
sensations)
in
his
himself
does
sensory perceptions(as Ghazill
Mi'ydr
al-'ilm),(1) we thenhave fourbasic typesofnecessaryknowledge:
(1) a priori knowledge,(2) sensory knowledge,(3) inductive
knowledgeand (4) second-handknowledge.
SomeMuslimtheoristsup to thetimeofGhazallwereunwilling
to include the last two of these types of knowledgeunder the
category of necessary knowledge. They argued that only
knowledge which is absolutely primitiveand underived, as
are the firsttwo types, deserveto be so classified. Inductive
knowledge and second-hand knowledge, they maintained,
are not absolutely primitiveand underivedbecause they are
the product of a process of reasoning. This reasoningis not
"manifest" (zdhir), as is the reasoninginvolved in ordinary
logical deduction. Rather, it is "hidden" (khafi). We may
sense its traces,but we are not fullyaware of it at the moment
when it occurs withinour intellects;it eludes us. "Hidden"
reasoningis subliminal.
Ghaztlt accepts the notion that inductive knowledge and
second-handknowledgeentail a "hidden" reasoningbut takes
the view that these types of knowledgemust be regardedas
necessarypreciselybecause the reasoningupon whichthey are
based is hidden. Where knowledge occurs without there
being a clear-cut consciouslyutilized logical basis which the
intellectis able to retain and referback to, it makes perfect
sense to say that the knowledge imposes itself upon the
intellect.
(2)
Implicitin the notionof "hidden" reasoningis a recognition

that inductive knowledgeand second-handknowledgecannot
be primitivein the same absolute sense that a priori truths
and sensoryknowledgeare primitive. If not primitive,they
must be derived; if the derivationdoes not occur by means
(1) Ed. Sulayman DunyA (Cairo: Dar al-Ma'arif, 1961), p. 187.

(2) Mustaffd,vol. I, pp. 132-134.

102
BERNARD
WEISS
of any conscious operation of the intellect, then a covert

operationmustbe posited.
A prioriknowledgeand sensoryknowledgewere not, among
the medieval Muslim theorists,opposing categories,as they
tended to be in the Western dialogue between "rationalists"and "empiricists." Rather, rational intuition and sensory
perceptionwere both accepted as sourcesof knowledge,having,
it would seem, a complementaryrelationshipto each other.
In a strictepistemologicalsense they are the ultimatesources
of knowledge. As such they stand apart from inductive
knowledgeand second-handknowledgewhich,as we have noted,
are sources only with referenceto logical processes.
One is temptedon the basis of Ghazalt's discussionto posit
three categoriesof knowledgerather than two: (1) primitive
knowledge,the ultimate source of all other knowledge,(2)
covertlyderivedknowledge,and (3) logicallydeducedknowledge.
However, this three-foldclassificationis apparentlynowhere
to be foundin the literatureofmedievalIslam. That literature
knows only the categories"necessary" and "discursive"; and
for this reason the placement of covertlyderived knowledge
becomes somethingof a problem.
Ghazalf apparently does not regard "hidden" reasoning
as entirelybeyond analysis. It may, he seems to say, be
reconstructedby the theorist. He reconstructsthe reasoning
involvedin inductionas follows:"If it were not the case that,
for example, rocks always fall, then there would necessarily
be some variation in our experience:sometimes,upon being
released, rocks would fly upwards or to one side or remain
in their place. The fact that in our experiencerocks have
always fallen is a sure indicationthat it is in the nature of
rocksto fall." (1) As forthe reasoningwhichunderliessecondhand knowledge, this, in Ghazali's reconstruction,entails
two "premises": "(1) Those (makingsuch-and-suchstatement)
circumstancesand motivations
are by virtue of theirdifferent
and their large numbersuch that nothingcould have united
(1) This reconstructionis phrased in my own words.
p. 46.
See Mustasfd, vol. I,

103
them all in (the perpetuationof) a falsehood;ratherthey can

only have concurredon the truth; and (2) they all concurred
in makinga statementabout an actuality(wdqi', i.e. a spatiotemporalphenomenon,not an opinionor rationalprinciple)."(1)
From these premises,so one gathers,it followsthat the statement is true. The two premisesseem to have some sort of
relationshipto the two conditionsof lawdlurmentionedearlier.
It would appear that these conditions enter, in some way,
into the "hidden" reasoning,although just how they enter
in is not clear, especiallyin view of what we have said earlier
about the conditions not constitutinglogical antecedents.
The logic entailedin the "hidden" reasoningis obscureat best,
and Ghazall does not choose to elaborate upon it. Even
under analysis the "hidden" does not come entirelyto light.
It remainssomethingof a mystery.
There is, for Ghazall, a certainsimilaritybetweeninductive
knowledge and second-hand knowledge which goes beyond
the fact that both are covertlyderived. Inductiveknowledge
results froma recurrenceof experiences. For example, one
may observe that upon applyinga certainlinimentto a pain
in somepartofthe bodythepain disappears. If thisexperience
keeps recurring,one will eventually reach the point where
knowledgewill occur,namelythe knowledgethat the particular
type of linimentused cures the pain in question. Similarly,
second-hand knowledge also results from a recurrence of
experiences,the repeated hearing of a statement concerning
a past event.
Furthermore,both inductive knowledge and second-hand
knowledge are preceded by a gradual growth of opinion.
After observing on a given number of occasions that pain
ends when a linimentis applied we findthe opinion forming
withinus that the linimentcures the pain; as this experience
continuesto recur a point is reached where the opinion gives
(1) Mustasfd,vol. I, p. 132. GhazAll seems to say on this page that the mind
is aware of these premises but not aware that it is aware of them; not aware, in
other words, of its own awareness. This is a curious statement, but it does not,
it seems to me, contradictthe characterizationof the "hidden" reasoning as essentially subliminal.

104
WEISS
BERNARD
way to knowledge. The same thinggoes forstatementsabout

past eventsand presentphenomenalyingoutsideour experience.
We do not, upon having our firstexperiencesof hearingfrom
a number of persons the statement "Shffi'~ existed," know
immediatelythat the statementis true; rather we firstfind
ourselves believing that the statementis probably true, and
only subsequently,when the higheststage of opinionhas been
transcended,do we know that it is true. In both cases we
are never aware of the precise momentin which knowledge
emerges. This eludes us. (1)
Despite these similarities,there is, however,an important
differencebetween inductive knowledge and second-hand
knowledge,which, though not expressly stated by Ghazall,
is clearly implied. Inductive knowledge is a generalization
from particular experiences,whereas second-handknowledge
is not. The knowledgethat the statement"Shafi'l existed"
is true is in no sense a generalizationfrom the particular
that
experiencesof hearingthis statement. The affirmation
is
is
not
an
a
a
true
affirmation
of
statement
such
general
truthabout the phenomenalworld; it does not belong to the
such as "rocks fall," "fire rises," "wine
class of affirmations
As
we have already noted, second-hand
etc.
intoxicates,"
is
a
knowledgethat takes us beyond the reach of
knowledge
is
our senses: it a knowledgeof the absent. Althoughit has
a formal linkage with the experience of original witnesses,
there is no question of its emergingout of our experiencein
the way thatinductiveknowledgeemergesout ofour experience.
To sum up: Ghazali is certain that there is such a thing
as knowledgeofpast events,such as the existenceof Muhammad
and his call to prophethood. This knowledge is, owing to
the absence of an empiricalbase, reducible to a knowledge
that statementsabout past events are true. Such knowledge
occurs spontaneously, that is to say without consciously
(1) Laoust notes that according to Ghazzali a new convert to Islam may be
excused for denyingmatters transmittedby tawdtur. Apparently,a new convert
is considerednot to have been as yet sufficiently
exposed to recurrentreportsamong
Muslims to have reached the point where a firmknowledgethat these matters are
true (i.e. a second-hand knowledge) can occur. Laoust, op. cit., p. 354.

105
utilized logical antecedents,whenever true (in the sense of

empiricallybased) statements about past events are heard
to rule out the possibility
froma numberof personssufficient
of collaborativefabrication. It does not occur when a false
statementis heard fromthis number of persons; nor does it
occur when a true statementis heard by a smaller number
of persons. Since no consciouslyutilized logical antecedents
are involved, it must be classified as necessary knowledge.
It does not, however,have the immediacyof a priorirational
truths and sense perceptions. For this reason a "hidden"
or subliminal reasoning must be posited. Unlike inductive
knowledge,which also is necessary and involves a "hidden"
reasoning, this knowledge involves no generalization from
particular experiences.
Even though GhazalM'stheorizingabout the knowledgeof
the truthof statementsabout past events leaves certain difficulties unresolved,it demonstratesa firmgrasp of the kinds
of issues which must be faced if such a knowledgeis to be
posited.
Bernard WEISS
(Montreal)


Bernard Weiss - The Theory of Tawatur According To Ghazali

Cargado por

Información del documento

Título original

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Bernard Weiss - The Theory of Tawatur According To Ghazali

Cargado por

Copyright:

Formatos disponibles

Maisonneuve & Larose

Knowledge of the past: The Theory of "Tawtur" According to Ghazl

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

mena, past and present.) If I insistdogmaticallythat I know

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

the language of laymen). When one attends a university

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

'ilm has much the same sense as "knowledge" in ordinary

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

the words which he delivered to mankind as from God and

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

The term lawdiur means literally "recurrence." As a

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

the richertype of knowledgeof a past event which the eye

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

tion the recurrenceof statementsabout past events produces

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

one person after another) until it has occurred a sufficient

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

the occurrenceof knowledge:(1) that the number(of persons

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

expounded by Ghazalt, be termed mutawdtirin the technical

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

small enough to make collaborationpossible, can easily gain

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

we have heard the statementfrom a few persons but only

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

actuallyexistedI can arguefromthisthat the numberofpersons

were not adequate.

Obviously, extremelylarge numbersmust necessarilybe in

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

The theorybehindthe concept of lawdiurand its conditions

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

statementsabout past events arise. How are such disputes

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

us back to square one, forwe would be dealingwitha situation

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

ed to make statementsabout these events or phenomena.

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

are not arrived at throughdeductionbut are simply there as

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

It may be noted that if we combine the second and third

Implicitin the notionof "hidden" reasoningis a recognition

(1) Ed. Sulayman DunyA (Cairo: Dar al-Ma'arif, 1961), p. 187.

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

of any conscious operation of the intellect, then a covert

See Mustasfd, vol. I,

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM

KNOWLEDGEOF THE PAST

them all in (the perpetuationof) a falsehood;ratherthey can

This content downloaded from 132.206.27.25 on Fri, 27 Dec 2013 13:51:37 PM