Ilm Al Wad A Philosophical Account

Brill
ʿIlm al-waḍʿ: An Introductory Account of a Later Muslim Philological Science

Author(s): Bernard G. Weiss
Source: Arabica, T. 34, Fasc. 3 (Nov., 1987), pp. 339-356
Published by: Brill
Stable URL: http://www.jstor.org/stable/4056913
Accessed: 24-10-2015 04:08 UTC
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.
Brill is collaborating with JSTOR to digitize, preserve and extend access to Arabica.
http://www.jstor.org
This content downloaded from 137.189.170.231 on Sat, 24 Oct 2015 04:08:30 UTC
All use subject to JSTOR Terms and Conditions
'ILM AL-WAD':
AN INTRODUCTORYACCOUNT OF A LATER MUSLIM
PHILOLOGICALSCIENCE'
BY
BERNARD G. WEISS
Of the traditionalMuslimphilologicalsciences(al-'ulumal-lughawiya),
one whichhas receivedvirtuallyno attentionin the Westis 'ilmal-wad'.
This neglectmay perhapsbe due to the relativelylate emergenceof this
science, as comparedto the other Muslimphilologicalsciences,which
gives it the appearanceof being an appendageto the Muslimscholarly
tradition,an afterthought,as it were, of the medievalMuslimintellect.
Anotherpossible reasonfor this neglectmay be that from the point of
view of Westernphilology 'ilm al-wad'is somethingof a curiosity;it is
not so easily categorizedas the otherphilologicalsciencesand does not
seem to reflectany of the traditionalinterestsof Westernphilological
scholarship.'Ilm al-sarf, for example, is quite readily categorizedas
morphology, 'ilm al-nahw as syntax, 'ilm al-ishtiqdq as etymology,
'ilm al-lugha as lexicography, and so on. 'Ilm al-wad', on the other
hand, has no apparentcounterpartamong the branches of Western
philology or linguistics;as a matter of fact, its subjectmatter, which
will be unfolded in the course of the present study, does not seem
to fall within the domain of the Western philologist's,or linguist's,
interests. If there is any category of inquiry at all, known to the
West,that suits 'ilmal-wad',it is probablythat partof philosophywhich
is called?philosophyof language>>. GardetandAnawatihavereferredto
'ilm al-wad' as ose'mantique>>2.
1 My initialstudy of the subjectmatterof this articlemay be foundin the thirdmain

sectionof my doctoraldissertation(vLanguagein OrthodoxMuslimThought:A Studyof
Wad'al-Lughaand Its Development>,PrincetonUniversity,1966).The presentstudy
incorporatesa numberof key terminologicalchangesand includesobservationsthat were
not containedin the earlierstudy,althoughit does not go into as muchdetailon certain
points.
2 L. Gardetand M.-M. Anawati,Introductiona la theologie musulmane (Paris, 1948),
p. 135.
Arabica, tome xxxiv, 1987.
340 BERNARD G. WEISS [2]
The beginningsof 'ilmal-wad'may be tracedback to a short treatise

on the subject of wad' written in the fourteenthcenturyby the great
theologian 'Adud al-Din al-Iji (d. 757/1355) and entitledal-Risdlaal-
wad'iya.Upon this treatisea numberof commentariesand glosseswere
writtenin subsequentcenturies,the most renownedof these being the
commentariesof 'All ibn Muhammadal-Jurjini(d. 816/1413),'Ala' al-
Din 'All ibn Muhammad al-Qushjl (d. 879/1474), Abu'l-Qasim al-
Samarqandi (fl. 888/1483) and 'Isam al-Din al-Isfara'ini (d. 944/
1537)3.However,despite the abundanceof this literatureof commen-
tary,Tashkopruzadeas late as the mid-sixteenthcenturystatedthat 'ilm
al-wad'had not yet becomea <(codified>> (mudawwan) scienceand that
the treatiseof al-Ijiwas but a drop in the ocean of its subjectmatter4.
If we take the existenceof manuals(khuldsdtor theirequivalent)to be
the ear-markof a fully fledgedcodifiedsciencein Islam,then we should
perhaps assign the coming-of-ageof 'ilm al-wad'to the period after
Tashk6pruzade, whenthe firstmanualson wad'beganto appear.A num-
ber of such manualswerewrittenand publishedwithinthe last hundred
years. The following,for example,may be found in the Azhar mosque
library:MuhammadHajarzade,Risdlafi'l-wad' (Istanbul, 1890);'Abd
al-Malikal-Patni,'Aqdal-La'dli(Cairo:al-Matba'aal-Sharafiya,1887);
Yuisufibn Ahmadal-Dijwi,Khuldsatal-wad'(Cairo:Matba'atal-Nahda,
1915); 'abd al-Rahman Khalaf, Khuldsat'ilm al-wad' (Cairo: Mat-
ba'at al-Sa'ada,1916);'Abd al-Khaliqal-Shubriwi,al-Minhaal-ildhiya
fi'l-qawd'idal-wad'iya(Cairo: Matba'atal-Sa'ida, 19-); and Muham-
mad al-Husayni al-ZawThiri,al-Mulakhkhasft 'ilm al-wad' (Cairo,
19-)5. Until the early part of this century,'ilmal-wad'has occupieda
ratherhigh place in the curriculumof religiousstudiesin the Azhar6.
The termwad',whichindicatesthe subjectmatterof the scienceunder
consideration,is a verbal noun or infinitive,the ordinarymeaningof
which is ((to put down>>,?to place>>,?to posit>>,etc. In the technical
3 In my own studyof theseworksI havemadeuse of manuscriptsfoundat Princeton.

See R. Mach, Catalogue of Arabic Manuscripts (Yahuda Section) in the GarretteCollection,
PrincetonUniversityLibrary(Princeton,1977),pp. 293-7.As this catalogueindicates,al-
Qushjihas sometimesbeenconfusedwith al-Samarqandi. Cf. C. Brockelmann,Geschichte
derArabischenLitteratur(Leiden,1949),vol. II, pp. 268-9; vol. SII, pp. 288-9.
4 Ahmad Tashkopriizade, Miftdh al-sa'dda wa-misbah al-siyada (Haydarabad,1910),
vol. I, p. 110.
I These works are listed in Fihris al-kutub al-mawjsida
bi'l-naktaba al-azhariya (Cairo:
Matba'atal-Azhar,1946-52),vol. IV, as follows:Hajarzade,pp. 53-4; al-Patni,p. 58; al-
Dijwi,p. 51; Khalaf,p. 50; al-Shubrawi,p. 59; al-Zawahiri,p. 58.
6 Gardetand Anawati,loc. cit.
[3] 'ILM AL-WAD' 341
vocabulary of traditional Muslim philology, the term refers to the

<positing>of patternedvocal sounds (alfaz, literallyoemissions?)-to
be called, hereafter,vocables-as designationsfor particularmeanings.
Indirectly,therefore, the term refers to the invention of language,
although the object of wad' is normally alfaz rather than lugha
(<language>).What is immediatelyposited is all the vocables which,
taken together,constitutelanguage,with the resultthat languagein its
entiretytakes on a positedcharacter.These vocablesembracenot only
words as such but also such constituentsof languageas affixesans syn-
tactic structures.In the case of words which are made up of indepen-
dently significativeformal and materialcomponents(to be explained
later),two separatepositingsmust be postulatedfor each word, one for
the formalcomponentand one for the material.
The oposited>-we would now tend to say oconventional?-charac-
ter of languageis an importantassumptionin medievalMuslimthinking.
An attempt on the part of an early Muetazilitenamed'Abbad ibn
Sulaymanto attributelanguageto an innate predispositionin man to
imitatethe sounds of natureand to referto objectsby means of vocal
imitationsof soundsassociatedwith those objects,thus in effectplacing
languageentirelywith the naturalorder,was unanimouslyrejectedby
later Muslim thinkers.Of the two principaltheories of the origin of
languagewhichhad beenentertainedby Greekphilosophers,namelythe
so-calledphysis(tab) and thesis(wade)theories,Muslimlearnedopinion
unanimouslyopted for the latter. Though the identity of the original
inventoror inventorsof languagewas neveragreedupon, the essentially
supranatural,i.e. posited, character of language was raised above
virtuallyall doubt.The wordsky does not, it was held, signifythe object
so named becausethe very sound of the word evokes the idea of the
object;but for the fact that it has been positedas a designationfor that
objectit might as easily signifyearth or wateror fire7.
Thuslanguageis seento haveits originin a quitedeliberate,conscious
assigningof unitsof vocalsound,or vocables,to ideas(unitsof thought).
Languageis clearlythe productof mind, not its precondition.Behind
everysignificantunit thereis a Positor(wdd), or Inventor.Althoughthe
questionof whetherlanguageas a whole was the productof one such
Positor (God) or many (membersof a primordialhuman society) was
neverresolved,it was the practiceof writersin 'ilmal-wad'to speak of
7 See my <<TheMedievalMuslimDiscussionsof the Originof Language>,Zeitschrift

der DeutschenMorgenlandischen
Gesellschaft,Band 124,Heft 1 (1974).
<the Positor>>in a generic sense. This somewhatanonymousPositor

lurksin the backgroundof all medievalMuslimdiscussionof the positing
of language.
What is posited, in the final analysis,is a fixed relationshipbetween
two otherwiseindependentsets of entities:alfaz and ma'ani. Language
comes into being when the two are, so to speak, wedded. Meanings
(which before their associationwith vocables stand as mere ideas) are
conceivedas discrete,self-containedentities, to which vocablescan be
correlatedin an almost one-to-onefashion. Languageis the totality of
vocables togetherwiththe totality of their meanings.Separatethe two
andpresumablyone is left with meaninglesssoundson the one handand
non-verbalizedideas on the other. Of course, vocables, as patterned,
repeatedsounds,wouldnot existwithouttheirmeanings.The patterning
itselfis the work of mind.But a sound-patternmustbeginwith an initial
eruption of sound (suggestedby the word laft itself), which taken by
itselfis quitearbitraryand even accidental,consideringthe widerangeof
sounds which the human voice is capable of emiting. It is this freely
producedoriginalsound which (like the babblingof an infant)can be
consideredas intrinsicallymeaningless,as endowedwith meaningby the
Positor. The sound by virtue of its positing becomesan expression,a
patternedvocal representationof meaning.
Clearly,the positingof language,as conceivedby the writersin 'ilmal-
wad',amountsto a sort of name-giving.Everysignificantunit of sound
in a language,be it a word,an affix,a syntacticstructure,a word-formor
whatever, is viewed as having been assigned to a definite meaning
somewhat(but not exactly)in the way a parentassignsa nameto a child.
Thus man (insdn)is assigned to ?man>>,from (min) to ?from? (or,
perhaps better, <(from-ness?),-s (-un, -dt) to what might be called
opluralness>>, and so on. The meaningsof vocables other than words
may, when statedin the above manner,often appearawkward,but that
such meaningsexist is never deniedby the writersin 'ilmal-wad'.For
them it is inconceivablethat any constituent of language would be
posited except as the designationfor some meaning.
The assignmentof vocablesto meaningsby the Positoris consideredto
be final and irrevocable.Once a languagehas been posited,it has, in its
originalform, a normativecharacteramong the people who speak it.
New wordsmay be coinedand newmeaningsattachedto old words,but
only for specializedneeds and purposes; these new accretionsnever
become a part of the basic language of the people; they are merely
graftedonto it as a technicalidiom, suitablefor the variouscrafts and
sciences.The basiclanguageis all-sufficientso far as man'sfundamental

needs are concernedand requiresno alterationor amending.What has
been posited becomes a deposit,a given; it is, in fact, a sunna.In its
broad,non-religioussense,the termsunnameanssimplyopre-established
custom>>,i.e. that which has been posited in the past as a norm for all
subsequentactivity.The sunnais the totalityof givensin the sociallife of
man. Accordingly language constitutes a sunna, as the title of a
philologicalclassic,Ibn Faris'sal-Sahibiffifqhal-lughawa-sunanal-'arab
ft kalamihd,suggests.
The normativecharacterof the originalgivens of languageis rein-
forced by religiousconsiderations.One of the central affirmationsin
Islamis that God has spokenand that His Wordhas beenenshrinedin a
sacred Book. The purpose of all speech-God's or any other's-is
communication;God has spoken in orderto communicateHis Will to
man, and He expects man to understandand to comply in grateful
obedience. Thus the Book, together with the canons of Prophetic
traditionwhichinterpretits meaning,is a supremegivenupon whichthe
well-beingof mandependsutterly-and is to remainsuchuntilthe Day of
Resurrection.The Book addressesman in a particularlanguage,namely
the Arabiccurrentat the time of its revelation.Consequently,if manis to
understandthe Book he must know that languageand know it well.
Sincethat languageexistedindependentlyof and priorto the revelation,
it constitutes a given in its own right. If the Divine Word is to be
understoodand obeyedin all ages down to the Day of Resurrection,the
languagein whichit is expressedmustbe preservedintactthroughtime-
not only in books of philology,but also on the tonguesof people.Thus
for the communityof Muslimsthe (<languageof the Arabs>> (lisdnal-
'arab,al-'arabiya)representsa fixeddepositof vocablesand correspond-
ing meaningswhichmust neverbe alteredor abandonedso long as that
communityseeks to understandand do God's Will. If the Book is to
prevailto the end of time,its expressionsmusthave a fixedand constant
meaningavailableat all times to those who seek to live by its dictates.
This semanticconstancypresupposesthe absolutenormativenessof the
<original>> and thereforeessentialArabiclanguage.
In view of the foregoing,it can perhapsbe betterunderstoodwhy the
scholarsof late medievalIslamshouldhave devotedan entirescienceto
the systematicexplication(or exploration)of the positing of vocables.
The notion that languagein its entirelyis positedpromptedspeculation.
'Ilm al-wad' explores this notion to the fullest degree possible. This
entailsa breakingdown of languageinto its constituentelementsso that
the positedcharacterof each elementcan be demonstrated,the assump-

tion being that the posited characterof the whole presupposesthe
posited characterof its parts. Languageis thus an aggregateof posited
entities.Eachentityowes its existenceto an act of a Positor. 'Ilmal-wad'
is markedby its thoroughness.It wishesto leaveno elementof language
unaccountedfor. It is nothing less than the posited characterof the
totality of languagethat it seeks to explore.
As we will see, the constituentelementsof language-the alfdz with
theirma'aini-arenot all on one and the sameplane.Rather,they forma
hierarchy:simplerelementsarethemselvesconstituentsof morecomplex
elements,with the syntacticstructuresrepresentingthe highestdegreeof
complexity(in absorbingall otherelementswithinthem).At everylevel
of complexity,the correspondencebetweenalfdz, and ma'ani,physical
languageand thought,is assumed.
The analysiscarriedon in 'ilmal-wad'proceedsin two stages.First it
developstypesor categoriesof positing(aqsdmal-wad-);thenit considers
each of the principalclassesof vocablespostulatedby Arabiclinguistic
sciencewith a view to determiningthe type of positingupon whicheach
is based.
Thetypesof positingdevelopedin the literatureof 'ilmal-wad'fall into

two sets:
FirstSet
1. Al-Wad'al-'ammli-mawdiu' lahu 'amm(The generalpositing of a
vocablefor a generalmeaning).
2. Al-Wad'al-'ammli-mawdi'lahukhdss(The generalpositing of a
vocablefor a particularmeaning).
3. Al-Wad'al-khdssli-mawdui'lahukhdss(Theparticularpositingof a
vocablefor a particularmeaning).
SecondSet
1. Al-Wad'al-shakhsi(the specificpositing).
2. Al-Wad'al-naw'i(the genericpositing).
For the sake of brevityin the followingdiscussionI shall referto the

first set as TypesA, B and C and the second set as Types I and II.
The first set was first set forth by 'Ali ibn Muhammadal-Jurjani(d.
1413 A.D.) in his commentarieson Iji's al-Risala al-wadeiya.The
groundworkfor this set of types was laid by Iji himself,whose Risdlais
concernedprimarilywith Type B. Oncethis type had beenintroduced,it
was a naturalstep to the formationof the other types.
This firstset of typesclassifiesthe positingsof vocableswith reference
to two things:(1) a conceptwhichexistsin the mindof the Positorat the
time of the positingof a vocableand (2) the meaningwhichthe vocable
has in actual speech situations (that is to say, the meaning which it
acquiresas a resultof the positing).In the case of Type A and Type C
positing the concept in the mind of the Positor is identicalwith the
meaningwhichthe vocablehas in actualspeechsituations;in the case of
Type B positingit is not.
In attemptingto gain an understandingof this first set of types, the
best approachwill be to begin with Type B, the one with which Iji's
Risdlais primarilyconcerned.Oncethistypeis understood,the othertwo
will fall readilyinto place.
TypeB was developedby Iji as a meansof accountingfor the positing
of certainvocablesthe meaningsof whichcouldnot havebeen knownat
the time of their positing. We may take the pronoun he (huwa)as an
example.If we understandthe meaningof a vocableto be that to which
the vocable refersin an actualspeechsituation,then the meaningof he
can only be a particularmale individualwhom a speakerhas in mind
when using this pronoun. Unlike vocables such as man (insdn),which
have the same meaningin all speechsituations,he will have a different
meaningin differentspeechsituations,sincethe.speakeror speakerswill
not necessarilybe referringin every speechsituationto the same male
individual.How, then, can he be said to have been posited as the
designationof a meaning?If its meaningcan be knownonly in concrete
speech situations, then to what could it have been assigned in the
momentof its originalpositing,whichnecessarilytook placepriorto all
speechsituations?
Earlierthinkershad attemptedto overcomethis dilemmaby postulat-
ing generalideasas the meaningof vocablessuch as he. On this view the
Positor assignedhe to the idea of ?a single masculineobject which is
absent(ghd'ib)from the speechsituation>> 8. This idea is thereforeto be
consideredits meaning.oMeaning>> in this context has a strongresem-
I It is not clearwhatrelationshipis supposedto existherebetweengrammaticalgender

and sex. The 'ilmal-wad'literaturedoes not expatiateon this point.
blanceto lexicaldefinition.If one wereto look up the omeaning?> of he in

a lexicon,one would certainlynot expectto find an exhaustivelist of all
those particularsto whichhe refersin countlessspeechsituations;rather
one wouldexpectto find some sort of abstraction.Indeed,he does bring
to the mind,even apartfromits usagein actualspeechsituations,certain
vague contents.
Thisway of thinkingwas unsatisfactoryto Iji,principally,one gathers,
becauseit did not providethe properlinkagebetweenthe positingof he
and its real meaning in actual speech situations. If he was to be
considered,like man,to be assignedto the meaningwhich it has in the
minds of speakers,then anotherview of the matterWouldhave to be
developed.Accordingly,Il advancedthe theorythat he and its likes are
assigned to their meanings, not directly, but throughthe mediumof
general concepts. To state the matter more precisely:the Positor, in
positing he, contemplatesthe general concept of ?a single masculine
object which is absent from the speech situation>>,but instead of
assigninghe to this conceptas suchhe assignsit indeterminately for each
and everyparticularsubsumedunderthe concept.The phrase?each and
everyparticular>> is crucial;it is meantto be a translationof kull wdhid
min al-mushakhkhasdtbi-khusasih,a key phrase formulated by Iji
himselfin his Risdla.The Positor does not assignhe to a class, viz. the
class of all those particularssubsumedunderthe generalconcept.In an
actual speech situationhe obviouslydoes not refer to such a class; it
refersratherto a singleparticular.Consequently,it mustbe stressedthat
he is posited in such a way that when it is used one and only one
particularis to be understood.Abu'l-Qasimal-Samarqandielucidates
the phrase?each and everyparticular>> by meansof a disjunction.He is
said to be assigned to ?this particular,or that particular,or that
particular,etc.>> of the particularssubsumedunderthe generalconceptof
?a singlemasculineobjectabsentfromthe speechsituation>> . Sincehe is
thus assignedindeterminatelyto ?each and everyparticular?subsumed
underthe generalconcept,the identityof the particularto whichhe refers
in a given speech situation cannot be known from the vocable itself.
Rather, it is known from the context (qarina)of the speech situation
itself;it is the contextin whichthe vocableis used,not the vocableitself,
that enablesthe hearerto identifythe actualindividualobjectreferredto.
Sincethat whichthe Positordirectlyconsidersin positingthe vocable
9 Abu'l-Qisimal-Samarqandi,Sharh'ala'l-risilaal-wad'iya(Princeton:YahudaMs.
5997), fol. 12b.
he is a general concept embracinga large numberof particulars,the

positingitselfis classifiedas general(al-wad'al-'amm).However,in order
to distinguishthe positingof vocablessuch as he from that of vocables
suchas man,whichalso entailgeneralconcepts,the 'ilmal-wad'literature
addsto the formerthe furtherqualificationli-mawdu'lahukhdssin order
to make it clear that the generalconceptwhich the Positorconsidersis
not itself the meaningto whichthe vocableis assigned;that meaning,it
must be stressed,is particular(khdss).The full Arabic designationfor
this type of positingis, therefore,al-wad'al-'ammli-mawdu'lahukhdss.
Whatis to be understoodfromthis phraseis that a vocableis, by means
of a generalpositing (a positing undertakenthroughthe mediumof a
generalconcept),assignedto a particularmeaning.
In the case of a vocable such as man, on the other hand, the general
concept present in the mind of the Positor happensto constitutethe
actualmeaningto whichthe vocableis assigned(and whichit is to have
in actualspeechsituations);thereforeboth the positingand the meaning
are general.Hencethe designational-wad'al-'aimmli-mawdu'lahu 'amm
The remainingcategoryfollows close behind. The Positor of a proper
namesuch as Zayd has in his mindthe conceptof a particularindividual
and assigns the name to that concept10. Here again the concept
consideredby the Positor is identicalwith the meaningto which the
vocableis assigned;however,in this case both areparticularratherthan
general.Henceal-wad'al-khdssli-mawdii'lahukhass.Thusthe firstset of
types of positing-which I have chosen to referto as Types A, B and
C-is complete.
Whereasthe firstset classifiesthe positingsof vocableswith reference
to conceptsand meanings(whetherexistingin the mindof the Positoror
of the speakersor of both), the secondset classifiesthem with reference
to the vocablesthemselves.The termsnaw'iand shakhsi(translatedhere
as <generic>> and ?specific>>)are in themselves similar (though not
identical)in meaningto the terms 'ammand khass (here renderedas
?general>and ?particular>>). It is importantto understand,therefore,
10 It is unclearhow the relationshipbetweenthe Positor and, say, the parentwho

christensa childis to be understood.It is unlikelythat the two areseenas identicalby the
writersin 'ilmal-wad'.A parentemploysa pre-existentname in christeninga child; the
namehas, apparently,alreadybeenposited.Yet if thePositorpositsa namefora particular
individualwho does not existas the timeof the positing,it is difficultto understandexactly
what arisesin his mind.Unlikewordslike he, a namesuchas Zaydis not positedthrough
the mediumof a generalconcept.The Positorthereforeapparentlyconceivesof particular
individualsas unidentifiedentities and posits names (Zayd, Ahmad, Fatima, etc.) as
designationsfor them withouthimselflinkingany one nameto any one individual.
348 BERNARDG. WEISS [10]
that the two sets of termsdifferwith respectto criteriaof classification.

Though similarin meaning,they representtwo differentthough com-
plementaryapproachesto the classificationof the positingsof vocables.
As technical terms, these terms should thus be seen as conveying
muchmore than theirordinarymeanings;they shouldalwaysbe linked
to the methodsof classificationwhich they are meantto represent.
The secondset of types of positing(TypesI and II) may be explained
somewhat more succinctlythan the first set. The Positor may posit
vocables either individuallyor in groups; that is, he may posit a
particularvocable without referenceto any featurewhich the vocable
may have in commonwith othervocables,or he may focus his attention
on some feature-a patternor form-which a pluralityof vocablesmay
have in common and summarilyposit all those vocablesin which this
featurewill appear.In the formercase, the positingis specificin that it
has as its objectone vocableand no others.In the lattercase,it is generic
sinceit has as its objecta patternor formcommonto a groupof vocables
by virtue of which the group constitutesa genre, or class. Thus, for
example,the positingof the propername Zayd is specific(Type I) and
that of darib, katib, hdfz, 'alim, shdkir, kdflr and all other words that
adhereto thefdil form is, in respectto that form, generic(Type II).
It must be pointed out straightawaythat a vocable can never be
constitutedby means of a genericpositing alone, for such a positing
establishesonly one componentof a vocable: its form. Alongsidethe
formof a vocabletheremustexist its <matter?),whichowes its existence
to a positingdistinctfrom that of the form. The vocableis thus a union
of form and matter.This distinctionof form frommatter,whichwas of
coursea legacyfromthe Greekphilosophers,playsa fundamentalrolein
'ilm al-wad',not to mention the other branchesof Arabic philology.
Although the linguisticmatter postulatedby the philologistsmust be
distinguishedfrom the matter encounteredin the physics and meta-
physicsof the philosophers,it resemblesthat matterin that it constitutes
a principleof individuation.In themselves,forms are free-floatingand
transcendent;in matter they find their ?resting place?),become em-
bodied, individuated.
Vocableswhichare brokendown by the Arabicphilologistsinto form
and matter may be described,following Massignon'slead1",as hylo-
morphic.Theyincludemany(but not all, as we will see presently)words
" L. Massignon, (<La structure primitive de I'analyse grammaticale en Arabeo,

Arabica, I, fasc. 1 (1954), p. 12.
[11l ] 'ILMAL-WAD' 349
and all syntacticstructures(murakkabdt).In the case of hylomorphic

words, the form of the word consists of the vowels, sukansand ?addi-
tional>> (zd'id)consonants(thatis, conconantsotherthanthe radicalswhich
formthe root), whereasthe matter,or materialsubstrate,consistsof the
bare radicals. Any consonant may, of course, be a radical. Taken
together,the consonants(hurij)are,so to speak,the raw,inanimatestuff
out of whichwords are built throughthe superimpositionof animating
forms;they are a stock from whichradicalscan be selectedto form the
materialbasis of words.The wordddrib,for example,is madeup of the
fa'il form,representedin the worditselfby its vowels,and of the radicals
D-R-B, which constitutethe materialsubstratethroughwhich the fa'il
formis individuated,or concretized2. In the case of syntacticstructures,
the form consists of the syntacticpatternas such, be it a sentenceor
constructphraseor attributivephraseor whatever,whereasthe material
substrateconsists of the words that make up the structure;in this case
wordsratherthan radicalconsonantsarethe stuffupon whichformsare
superimposedand throughwhichthey are individuated.Matteris thus a
relativecategory.The case of compoundsconsistingof wordsplus their
suffixes are similar.The masculineplural suffix -u7n,for example,is a
form which becomesindividuatedby being attachedto a word such as
muslim,whichconstitutesits materialsubstrate.
The basis for breaking vocables down into formal and material
componentsis, in the 'ilmal-wad'literature,semantic.Form and matter
each have their separatemeaning,which they contributeto the total
meaningof the vocable.Eachthereforemust, as we have noted, have its
separatepositing,with the resultthat we must postulatea dual positing
in the case of hylomorphicvocables.In the case of certainvocables-all
of them words-whose form and radicals are deemed not to have
separatesemanticvaluesthe hylomorphicdualityis ignored.Withouta
correspondingdualityin meaning,the form-matterdistinctionloses its
significancefrom the point of view of 'ilmal-wad';it becomesa purely
superficialphenomenon having no relevanceto the analysis of the
positingof vocables.Take the propernamezayd, for example.One can
12
It must be emphasizedthat a form such as fa'il, though it may be describedas a
featurewhicha numberof words(e.g. ,drib,kdtib,etc.)havein common,cannotbe defined
as such;for if this werethe definitionof form,thenit couldbe arguedthatthe radicalsD-
R-B in the wordsdarib,daraba,darb,madrab,etc. constitutea form,for they areindeeda
featurewhicha numberof wordshavein common.Theform-matter distinctionin medieval
Muslimphilologyrestsupon the notion that it is the radicalsthat constitutea material
substrateand that what exists within particularwords over and above their radicals
constitutesthe formalcomponentof thesewords.
arguethat it has both a form(fact)and a matter(the consonantsZ-Y-D).

However,thesedo not have separatesemanticvaluesand thereforeneed
not be distinguishedfrom each other for the purposeof analysisof the
positingof the name.They both ariseout of one and the samepositing.
(The <adjectival>> characterof some names will not be gone into here.
Sufficeit to say that it has no bearingon the analysisof the positingof
namesqua names.)
For want of a better term, we will describewords which (despitea
superficialhylomorphicduality)arenot brokendowninto a semantically
distinguishableform and matteras non-hylomorphic.
The notion of a dual positing in the case of hylomorphicvocables
raisesa questionwhichmay have occurredto the reader.If the positing
of a vocable with respect to its form is generic(Type II), how is its
positingwith respectto its matterto be classified?The answerto this
questionwill unfold later in this study.
II
The typesof positingof vocablesdescribedaboveprovidethe basisfor

an exhaustiveaccountof the positingof language,whichis the primary
objectiveof 'ilmal-wad'.The presumptionbehindthis accountis that the
positingof any vocableis typablein accordancewith both sets of types.
The two sets, in combination,yield a typology consistingof six types,
whichmay be arrangedas follows:
1. Type IA (Al-Wad'al-shakhsial-'ammli-mawdiulahu 'amm).
2. Type IB (A1-Wad'al-shakhsial-'ammli-mawdu'lahukhdss).
3. Type IC (Al-Wad'al-shakhsial-khassli-mawdu'lahukhdss).
4. Type IIA (Al-Wad'al-nawi al-'ammli-maw.diu lahu 'amm).
5. Type IIB (Al-Wad'al-naw'ial-'ammli-mawdiu- lahukhdass).
6. Type IIC (Al-Wad'al-naw'ial-khdssli-mawdziV lahukhdss).
In principleone could give an accountof the positingof languageby
proceedingfrom vocableto vocable,typingthe positingof each one in
accordancewith the above typology, until all of the countlessvocables
which make up languagehad been accountedfor. This procedureis,
however, both impracticableand unnecessary.'Ilm al-wad' proceeds
ratherby dealingwith vocables,not individually,but by class. In this
way it is ableto producean exhaustiveaccountwithoutdescendingto the
level of individualvocables. Thus, ratherthan attemptingto type the
positingof zayd, 'amr,ahmad,'ali,etc., it typesthe positingof the proper
[13] 'ILM AL- WAD' 351
name (al-'alamal-shakhsi);from this typing one knows the typing of

individualpropernames.
Sincethe manualsof ilmal-wad'varysomewhatin theirtypingof the
positingof certainclassesof vocables,a full discussionof thesemanuals
would requirethat properattentionbe given to the differencesbetween
them in this respect.As this is not possiblewithinthe limits of a single
article,we will arbitrarilyselect one of the manuals as a sample and
considerits typingof the variousclasseswithoutattemptingcomparisons
with other manuals. It may be noted in passing, however, that the
differencesbetweenthe manualshaveto do primarilywith the last three
types of positing(TypesIIA, IIB and IIC).
The manualto be consideredhereis the Risdlafi'l-wad'of Muhammad
Hajarzade'3.Hajarzadetypes the classesof vocablesas follows:
Type IA: the genericnoun (ism al-jins);the infinitive(masdar);the
materialsubstrateof the verb (maddatal-fi'); the materialsubstrateof
the deinfinitival14 noun (mdddatal-mushtaqq).
Type IB: the personal pronoun (mudmar);the relative pronoun
(mawsul);the demonstrativepronoun(ismal-ishdra);the particle(harj).
Type IC: the proper name (al-'alamal-shakhsi);the generic name
('alam al-jins).
Type IIA: the form of the verb (hay'atal-fi') as a designationfor a
time-determination;the form of the deinfinitivalnoun (hay'at al-
mushtaqq);the elative(tafdio;the diminutive(tasghir);the relativesuffix
-i (nisba);the dual suffix(tathniya);the pluralsuffixor form (jamt);the
vocative(munda-).
Type ITb:the form of the verb (hay'atal-fi'l) as a designationfor an
ascriptivenexus; the sentencestructure(al-murakkab at-tdmm,literally
((thecompletesyntacticstructure>>); the constructphrase(al-murakkab
al-ki4df);the attributivephrase (al-murakkabal-tawsifi);the definite-
article-plus-nounphrase(al-muhalldbi-'al).
Type IIC: the morphologicalterminustechnicus(ism al-sigha).
It will be noted that the above schema embracesboth whole (non-
hylomorphic)words and formaland materialcomponentsof words, as
well as formalcomponentsof syntacticstructures.For the sake of clarity
13
See above,note 5.
14 The term is inspiredby Wright'sterm vdeverbal>> while at the same time being a
correctionof it. The noun in questionis, as we shall see, derivedfrom the infinitive,not
fromthe verb.See W. Wright,A Grammarof theArabicLanguage(Cambridge,1955),vol.
I, p. 100.
we may groupthe classesof vocablesconsideredby Hajarzade,or their

components,as follows:
Whole (non-hylomorphic) words: the proper name, the generic name,
the generic noun, the infinitive, the personal pronoun, the relative
pronoun, the demonstrativepronoun, the particle, and the morpho-
logical terminustechnicus.
Formal components of words: the form of the verb, the form of the
deinfinitivalnoun,the elativeform,the diminutiveform,the pluralform,
the dual suffix, the relativesuffix.
Material componentsof words: the material substrate (or matter) of the
verb, the materialsubstrateof the deinfinitivalnoun.
Formal components of syntactic structures: the sentence form, the
attributivephrase, the constructphrase,the vocative phrase(vocative
particle plus noun), the definite-article-plus-noun phrase (all these
phrasesbeing consideredqua forms).
A numberof points call immediatelyfor clarification:
1. Most of the formal components mentionedwithin this schema
embracea variety of actual morphologicalpatterns.The form of the
infinitiveis particularlydiversein respectto actual patterns.Only the
elative and diminutiveforms and the syntacticforms are limitedto a
single pattern.
2. The pluralform embracestwo main types of patterns:the broken
pluralpatternsandthe pluralsuffixes.The 'ilmal-wad'writerslumpboth
typestogetherunderthe one heading.Theysee no reason,apparently,to
give separatetreatmentto each, although it is worth noting, for our
purposes,that the brokenpluralpatternshave radicalsas theirmaterial
substrate,whereasthe pluralsuffixesare individuatedin whole worlds(-
un, for example,is individuatedin wordssuch as muslim).
3. It will be noted that of the twelveformalcomponentsof words or
syntactic structuresmentioned in Hajarzade'sschema the material
counterpartsof only two are mentioned:the materialsubstrateof the
verb and that of the deinfinitivalnoun. Why mention only these, one
may wonder, and not the materialsubstratesof other words (elatives,
diminutives and the like) or the material substrates of syntactic
structures?If the hylomorphicdistinctionis based on the notion of
separatepositings(and separatesemanticvalues),then why do we not
find in Hajarzade'sschema twelve materialcomponentsto match the
twelveforms?Why only two?
In answerto thesequestions,we maynote firstthat of the twelveforms
seven (the dual suffix, the relative suffix, the sentence structure,the
[15] 'ILM AL- WAD' 353
construct phrase, the attribute phrase, the vocative phrase and the
definite-article-plus-nounphrase)have wordsas theirmaterialsubstrate
and an eighth (the plural)has words as its substratein some cases, i.e.
when it consists of a plural suffix; and words are already included
within the schema in their own right, whether treated as wholes or
broken down into formal and materialcomponents.This leaves five
formswhichhave(or, in the case of the plural,mayhave,if it is a broken
plural) radicalsas their materialsubstrate.Of these five the material
substratesof three(the elativeform,the diminutiveformand the broken
pluralform) are not mentioned.Why not? Perhaps(and this is only a
guess) these three forms were viewed as being built either upon other
formswhose materialsubstratesare mentioned(i.e. deinfinitivalnouns)
or upon non-hylomorphicwords (having no distinct material com-
ponents)and as thereforenot requiringthat theirmaterialsubstratesbe
mentioned.The elative form, for example,is built upon an adjectival/
participalform(ismal-fdailor sifa mushabbaha bi-ismal-faiti),whichfalls
withinthe categoryof the deinfinitivalnoun. Sincethe materialsubstrate
of the deinfinitivalnoun is mentioned,there would (if this guess be
correct) be no need to mention that of the elative. Similarly, the
diminutivemay be built eitherupon a deinfinitivalnoun or (as is more
common) a non-hylomorphicword. Kulayb,for example,is built upon
kalb, which has no separate(semantically-based) materialcomponent.
The same is true of brokenplurals.
4. The class of the deinfinitivalnoun (mushtaqq)is a highly com-
prehensiveone whichincludesparticiples/adjectives (asmd'al-fa-'il,asma'
al-mafal and the sifat), nouns of instrument(asmd'al-dla), nouns of
place and time (asma'al-makdnwa'l-zamdn)and all other nouns which
are consideredto be deriveddirectlyfrom an infinitive(masdar).This
process of derivationis called by the Arabic philologiststhe (<minor
etymology? (al-ishtiqdqal-asghar).It is the etymology of forms, in
constrastto the <majoretymology?,whichis the etymologyof radicals
(i.e. materialsubstrates).The ?minoretymology?entailspurelyformal
transformations:the radicalsof the etymon(masdar)are retained,while
new formalelementsare introduced.
5. The infinitive(masdar,or ism al-fi') is treatedas non-hylomorphic
becauseof its status as an etymon. An etymon has no other function
semanticallythan to signifya simple,i.e. undivided,meaning.Its form
thereforehas no separativesignificativefunctionand henceno separate
positing.An etymon such as darbis thus on a par with nouns like kalb
and insdn,which also signify undividedmeaningswhich do not entail
separatesignificativefunctions for form and matter. When a noun is

derived from an infinitive, the meaning signified by the infinitive is
carriedover to the deinfinitivalnoun by meansof the radicals,while an
additionalmeaningis conveyedby a new form which is superimposed
upon the radicals. The result is a word with a semantically-based
hylomorphicdualism.Verbsalso arederivedfrominfinitivesin the same
manner.The minoretymologyis thus concernedwith the derivationof
both verbs and deinfinitivalnouns from infinitives.
Deinfinitivalnouns are so called in order to distinguishthem from
nouns which are not derivedfrom infinitives.Verbs, though as much
derivedfrom infinitivesas deinfinitivalnouns and as much deserving
thereforeto be calledmushtaqq,arenot normallyso calledsinceall verbs
are so derived, and there is no need to distinguish a category of
deinfinitivalverbs from a categoryof verbs which are not so derived.
Nouns, from the etymologicalpoint of view, fall into three categories:
infinitives(or verbalnouns),deinfinitivalnouns and oimmobile>> (idmid,
i.e. etymologicallynon-productive)nouns. Wrightdefinesthe immobile
noun as a noun
that is not itself a nomenactionisor infinitive,nor derivedfrom a nomenactionis,
and whichdoes not give birthto a nomenactionisor verb,as rajul,a man, batta,a
duck; opposedto ism mushtaqq,a noun that is derivedfrom a nomen actionisor
verbalroot, as kdtib,a writer,qdtil,slain' .
This definition,if the phrase<(whichdoes not give birth to a nomen

actionis (Wright'sdesignationfor the infinitive)or verb? is corrected
to read <which does not give birth to a deinfinitivalnoun or verb>>,
providesa good capsulizedsummationof the theoryof derivationunder
consideration(althoughWrighthimselfdoes not adherestrictlyto this
theoryin his own analysis).Morphologicalderivationas such turns on
two primarycategories:(1) nounsfromwhichverbsor othernounsmay
be derived,i.e. infinitivesand (2) the verbsand othernounswhichare so
derived.The <immobile>> noun (to which particlescan, by implication,
be assimilated)stands outside this basic etymologicalrelationship.It is
itselfneitherderived,nor can it serveas a sourcefromwhichotherwords
may be derived:it is etymologicallynon-productive,is not an etymon.
6. It will be noted that the form of the verb appears twice in
Hjarzade'sschema: under Type IIA and Type IIB. The basis for this
doubleclassificationof the formof the verbis the compositecharacterof
its meaning.A verb such as darabadoes not simply signify an act of
15 Wright, op. cit., vol. I, pp. 106-7.
[17] 'ILMAL-WAD' 355
striking(a meaningwhichit inheritsfromthe infinitivedarb,fromwhich

it is derived,and which is carriedby its radicals);it signifiesan act of
strikingas ascribedto somesubject(a subjectwhichis itself,however,not
includedin the verb's meaning)and as occurringin one of three time
categories,namelypast, presentor future).It is by virtueof these latter
componentsof meaning,signifiedby its form,that the verbgoes beyond
the significationof the infinitivefrom whichit is derived.Thusthe verb,
taken as a hylomorphicunity, signifies three meaning-components:
hadath(<<event>> or ?process>>: a generalcategoryunderwhich actions
such as strikingare subsumed),nisba(an ascriptionof the action to a
subject,i.e. the striker)and zaman(a time-determination). Hajarzade's
schemaclassifiesthe formof the verbone way with respectto the second
of these components(nisba)and anotherway with respectto the third
component(zamdn).Whetherwe are to understandfrom this that the
formof the verbis the resultof two separatepositings(suchthat the verb
as a whole-form togetherwith matter-is the productof a triplerather
thandual positing)or of a singlepositingwhichis doublyclassifiedis not
clear.Both possibilitiespresentdifficultieswhichwe will herepass over.
7. While the typing of the positing of relative and demonstrative
pronouns as Type B (al-wadeal-'amm li-mawdit lahu khdss) can be
readilyunderstoodon analogywith the explanation,givenabove, of the
Type B positingof pronouns(our examplewas he), the Type B positing
of the particle,the form of the verb as a designationfor an ascription,
and the syntacticstructurescallsfor comment.The meaningof a particle
is viewed as tied to a particularcontext: it is ma'ndfi ghayrihi.The
particlefi (<in>>),for example,does not have in actualdiscoursea self-
containedgeneralmeaning(somethinglike ?in-ness>>); rather,it always
has a particular?<meaning>> ((<inthe house>>,
?<inBaghdad>>, etc.). Thatis,
it alwaysrefersto a particular?in-ness>>, one that involvesa particular
place or entity. The Positor obviouslycannotenvisionall the particular
?in-nesses>>which will arise in innumerablespeech situations and
thereforemust posit the wordft for those particularinstancesof ?in-
ness>> by meansof the generalconceptof ?in-ness>> 16. As for the form of
the verb as a designationfor an ascriptionand the syntacticstructures,
here we have ?<ascription>>, or ?ascriptivenexus>>,as a meaning17.Like
16
The <meaning>> of particleswas a subjectof specialinterestin 'ilmal-wad and short
works were devotedespeciallyto it, for example,al-Risdlaal-harflya and al-Risdlaal-
mir'dtiya,both by 'All ibn Muhammadal-Jurjani(to be found in the PrincetonGarrett
Collection).
17 Thisconsideration of <ascription>>
(nisba)as the meaningof syntacticstructuresdoes
not seem,however,to applyto the vocativephraseor the definite-article-plus-noun phrase.
356 BERNARDG. WEISS [18]
oin-ness>,the <ascription>>underconsiderationin actual speechsitua-

tions will always be an ascriptionof one particularitem to another.
Sincethe Positorcannotenvisionall the particularascriptionsthatmight
arisein actual speechsituations,he posits the form of the verb and the
forms of the syntactic structuresby means of the general concept of
<<ascription>.
The foregoingsummaryaccountof the subjectmatterof 'ilmal-wad'is

not meantto be completeor to answerall questionsthat may arisein the
mind of the reader.Its purpose,rather,has been to give the reader,and
especiallythe Islamicist,a generalunderstandingof this hithertoover-
looked but nonethelessimportantMuslimphilologicalscience.To the
Western mind, schooled by modern linguistics and philosophy of
language,the outlookexhibitedin the 'ilmal-wadeliteraturewill perhaps
appearstrangeor even bizarre.It is thereforecruciallyimportant,if we
areto gain a genuineappreciationfor whatthe 'ilmal-wad'scholarswere
attemptingto do, that we place theirwork in the context of a religious
and culturaltraditionthat had a large stake in the positedcharacterof
languageand in the constancyof meaning.

Ilm Al Wad A Philosophical Account

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ilm Al Wad A Philosophical Account

Uploaded by

Copyright:

Available Formats

Brill

ʿIlm al-waḍʿ: An Introductory Account of a Later Muslim Philological Science

1 My initialstudy of the subjectmatterof this articlemay be foundin the thirdmain

Arabica, tome xxxiv, 1987.

The beginningsof 'ilmal-wad'may be tracedback to a short treatise

3 In my own studyof theseworksI havemadeuse of manuscriptsfoundat Princeton.

vocabulary of traditional Muslim philology, the term refers to the

7 See my <<TheMedievalMuslimDiscussionsof the Originof Language>,Zeitschrift

<the Positor>>in a generic sense. This somewhatanonymousPositor

sciences.The basiclanguageis all-sufficientso far as man'sfundamental

the positedcharacterof each elementcan be demonstrated,the assump-

Thetypesof positingdevelopedin the literatureof 'ilmal-wad'fall into

For the sake of brevityin the followingdiscussionI shall referto the

I It is not clearwhatrelationshipis supposedto existherebetweengrammaticalgender

blanceto lexicaldefinition.If one wereto look up the omeaning?> of he in

he is a general concept embracinga large numberof particulars,the

10 It is unclearhow the relationshipbetweenthe Positor and, say, the parentwho

that the two sets of termsdifferwith respectto criteriaof classification.

" L. Massignon, (<La structure primitive de I'analyse grammaticale en Arabeo,

and all syntacticstructures(murakkabdt).In the case of hylomorphic

arguethat it has both a form(fact)and a matter(the consonantsZ-Y-D).

The typesof positingof vocablesdescribedaboveprovidethe basisfor

name (al-'alamal-shakhsi);from this typing one knows the typing of

we may groupthe classesof vocablesconsideredby Hajarzade,or their

separatesignificativefunctions for form and matter. When a noun is

This definition,if the phrase<(whichdoes not give birth to a nomen

15 Wright, op. cit., vol. I, pp. 106-7.

striking(a meaningwhichit inheritsfromthe infinitivedarb,fromwhich

oin-ness>,the <ascription>>underconsiderationin actual speechsitua-

The foregoingsummaryaccountof the subjectmatterof 'ilmal-wad'is

You might also like