The history of
orientalism is quite peculiar. According to a few of them the history of Islam
and Muslims is quite possibly a lie. They also claimed that Arabic sources on
Islam are inherently unreliable whereas non-Islamic sources and speculative
opinions are given an aura of truthfulness. As far as the Qur'an is concerned,
it was not the revelation given to the Prophet, but simply a compilation of
stolen liturgical material from the mass of Judeo-Christian and Zoroastrian
traditions. One such example of an orientalist belonging to this class was that
of Reverend Alphonse Mingana. Mingana attempted to teach Muslims about the
transmission of their sacred Book down to even the Arabic alphabet! His hypothesis
was that the Qur'an had strong imprints of Syriac. The "author"
integrated a host of Syriac loan words into the language and thus brought about
the linguistic revolution of what is now called the Qur'an.[1] Mingana
catalogued the alleged "Syriac" vocabulary in the Qur'an and argued
for the widespread presence of Syriac Christianity and its important role in
the origins of Islam. His work, along with the more comprehensive work of
Arthur Jeffery's The Foreign Vocabulary Of The Qur'an,[2] gave impetus for
further research into the connection between the "foreign" vocabulary
of the Qur'an and the historical circumstances of its appearance. Recently,
Mingana's work was given a resurrection with a new twist by Christoph
Luxenberg's Die syro-aramäische Lesart des Koran: Ein Beitrag zur
Entschlüsselung der Koransprache.[3]
As far as the origins
of the Arabic language is concerned, Mingana claims complete ignorance about
it. He goes on to claim that in Makkah and Madinah, the written language
"must have been" either Syriac or Hebrew:
If all the signs do not
mislead us, very few oracular sentences, if any, were written in the time of
the Prophet. The kind of life he led, and the rudimentary character of reading
and writing in that part of the world in which he appeared, are sufficient
witnesses in favour of this view. Our ignorance of the Arabic language in its
early period of its evolution is such that we can not even know with certainty
whether it had any writing of its own in Maccah and Madinah. If a kind of
writing existed in these two localities it must have been something very
similar to Estrangelo [i.e., Syriac] or the Hebrew character.[4]
As for the Arabic
vowels, he dismisses the value of Arab authors and instead relies on Aramaic
writers and his own speculative opinions. He says:
The first discoverer of
the Arabic vowels is unknown to history. The opinions of Arab authors, on this
point, are too worthless to be quoted... If we may advance an opinion of our
own, we think that a complete and systematic treatise on these vowels was not
elaborated till the latter half of the VIIIth century, and we believe that such
an attempt could have been successfully made only the under the influence of
the school of Baghdâd, at its very beginning. On the one hand, besides the
insufficiency of the grounds for assuming an earlier date, we have not a
manuscript which can be shewn to be before that time, adorned with vowels; on
the other hand, the dependence of these vowels on those of Armaeans obliges us
to find a centre where the culture of the Aramaic language was flourishing, and
this centre is the school of Baghdâd, which was, as we have already stated,
under the direction of Nestorian scholars, and where a treatise on Syriac
grammar was written by the celebrated Hunain.[5]
He also asserted that:
The foundation of the
Arabic vowels is based on the vowels of Aramaeans. The names given to these
vowels is an irrefragable proof of the veracity of this assertion. So the Phath
corresponds in appellation and in sound to the Aramaic Phtâha....[6]
Following closely in
the footsteps of Mingana, Luxenberg claims that before the emergence of Arabic
literature, the principal language of writing was syro-aramäische or Syriac.
This lead him to assume that the origins of the literary Arabic and the Qur'an
must be sought in Aramaic and Christian communities. This assumption is taken
further to claim that Makkah was not an Arab settlement but an Aramaic colony
and that the residents of Makkah spoke aramäische-arabische Mischsprache.[7]
This language, apparently not known or understood outside of Makkah(?), soon
went into a state of oblivion and no reliable tradition existed to prove its
existence.[8] Hence, according to Luxenberg, the early Muslim scholars, writing
about a century and a half after the Prophet, were under the false impression
that the Qur'an was written in classical Arabic; therefore, it was no surprise
that they did not understand what they were reading.[9] In this regard,
Luxenberg represents a radical break from everyone else, including Jeffery and
Mingana.
Under the cloak of
these assumptions, Luxenberg begins his quest to find the "real"
Qur'anic text using his own graphic and linguistic methods. It is his
assumptions for the graphic side of his analysis that interests us in this
paper. By claiming that the early Arabic documents lack diacritical points and
vowel markers, Luxenberg takes liberty to alter diacritics and change the
vowels at will.
Luxenberg's work has
been given wide publicity by the New York Times (Alexander Stille and Nicholas
Kristoff), The Guardian and Newsweek. Is his book a path-breaking discourse or
is it yet another headline grabbing exercise? This has prompted us to evaluate
the claims of Luxenberg and inspect the foundations which these claims rest
upon. In this paper, we would like to examine the assumptions of Mingana and
Luxenberg concerning the origins of various aspects of the Arabic script. We
will also compare the Arabic script with the Syriac script and its development.
It will be shown that both Mingana and Luxenberg were wrong in their
assumptions concerning the Arabic script.
2. Origins Of The
Arabic Script
As mentioned earlier,
Mingana claimed ignorance about the evolution of the Arabic script and the
presence of an Arabic alphabet during the advent of Islam. He then went on to
say that in Makkah and Madinah, the written language "must have been"
either Syriac or Hebrew. As for Luxenberg, he claims that:
When the Koran was
composed, Arabic did not exist as a written language; thus it seemed evident to
me that it was necessary to take into consideration, above all, Aramaic, which
at the time, between the 4th and 7th centuries, was not only the language of
written communication, but also the lingua franca of that area of Western Asia.
As far as the history
of Arabic as a written language is concerned, it is best depicted by the
following pre-Islamic as well as early post-Islamic Arabic inscriptions that
show the progressive development of the Arabic script. The inscriptions below show
that the Arabic script before the advent of Islam clearly had a well-developed
alphabet.
Raqush Inscription (Jaussen-Savignac 17): The
Earliest Dated Pre-Islamic Arabic Inscription (267 CE).
Healey and Smith have
hailed it as the earliest dated Arabic document.[10]
Jabal Ramm Inscription: A Fourth Century
Pre-Islamic Arabic Inscription.
This inscription is the
second oldest so far discovered in the Arabic alphabet after the Raqush
inscription. The grammar in this inscription is straightforward classical
Arabic.[11]
A Pre-Islamic Arabic Inscription At Umm
Al-Jimal.
The grammar in this
inscription is straightforward classical Arabic.[12]
Zebed Inscription: A Pre-Islamic Trilingual
Inscription In Greek, Syriac & Arabic From 512 CE.
As the name suggests,
it is a trilingual inscription. The Arabic, though, does not translate the
Greek but merely lists six names, not all of which are mentioned in the Greek.
Jabal Usays Inscription: A Pre-Islamic Arabic
Inscription From 528 CE.
This is the only
pre-Islamic Arabic inscription with historical content.
Harran Inscription: A Pre-Islamic Arabic
Inscription From 568 CE.
A Greek-Arabic
bilingual inscription from Harran, near Damascus, Syria.[13]
Inscriptions Near Madinah Of The Early Years
Of Hijra [c. 4 AH].
Also see inscriptions
[B] and [C]. These set of inscriptions are from Mount Sal‘ near Madinah. They
have been dated to c. 4 AH using internal evidence.[14] The monumental script
(i.e., Kufic) in the above inscriptions is quite interesting. Grohmann compares
the script of the inscriptions with those in the early Kufic Qur'anic
manuscripts. He says that the resemblance is "very striking."[15]
These inscriptions
detailed above provide ample evidence of a well articulated Arabic alphabet and
are sufficient to refute the speculative assumptions of Mingana and Luxenberg.
Furthermore, Bellamy commenting on the inscriptions from Jabal Ramm, Umm al-Jimal
and Harran says:
Anyone who takes a
close look at these inscriptions and compares them with the sample of Koran...
will discern a great many letterforms that have not been changed at all, or
very little, in the sixteen hundred years that have elapsed since the earliest
one was written.[16]
We should also point
out that Nabia Abbott also refuted the arguments of Mingana using the earliest
known Arabic papyrus PERF No. 558 [22 AH] originating from Egypt. If Arabic was
indeed so primitive in its homeland during the advent of Islam, as claimed by
Mingana, how can one rationalize its practical use in Egypt in such a short
time and that too in a well-developed cursive script? Abbott says:
The condition of Arabic
writing in Muhammad's time is indicated by PERF No. 558 (our plates iv-v), an
Arabic papyrus of the reign of ‘Umar dated AH 22 and written in a fairly well
developed manuscript hand in the distant province of Egypt, where Greek and
Coptic were the written languages in general use. If written Arabic was so
primitive and rare in its own homeland at the time of Muhammad's death, how do
we account for its practical use in Egypt only a short dozen years after that
event? Again to grant the incomplete development of orthography would give us
reason to suspect only the orthographic accuracy of early Qur'anic editions but
not the possibility of their existence. In this connection it is interesting to
note that nowhere in the traditions of the earliest transmission of the Qur'an
is there any hint of serious orthographic or vowel difficulties; rather it is
the differences in the Arabic tribal dialects and differences arising out of
foreigner's use of Arabic that seem to demand attention. The foregoing
considerations lead one to believe that, if we allow for such common mistakes
as writers and copyists are liable to make, the Arabic writers of Muhammad's
time and of the time of early Caliphs were able scribes capable of producing an
acceptable edition of a written Qur'an despite the lack of all the improvements
of modern written Arabic.[17]
Luxenberg mentions the
pre-Islamic Arabic inscriptions in Grohmann's classic Arabische
Paläographie.[18] Deducing from the early form of Arabic alphabets, he says
that it is safe to assume the cursive syro-aramäische script [i.e., Syriac]
served as a model for the Arabic script.[19] What now becomes almost
unbelievable is that Luxenberg uses Grohmann's Arabische Paläographie as a
source to support his argument that the syro-aramäische script served as a
model for the Arabic script. Grohmann in this book, in fact, was one of the
earliest scholars to refute the origins of Arabic script from Syriac
script.[20] T. Nöldeke was the first to establish the link between the
Nabataean and Arabic scripts in 1865, which later confirmed against J.
Starcky's Syriac thesis by Grohmann. The affiliation between Nabataean and
Arabic scripts has now been fully documented by J. Healey. He says:
The development of the
Nabataean script in the 2nd, 3rd and 4th centuries A.D. is usually seen as a
progression from form derived from earlier Aramaic towards forms out of which
the early (western cursive) Arabic script developed, though we should note the
view of J. Starcky, based partly on the observation that Nabataean script,
unlike the Syriac and Arabic scripts, is essentially suspended from an upper
line, that the origin of the Arabic script is to be sought in a Lahmid form of
the Syriac script. This view has met with little support. The Nabataean origin
of the Arabic script is now almost universally accepted.[21]
Similar conclusions
were also reached by Nabia Abbott,[22] Kees Versteegh[23] and Beatrice
Gruendler.[24] One should also note that the origins of an early cursive Arabic
script has nothing to do with the syro-aramäische script of Luxenberg; rather
it is from the Nabataean script from where it originated. Now that Luxenberg's
hypothesis of the syro-aramäische script being the "model" for the
Arabic script is conclusively refuted, let us now move on to the origin of
diacritical and vowel marks in Arabic script.
3. Diacritical &
Vowel Marks In Arabic From Syriac?
The diacritical (or
skeletal) and vowel marks in early days of Islam were termed as nuqat (or
dots). Skeletal dots differentiate the graphemes or the letters sharing in the
same skeleton such as ح from ج. These are known as nuqat al-i‘jām and was
familiar to the Arabs prior to the advent of Islam. The vowel marks or nuqat
al-i‘rāb (or tashkīl), which can take the form of dots or conventional
markings, were invented by Abu al-Aswad al-Duali (d. 69 AH / 688 CE) as we
shall see below.[25] Let us now look into the issue of borrowing.
3.1 DIACRITICAL MARKS
It has been claimed by
scholars, with some reservations, that the origins of diacritical and vowel
marks originate from Syriac.[26] We have already seen the opinions of Mingana
earlier.[27] Luxenberg opines that the diacritical dots for ܕ
(dolath) and ܪ (rish) in Syriac may have served as the
basis for the Arabic alphabet.[28]
In the Syriac alphabet,
only two characters possess diacritical dots: ܕ
(dolath) and ܪ (rish). By comparison the Arabic
alphabet contains a total of fifteen dotted characters: ب، ت، ث، ج، خ، ذ، ز، ش،
ض، ظ، غ، ف، ق، ن، ة. Imagining that the Arabs borrowed their multitudinous dots
from the Syriac becomes a difficult proposition.[29] Moreover, we have clear
pre-Islamic evidence of the usage of diacritical dots, e.g., the Raqush
Inscription [267 CE] has diacritical points on the letters د، ش and ر; the
Jabal Ramm Inscription [4th century CE] has diacritical points for the letters ج،
ي and ن; and a curious inscription from Sakakah contains dots associated with
Arabic letters ب، ت and ن.
Coming to the time of
the advent of Islam, the earliest dated papyrus PERF No. 558 [22 AH / 642 CE]
shows numerous diacritical dots on the letters ج، خ، ذ، ز، ش and ن. Dotting is
also seen for the letters ز، ق and ن in P. Mich. 6714 - a bilingual papyrus
from 22 - 54 AH / 642 - 674 CE. There are also examples of diacritical dots in
the Islamic inscriptions, e.g., an inscription at Wadi Sabil [46 AH / 666 CE]
shows a dot below ب; and an inscription near Ta'if on a dam built by Caliph
Mu‘awiya [58 AH / 677 CE] shows the use of consonantal points for ي، ب، ن، ث، خ،
ف and ت.
Given the fact that all
of the above material was published before Luxenberg published his book, it is
surprising to see his claim (quoting Blachère) that Islamic tradition is unable
to pinpoint when the diacritical points were finally "fixed" - a
process that took over three hundred years.[30] If we take evidence from the
inscriptions and papyrus that predate ‘Uthman's mushaf we find that there are
ten dotted characters (out of fifteen) that have the same dot pattern as used
today. Not surprisingly, Gruendler, a specialist in Arabic script, using the
examples of inscriptions, papyri and coins from early Islamic times, says:
The diacritic system
had completed its development in the first half of the first Islamic century,
although points (or strokes) were used selectively and sporadically - being
regarded rather as an additional clarification than as an integral part of the
alphabet.[31]
We know that the
origins of Arabic script and the diacritical dots has nothing to do with the
syro-aramäische script of Luxenberg. Consequently, it would not be too
surprising that the diacritics may have come from the Nabataean script to the
Arabic script. Healey says:
... we may suspect that
the concept of diacritics came to the Arabs with the Nabataean script, ...[32]
It is clear that
Luxenberg is already incorrect on two counts, i.e., the origin of the Arabic
alphabet as well as the diacritical dots to differentiate between the letters
sharing in the same skeleton. Let us now see how he fares over the issue of
vowel marks.
3.2 VOWEL MARKS
As we have seen
earlier, Mingana had claimed that the origin of Arabic vowels is unknown to
history and said that the opinions of Arab authors are too
"worthless" to be quoted. Instead he advanced his own
"opinion" (worthless or otherwise) by saying that the foundation of
the Arabic vowels is based on the vowels of the Syrians. The only proof offered
by Mingana was the similarity in the names of vowels in Syriac and Arabic. The
fatha of Arabic corresponds in appellation and in sound to the Aramaic
phtâha.[33]
Luxenberg, on the other
hand, brings another dimension into the whole issue of vowel signs. He claims
that the Arabic vowel system for the designation of the short vowels a, u and i
by points, was after the model of the earlier syro-aramäische vocalization
system. It is also claimed that the addition of dots for the short vowels at
various locations was introduced in the reign of ‘Abd al-Malik b. Marwan [r.
685-705 CE]. Luxenberg applies his trial-and-error technique on the sab‘at
ahruf of the Qur'an and connects it to the seven vowel signs of Syriac, the
writing system developed by Jacob of Edessa [d. 708 CE]. Tabari [d. 310 AH /
923 CE] also mentions a tradition which says that there were five readings
(i.e., khamsah ahruf) of the Qur'an, which Luxenberg suggests correspond to the
five vowel signs of the Western Syrians.[34]
The common theme in the
arguments of both Mingana and Luxenberg is their use of speculation from which
they claim the Syriac origins of Arabic vowels. In other words, the Syriac
vocalization system was already in place before the Arabs borrowed it from
them. They differ only in their use of the sources. Mingana rejects the
opinions of the Arab authors as "worthless" whereas Luxenberg is all
too happy to embrace the opinion of an Arab author to support his hypothesis of
Syriac origins.
Let us first take the
case of Mingana. His only proof for the claim that the foundation of the Arabic
vowels is based on the vowels of Syrians is that the fatha of Arabic
corresponds in appellation and in sound to the Aramaic phtâha. Jacob of Edessa
[d. 708 CE] was the first person to introduce vowels in Western Syria.[35]
The use of the vowel
names, however, appear in the thirteenth century CE in Bar Hebraeus' [d. 1286
CE] writings. It is assumed that Bar Hebraeus may have followed the terminology
which had been in part introduced before by Jacob of Edessa, for the Syriac
sources on detailing origins of Syriac orthography and grammar are late. Elias
bar Sinaya [c. 11th century CE] was perhaps the first person to give the names
to the vowels in Eastern Syria (Figure 1). In Western Syria, the five named
Greek vowels appeared sometime after 839 CE, as suggested by the dated
manuscripts.[36]
Figure 1: Evolution of
Syriac vowels over the centuries.[37]
As for the Arabic
vowels, Abu al-Aswad al-Du'ali [d. 69 AH / 688 CE] was the first one to have
invented them. Ibn al-Nadim [d. 385 AH / 995 CE] in his Al-Fihrist says:
Abu ‘Ubaydah narrated:
Abu al-Aswad derived
grammar from ‘Ali ibn Abi Talib, for whom there be peace, but he did not
disclose to anyone what he had learned from ‘Ali, whose countenance may Allah
honor, until Ziyad [the governor of Basrah from 45 to 53 AH] appointed him for
the composition of something to serve as a guide to the people, so that they
could understand the book of Allah. Abu al-Aswad asked to be excused from this
task, until one time when he heard a reader recite, Allah is quit of the
idolators and of His Apostle [Qur'an 9:3, reading rasulihi, which should have
been read as God is quit of the idolators and so is His Apostle (reading
rasuluhu)]. Then he said, "I never supposed that the condition of the
people would come to this!" So he returned to Ziyad and said, "I will
do what the emir has ordered. Let there be sought for me a scribe who is
intelligent and obedient to what I say." They brought, therefore, a scribe
from the ‘Abd al-Kays Tribe, but he [Abu al-Aswad] was not satisfied with him.
Then they came with another one, about whom Abu al-‘Abbas al-Mubarrad said,
"I regard him to be one of those [who are intelligent]." So Abu
al-Aswad said [to the new scribe], "If you see that I open my mouth in
pronouncing a letter, place a mark above, on top of it. If I close my mouth
[making a u sound], place a mark in front of the letter, and if I split [my
lips] double the mark." So this was the marking system of Abu
al-Aswad.[38]
However, it is the Arabic
original that details the dotting that is of interest here. Abu ‘Amr al-Dani
says:
Muhammad Ibn Yazid
al-Mubarrid said : When Abu al-Aswad ad-Du'ali formalized the grammar, he said
: "Seek me a man skilled at taking notes." We looked for such a man,
and could not find anyone except in (the tribe of) ‘Abd al-Qays. Abu al-Aswad
told him : "When you see me pronounce a letter, if I bring my lips
together [fa-dammatu], put a dot before the letter; if I bring my lips together
[fa-dammatu] with nasalization, put two dots before it. If I lower my lips
[kasrtu], put a dot beneath the letter; if I lower my lips [kasrtu] with
nasalization, put two dots beneath it. If you see me open my lips [fathatu],
put a dot above the letter; if I open my lips [fathatu] with nasalization, put
two dots above it.
Abu al-‘Abbas
(Al-Mubarrid) said : Hence the dotting in Basrah remains in ‘Abd al-Qays until
now.[39]
Here we see that Abu
al-Aswad al-Du'ali [d. 69 AH / 688 CE] was responsible for the dot notation of
the three vowels and the nunation and that the names of the vowels (fatha,
damma and kasra) are connected with their articulation. Thus this first work of
Arabic grammar was an attempt to describe the structure of the language, an
accurate realization of the phonetic values of the short vowels.
Abu al-Aswad al-Du'ali
was the inventor of the vowels as well the names associated with their
articulation. Furthermore, Abu al-Aswad's scheme of vowelling is also seen in
some of the Qur'anic manuscripts from 1st/2nd century of hijra. Thus Mingana's
claim that the foundation of Arabic vowels is based on the vowels of the
Aramaeans becomes untenable. Moreover, we have already seen that the name
phtâha did not enter the Syrian phraseology until around the middle of the 9th
century; more than 150 years after the death of Abu al-Aswad al-Du'ali.
Furthermore, there are no corresponding appellations in Syriac for damma and
kasra of the Arabic vowel system. Hence Mingana's use of phtâha, just a single
vowel from Syriac, to claim that the Arabs borrowed their vowels from Syrians,
is rather disingenuous.
Similarly, Luxenberg's claim
that the designation of the short vowels a, u and i by points, was after the
model of the earlier syro-aramäische vocalization system and that they first
appeared in the reign of ‘Abd al-Malik b. Marwan does not hold any water
either. In the case of Syriac, the period from the third to the beginning of
the seventh century CE (i.e., until the advent of Islam) is marked by an almost
exclusive use of the diacritical point. Only towards the end of this period did
there begin to appear signs which denoted the vowels /e/ and /a/; prior to this
development, no specific signs were employed for the various vowels.[40]
In Syriac, a point is
placed either above or below the line and served a two-fold purpose. It was
used (a) to differentiate between the uses of y, w, ’ and h as vowel letters
and as consonant letters and (b) to differentiate between homographs. As for
the latter, a point can also differentiate homographs which have two vowel
phonemes in contrast: a upper point denotes a homograph which has a vowel of
the /a/ group as its distinctive phoneme, while a lower point (or the absence
of any point whatsoever) denotes a homograph which has a phoneme of the /i/ or
/u/ group.[41] This is out-of-step with the claim of Luxenberg that the
syro-aramäische vocalization system was used as a model for the designation of
the short vowels /a/, /i/ and /u/ in the Arabic script. The system of points in
Syriac was clearly inadequate and resulted in the evolution of two vocalization
systems, the Eastern (or Nestorian) and the Western (or Jacobite) as shown in
Fig. 1.
Like Mingana, Luxenberg
did not pay attention to the fact that the needs of the Arabic and Syriac
vocalizations were fundamentally different. The vocalization systems of Syriac
and Arabic had to cope with difficulties arising from the bivalent nature of
certain letters which had both consonantal (or semivocalic) and vocalic values.
The need to differentiate the former from the latter values existed in Arabic
as well as Syriac; it was, however, more acute in the latter, i.e., the Syriac.
In Arabic the situation is rather simple, since the letters ي، و and ء (i.e.,
w, y and ’) serve as vowel letters as a rule only when they do not have any
vocalization sign of their own. For Syriac, on the other hand, the differentiation
between the two kinds of values in question could not be based upon any such
automatic principle. Therefore, Syriac, unlike Arabic, had to develop other
methods to achieve the differentiation of vowel and consonantal notation.[42]
Hence, it is not surprising to see that the Arabic vowel notation took a
completely different route.
However, the course
that Arabic vowel notation finally took differs considerably from that of both
Aramaic and Hebrew. Of the Semitic languages which make use of vocalization
signs, Arabic is the only one to denote in its orthography all of the long
vowel phonemes but none of the short ones. Thus Arabic establishes a consistent
and clear-cut opposition between its internal and external vowel notation -
that is, between the values of its vowel letters and those of the respective
vocalization signs.... this opposition in marking is correlated with a
structural feature of language - the quantitative opposition of the vowel
phonemes.[43]
It is not surprising
that the dotting scheme adopted by Abu al-Aswad al-Du'ali was fundamentally
different from that of Syriac. Abu al-Aswad's scheme denoted the short vowels
(/a/, /i/ and /u/) and the nunation (/an/, /in/, /un/); as expected no long
vowels (/a:/, /i:/ and /u:/) were touched upon as the Arabic orthography takes
care of that. In contrast, in its most complete form the East Syrian system has
the signs for /a/, /a:/, /e/, /e:/, /i/, /o/ and /u/. The West Syrian system,
on the other hand, has signs for /a/, /å/, /e/, /i/ and /u/.
Let us now turn our
attention to another issue raised by Luxenberg. Firstly, he presents two hadith
traditions that deal with the differences of reading between the companions and
their resolution by the Prophet. These traditions relate to the revelation of the
Qur'an in the sab‘at ahruf or the seven modes.[44] This is represented by
Luxenberg as the indeterminate nature of the Qur'an's consonantal text. He also
argues that these are later stories which reflect what must have been a faint
recollection of the indeterminacy of the Arabic alphabet and hence the reading
of the text, which the later scholars interpreted as sab‘at ahruf. Secondly,
using his heuristic methodology, Luxenberg argues that the sab‘at ahruf of the
Qur'an allowed by the Prophet are connected to the seven vowel signs of Syriac,
the writing system developed by Jacob of Edessa. Furthermore, he cites a
tradition quoted by Tabari which says that there were five readings (i.e.,
khamsah ahruf) of the Qur'an, which, he suggests, correspond to the five vowel
signs of the Western Syrians.
The crux of the issue
here revolves around the dating of the tradition of the revelation of the
Qur'an in the sab‘at ahruf. This will then tell us whether the hadith is early
or late. If we plot the isnad (i.e., the chain of transmission) bundle of the
hadith of ‘Ubayy b. K‘ab (as mentioned by Luxenberg), we should be able to draw
some conclusions about the origins of this tradition of revelation of the
Qur'an in the sab‘at ahruf. The full text of the hadith reads:
Ubayy b. Ka'b reported:
I was in the mosque when a man entered and prayed and recited (the Qur'an) in a
style to which I objected. Then another man entered (the mosque) and recited in
a style different from that of his companion. When we had finished the prayer,
we all went to Allah's Messenger (may peace be upon him) and said to him: This
man recited in a style to which I objected, and the other entered and recited
in a style different from that of his companion. The Messenger of Allah (may
peace be upon him) asked them to recite and so they recited, and the Apostle of
Allah (may peace be upon him) expressed approval of their affairs (their modes
of recitation). And there occurred In my mind a sort of denial which did not
occur even during the Days of Ignorance. When the Messenger of Allah (may peace
be upon him) saw how I was affected (by a wrong idea), he struck my chest,
whereupon I broke into sweating and felt as though I were looking at Allah with
fear. He (the Holy Prophet) said to me: Ubayy, a message was sent to me to
recite the Qur'an in one dialect, and I replied: Make (things) easy for my
people. It was conveyed to me for the second time that it should be recited in
two dialects. I again replied to him: Make affairs easy for my people. It was
again conveyed to me for the third time to recite in seven dialects. And (I was
further told): You have got a seeking for every reply that I sent you, which
you should seek from Me. I said: O Allah! forgive my people, forgive my people,
and I have deferred the third one for the day on which the entire creation will
turn to me, including even Ibrahim (peace be upon him) (for intercession).
The isnad bundle of
‘Ubayy's hadith in the well-known collections of hadiths is given below.[45]
This hadith and its
variants in the form of a slightly shorter text or a change in wording are
extensively recorded in the well-known collections of Ahmad ibn Hanbal [d. 241
AH / 855 CE] in his Musnad, Muslim [d. 261 AH / 874 CE] in his Sahih, Abu Dawud
[d. 275 AH / 888 CE] in his Sunan and al-Nasa'i [d. 303 AH / 915 CE] in his
Sunan. They all predate Tabari [d. 310 AH / 923 CE]. However, one can argue
that the tradition of the revelation of the Qur'an in the sab‘at ahruf is not
present in collections earlier than the Musnad of Ahmad ibn Hanbal. This can be
countered by saying there exist independent traditions from ‘Umar b. al-Khattab
and ‘Abdullah b. ‘Abbas from the Prophet that mention the revelation of the
Qur'an in the sab‘at ahruf. The hadith from ‘Umar b. al-Khattab reads:
Narrated 'Umar bin
Al-Khattab:
I heard Hisham bin
Hakim reciting Surat Al-Furqan during the lifetime of Allah's Apostle and I
listened to his recitation and noticed that he recited in several different
ways which Allah's Apostle had not taught me. I was about to jump over him
during his prayer, but I controlled my temper, and when he had completed his
prayer, I put his upper garment around his neck and seized him by it and said,
"Who taught you this Sura which I heard you reciting?" He replied,
"Allah's Apostle taught it to me." I said, "You have told a lie,
for Allah's Apostle has taught it to me in a different way from yours." So
I dragged him to Allah's Apostle and said (to Allah's Apostle),
"I heard this
person reciting Surat Al-Furqan in a way which you haven't taught me!" On
that Allah's Apostle said, "Release him, (O 'Umar!) Recite, O
Hisham!" Then he recited in the same way as I heard him reciting. Then
Allah's Apostle said, "It was revealed in this way," and added,
"Recite, O 'Umar!" I recited it as he had taught me. Allah's Apostle
then said, "It was revealed in this way. This Qur'an has been revealed to
be recited in seven different ways, so recite of it whichever (way) is easier
for you (or read as much of it as may be easy for you)."
The isnad bundle of
this hadith and its variants in the form of a slightly shorter text or a change
in wording is drawn below.
This tradition of the
revelation of the Qur'an in the sab‘at ahruf through ‘Umar was recorded by
Ma‘mar b. Rashad[46] [d. 153 AH / 770 CE] in his Jami‘, Malik b. Anas [d. 179
AH / 795 CE] in his Muwatta,[47] al-Shafi‘i [d. 204 AH / 819 CE] in his Musnad,
Ahmed ibn Hanbal [d. 241 AH / 855 CE] in his Musnad, Bukhari [d. 256 AH / 870
CE] in his Sahih, Muslim [d. 261 AH / 874 CE] in his Sahih, Abu Dawud [d. 275
AH / 888 CE] in his Sunan, al-Tirmidhi [d. 279 AH / 892 CE] in his Sunan and
al-Nasa'i [d. 303 AH / 915 CE] in his Sunan.
This isnad bundle shows
that the earliest known occurence of this hadith is in a collection of Ma‘mar
b. Rashad. In other words, this hadith was already known and in circulation in
the first half of the second Islamic century, if we consider the death of
Ma‘mar as a terminus post quem.
Is that the final word
on the dating of this hadith? It is also interesting to note that the isnads in
the above bundle intersect at Muhammad b. Muslim, i.e., Ibn Shihab al-Zuhri [d.
124 AH / 741 CE]. He is the common link. This is also corroborated by studying
the isnad as well as by comparing different matns (i.e., the text of the
hadith) as seen in various collections.[48] A very short matn of the sab‘at
ahruf hadith also exist that has al-Zuhri as the common link:
Narrated Ibn Abbas:
Allah's Apostle said, "Gabriel read the Qur'an to me in one way (i.e.
dialect) and I continued asking him to read it in different ways till he read
it in seven different ways."
One can claim that
al-Zuhri might have invented the sab‘at ahruf tradition and circulated it
widely as he was the common link.[49] However, there are arguments which speak
against the assumption that al-Zuhri invented them outright.[50] This is
because he received information not only from ‘Urwa b. al-Zubayr [d. 94 AH /
712 CE] but also ‘Ubaydullah b. ‘Abdullah [d. 126 AH / 743 CE]. ‘Urwa received
the hadith from ‘Abd al-Rahman b. ‘Abd al-Qari and al-Masoor b. Mukhramah; they
both heard it from ‘Umar b. al-Khattab. ‘Ubaydullah on the other hand, heard it
from ‘Abdullah b. ‘Abbas. Hence we have two traditions from al-Zuhri going back
to two different companions of the Prophet as two different chains.
Furthermore, if we compare the two traditions of the revelation of the Qur'an
in the sab‘at ahruf from ‘Ubayy and ‘Umar, al-Zuhri is absent in the isnad of
the former. This again corroborates that al-Zuhri, being the common link in the
hadith from ‘Umar, could not have been the forger of the hadith of the
revelation of the Qur'an in the sab‘at ahruf. In other words, we can safely
conclude that the hadiths of the revelation of the Qur'an in the sab‘at ahruf
were already in circulation in the first Islamic century before the death of
‘Urwa b. al-Zubayr [d. 94 AH / 712 CE].
If we compare this
dating with the fruits of Luxenberg's heuristic methodology suggesting that the
sab‘at ahruf are connected to the seven vowel signs of Syriac, the writing
system developed by Jacob of Edessa [d. 708 CE], we find that the connection
between sab‘at ahruf and the seven vowels of Syriac is patently false. What has
the recitation of a book in seven modes got to do with the seven vowels of
Syriac? Nothing. Furthermore, Jacob of Edessa's scheme of vowels did not gain
any currency among the West Syrians and remained unexpressed until Bar Hebraeus
[d. 1286 CE].
The precise vowel
sounds of Jacob's scheme continued, however, to remain unexpressed in writing
by the West Syrians. His innovations, as he had expected, were not accepted
into the conventional alphabet. They do not appear in any manuscripts except
those of his grammar, and they are ignored until the time of the industrious
Bar Hebraeus. Why was Jacob's experiment never more than a gallant interlude?
We can only surmise. Perhaps Syrian conservatism rebelled against the tampering
with the traditional form of the Bible text, which was the very foundation of
all literary, as of all theological, activity in Syriac. Perhaps the new system
would have disrupted too violently the fundamental structure of verbal and
nominal ground-forms....[51]
Are we supposed to
expect that a vowel scheme that did not even gain currency in West Syria in
Jacob's time to have influenced the tradition of the revelation of the Qur'an
in the sab‘at ahruf? Contemplating the full spectrum of evidence available to
us, the answer is clearly no!
It should also be
mentioned that Abu al-Aswad's scheme was transmitted by Yahya b. Ya‘mar [d. 90
AH / 708 CE], Nasr b. ‘Asim al-Laythi [d. 100 AH / 718 CE] and Maimum al-Aqran.
It was al-Khalil b. Ahmad al-Fraheedi [d. 170 AH / 786 CE] who finally replaced
the pattern of dots with specific shapes for the three short vowels; a small و
for the vowel /u/, a small ا for the vowel /a/, and a small part of ي for the
vowel /i/. He also changed the sign of shadda, using a small س. Al-Khalil's
scheme gained rapid popularity and as Versteegh puts it:
With al-[K]halil's
reform, the system of Arabic orthography was almost completed and, apart from a
very few additional signs, it has remained essentially the same ever since.[52]
On the other hand, as
we have observed, the standardization of the vocalisation system in Western and
Eastern Syria was less rapid than that of Arabic.
3.3 DOTTING IN THE
EARLY QUR'ANIC MANUSCRIPTS
Since diacritical marks
were already fixed before the first half of the first century of hijra and the
vowel marks invented a little later by Abu al-Aswad al-Du'ali [d. 69 AH / 688
CE], it should not be surprising to see the Muslims towards the end of first
century of hijra were already using the dotted manuscripts. Abu ‘Amr al-Dani
narrates a couple of reports that shed some light into this matter.
It was to us narrated
that Ibn Sirin owned a mushaf that was dotted by Yahya Ibn Ya‘mur [d. 90 AH /
708 CE]. And that Yahya was the first one to dot them. The three of these people
are among the eminent successors of Basra.[53]
The other report says:
Khalaf b. Ibrahim said:
Ahmad al-Makki told me: Al-Qasim told me:‘Abd al-Rahman b. Mahdi told me from
Hammad b. Zayd; from Khalid al-Hadda': I used to follow a mushaf from Ibn Sirin
[d. 110 AH / 728 CE] that was dotted.[54]
It should be added that
every centre appears to have practiced a slightly different convention at
first. For example, Ibn Ushta reports that the mushaf of Isma‘il al-Qust
[100-170 AH / 718-786 CE], the imam of Makkah bore a dissimilar dotting scheme
when compared with the ones used by the Iraqis.[55] The scholars of Sana‘a'
followed yet another framework.[56] However, by the close of the first century,
the Basran convention became so popular that even the Madinan scholars adopted
it.[57] It is not surprising to see that some of the Qur'anic manuscripts from
1st/2nd century of hijra show the evidence of a vowelling scheme adopted by Abu
al-Aswad al-Du'ali. The frequency of diacritic dots and vowels signs varies and
alongside fully-vowelled manuscripts one can find texts in which even the
diacritic dots are left out.
4. The Cover Story
Should Luxenberg's book
be judged by its cover?
Figure 2: The front
cover of Luxenberg's book
It is strange that a
book that boasts a title Die syro-aramäische Lesart des Koran: Ein Beitrag zur
Entschlüsselung der Koransprache (The Syro-Aramaic Reading of the Qur'an - A
Contribution to the Deciphering of Qur'anic Language) has a cover page that does
not commensurate its title. One would expect that the author would have
unearthed an important piece of evidence in the form of a manuscript, or an
inscription to show the evidence of syro-aramäische reading of the Qur'an. Such
an evidence on the cover page of the book would have befittingly matched the
flowery title. However, to everyone's surprise the title page is from a first
century Qur'anic manuscript MS. Arabe 328a located at the Bibliothèque
Nationale, Paris.[58] A facsimile copy of this manuscript was published by
Déroche and Noseda in 1998.[59] This manuscript is written in a hijazi script,
with no vowels and rare diacritical points.[60] Even more damaging to the
thesis of Luxenberg is that a recent study on this manuscript has concluded
that this hijazi manuscript is written in the qira'at of Ibn ‘Amir [d. 118 AH /
736 CE] - one of the readings later to be declared indisputably mutawatir by
Ibn Mujahid [d. 324 AH / 926 CE].[61] Even though there are no vowel marks and
a rare diacritical mark in MS. Arabe 328(a), there is the consonantal outline
of the text and, in a series of fragments as extensive as these, there are,
fortunately, enough consonantal variants to enable the precise determination of
the reading.
A related manuscript to
MS. Arabe 328(a) is MS. Or. 2165 at the British Library, London. They both lack
vowels. Unlike MS. Arabe 328(a), in MS. Or. 2165, the consonants are relatively
frequently differentiated by dashes, thus allowing the identification of
several more variants where a single consonantal shape is pointed in a
distinctive way. The study by Dutton has shown that this manuscript is
remarkably similar to first century manuscript MS. Arabe 328a in Bibliothèque
Nationale, Paris, and was written in the qira'at of Ibn ‘Amir.[62] Based on the
similarity between MS. Arabe 328a and MS. Or. 2165, he suggests re-dating this
manuscript to the time just before Umayyad Caliph Walid [r. 86-96 AH], i.e.,
within the period 30-85 AH with the latter end of this time scale being safer.
Other early Qur'anic manuscripts up to 3rd century of hijra that show the
evidence of the qira'at are KFQ93[63] (2nd century AH), KFQ28[64] (3rd century
AH), KFQ19[65] (3rd century AH) and KFQ16[66] (3rd/4th century AH), all of them
at the Nasser David Khalili Collection Of Islamic Art, London, United Kingdom.
We should also add Sibawayhi's [d. 170-180 AH] interest in the qira'at, whether
attributed to any authority, or purely hypothetical. This interest did not
clash with Sibawayhi's respect for the already established Qur'anic text.
Brockett concludes from his study of Qur'an readings in Sibawayhi's Kitab:
that while this earlier
stage was far freer from tradition and far less systematised, it was
nevertheless one in which the Qur'an text was firmly set within surprisingly
narrow bounds.[67]
According to Luxenberg,
the Arabic alphabet used in the Qur'an began as some kind of shorthand, a
mnemonic device not intended as a complete key to the sounds of the language.
He then concludes that the transmission of the text from Muhammad was not
likely an oral transmission by memory, contrary to one of the chief claims of
Islamic tradition.[68] Assuming that this is true, then how does one explain
the early evidence of the qira'at in the Qur'anic manuscripts right from the first
century of hijra? For Luxenberg's theory to work, the Qur'an has to be two
different things at the same time: on the one hand, a paleographically frozen
seventh century document that represents the work of Muhammad, and on the
other, a garbled text that has been modified by later Muslim scholars who were
clueless as to its meaning. It can't be both at the same time. Moreover, we
have already seen that the Muslims in the first century of hijra were already
involved in diacritical marks as well as vocalization of the Qur'an to ensure
the correct transmission of the Qur'an in written as well as in oral form.
Furthermore, Whelan's study of the Qur'anic inscriptions on the Dome of the
Rock and the literary sources mentioning the Qur'anic inscriptions in the Prophet's
mosque in Madinah and the presence of professional copyists of the Qur'an has
already demonstrated the evidence of codification of the Qur'an in the 7th
century or the first century of hijra.[69] Given these facts, Luxenberg's
position of an unstable Qur'anic text in the first two centuries of hijra
becomes untenable.
5. Now The Evidence!
The basic premise of
Luxenberg's book, namely that the Qur'an was written in a hybrid Arabic-Aramaic
language and borrows from Christian Aramaic writings - is dedicated to giving
examples from the Qur'an to support this premise. The argument is essentially
circular. In order to document his big idea of the Qur'an being an
Arabic-Aramaic document that draws on Christian Aramaic texts, he ignored
whatever did not fit - a common problem with holders of grand ideas. This has
resulted in him making claims that are contrary to well-established facts such
as the syro-aramäische script served as a model for the Arabic script and that
the Arabic vocalization was based on the model of the earlier syro-aramäische
vocalization system.
To further his grand
ideas about the Qur'an, Luxenberg claimed that Makkah was not an Arab
settlement but an Aramaic colony and that the residents of Makkah spoke
aramäische-arabische Mischsprache. He also claimed that Aramaic, just before
the advent of Islam, was not only the language of written communication, but
also the lingua franca of that area of Western Asia.
On the contrary, Arabic
was a lot more widespread in the Middle East before Islam than Luxenberg allows
for, and we have sufficient evidence for this in the form of inscriptions
(Figure 3).
Figure 3: Geographical
distribution of pre-Islamic Syriac, Nabataean and Arabic inscriptions.[70] Each
coloured dot represents where the respective inscriptions were found giving no
importance to their numbers. For example, in Petra (Nabataean) and Edessa
(Syriac) and their surrounding areas, literally hundreds of inscriptions were
discovered. They all are represented by a single coloured dot in the above
figure.
The geographical spread
of pre-Islamic Arabic inscriptions range from Zebed from the Syriac speaking
heartland in the north to Mada'in Salih in the south and from Abu Darag (Egypt)
in the West to Sakakah in the East. Syrian Aramaic or the Syriac was the
language which Luxenberg says the Qur'an was partially written in. The bulk of
the pre-Islamic Syriac inscriptions are confined to the Edessa region in modern
south Turkey. It is certainly a long way from the hijaz region and in particular
Makkah! The pre-Islamic Syriac inscriptions south of Damascus are almost
non-existent (an exception being the one at Jabal Usays, south east of
Damascus), except those written by travellers or pilgrims.[71]
Aramaic was not as
widespread by the late sixth century CE contrary to what Luxenberg claimed.
Greek had replaced it in many areas as the principal lingua franca of the
Middle East.[72] After Alexander opened the Near East to Greek language and
culture, the distribution of Hellenism throughout the cities and countryside
was irreversible. By the time Islam arrived on the world scene, Greek had
already became the predominant language of Western Asia. Bowersock points out
that:
... the powerful impact
of Greek culture can be seen everywhere a few centuries later in the early
Roman imperial province of Syria. The caravan city of Palmyra was completely
bilingual... But in the villages and rural areas of central and southern Syria,
Greek can once again be seen as the language of local piety, uniting worshippers
from shrines and holy places that lay far apart. It is too little appreciated
that in general Greek inscriptions are far more common in the countryside of
late antique Syria than Syriac ones... On the eve of Islam, Hellenism continued
to be a powerful force. Greek-speaking Christians had never been able to set
themselves altogether free from it, and Syriac-speaking Christians discovered
that they could not exist without incorporating it into their literature and
language.[73]
This is self-evident
when one surveys the material, especially the Greek inscriptions, where around
90% of those from the sixth-century Near East are in Greek.[74] Syro-Aramaic or
Syriac, the language which Luxenberg says the Qur'an was partially written in,
was principally spoken in the Edessa region, modern south Turkey, a long way
from the hijaz region and in particular Makkah. The shaded areas in Figure 4
show very approximately the regions where Aramaic was used during the time when
it flourished from c. 900 BCE to the Arab conquest and where the Aramaeans
(i.e., Syriac speakers) are known with fair certainty to have settled in this
period.[75] The evidence is provided by external references to the Aramaeans
and distinctive Aramaean names and religion. It is seen that there is a discrepancy
between the two features. In large areas where Aramaic was used it is clear
that the native populations were not Aramaean. The Persians and Parthians are
the clearest example and Indian rulers also used Aramaic.
Figure 4: Geographical
distribution of Aramaic and the Aramaeans (the people who spoke Syriac).[76]
Comparing Figs. 3 and 4
we again find that most of the Syriac speakers (i.e., Aramaeans) were confined
in between the Euphrates and Tigris rivers and to the west of Euphrates; lying
across the modern frontier of Turkey and Syria. Although Figs. 3 and 4 were
drawn using different sources, they correspond very well and thereby provide
corroborative evidence.
The closest to Arabic
inscriptions in terms of geography are the Nabataean inscriptions and this
proximity makes Luxenberg's hypothesis of syro-aramäische reading of the Qur'an
untenable as we shall soon see. The Nabataeans were in fact Arabs and they
spoke a non-Aramaic north Arabic dialect akin to the Classical Arabic.[77] The
Nabataean Aramaic was effectively a Schriftsprache, i.e., the language of the
inscriptions. The early Arabic inscriptions such as the ‘En ‘Avdat inscription
[c. 125 CE] and the Namarah inscription [328 CE] were written in the Nabataean
Aramaic script but in the Arabic language. The Raqush inscription [267 CE] is
the earliest dated Arabic text with Aramaic archaisms. This shows that the
Arabs were familiar with Aramaic. The distinctive feature of Nabataean by
comparison with other Aramaic dialects is its Arabic colouring or, to be
precise, colouring from an Arabian language allied in some way to what became
Classical Arabic.[78] Healey notes that some other early though undated
Nabataean inscriptions may in fact be in Arabic and that some of the thousands
of Nabataean graffiti are arguably in Arabic.[79] Aramaic had been very
important in the hijaz as early as the fifth/fourth centuries BC as seen from a
stela from Tayma in northern Saudi Arabia inscribed in Aramaic.[80] So, it
would be surprising if there were not quite a number of Aramaic words in the
hijazi dialect of Arabic, and that does not make it a mixed language.
Most, if not all, of
the socio-historical context of the origins of Islam can be explained by
considering the Nabataean milieu. In the hijaz, we principally have Nabataean
inscriptions, and these do not seem to be Christian at all in their
content.[81] The Nabataeans worshipped idols of gods and goddesses such as
Dushara, Allat, al-‘Uzza, Manat, Hubal, et al. some of which were also
worshipped by Arab tribes such as Quraysh in the hijaz as mentioned in the
Qur'an and in the Islamic literary sources.[82] A Syro-Aramaic Christian milieu
would have a hard time explaining this socio-historic context. The Nabataean
origins of the Arabic script further strengthens the case for a Nabataean
milieu.
6. Syriac In The Early
Islamic Centuries
Mingana and Luxenberg
have claimed that since Syriac provided the literary exemplars from which
Muhammad (or the early compilers?) worked, Syriac grammar and vocabulary should
be used to interpret it. This comes from their assumption, whether stated
openly or tacitly, that Syriac unlike Arabic was older and hence was already
fixed in orthography, lexicography and grammar. According to Mingana:
As we believe the
Kur'an to be the first Arabic book, its author had to contend with immense
difficulties. He had to adapt new words and new expressions to fresh ideas, in
a language that was not yet fixed by any grammar or lexicography... so the
author of the Kur'an has exhibited stylistic idiosyncrasies which stamp his
work as being somewhat different from the classical Arabic known to us from the
eighth century downwards; his style suffers from the disabilities that always
characterise a first attempt in a new literary language which is under the
influence of an older and more fixed literature. This older and more fixed
literature is, in our judgment, undoubtedly Syriac more than any other.[83]
It is true that the
Qur'an exhibits certain stylistic features not seen in the pre-Islamic poetry.
However, to claim that this is different from the Classical Arabic of eighth
century Arabia is problematic.
On the other hand,
Luxenberg's aramäische-arabische Mischsprache is an ill-defined concept and
this has given him an excuse to claim that the normal rules of neither of the
Arabic and Aramaic applies in a particular passage of Qur'anic text. This frees
him to make conjectures at whim about the meaning of a particular passage in
the Qur'an, even though those suggestions would otherwise be rejected as
ungrammatical. The hypothesis of a aramäische-arabische Mischsprache or a
"mixed language" appears to be little more than a convenient excuse
for high-handed interpretation of the Qur'anic text. As one can see, this hypothesis
leads to more problems, historically and linguistically, than it claims to
solve. Like Mingana, one of the biggest drawbacks of Luxenberg's hypothesis is
its inability to explain the origins of Classical Arabic. If one stays within
the bounds of the Nabataean milieu, the pre-Islamic existence of Classical
Arabic is easily documentable using the well-studied Namarah inscription.
Classical Arabic has
its origins pre-dating the Islamic period. The Namarah inscription [328 CE] is
one of the earliest inscriptions so far discovered in the classical Arabic
language and is written in Nabataean script. Bellamy, who had studied this
inscription, concludes by saying:
... we have added a
century and a half to the life of classical Arabic. This should come as no
surprise since the conservatism of the Arabic language is well known.[84]
The ‘En ‘Avdat
inscription [c. 125 CE], even earlier than the Namarah inscription was
apparently written in Classical Arabic. Furthermore, the two pre-Islamic Arabic
inscriptions from Jabal Ramm [4th century CE] and Umm al-Jimal [5th / 6th
century CE] have grammar that is straightforward classical Arabic. The language
in these inscriptions is closer to modern Arabic than the language of
Shakespeare is to modern English.[85] What can be concluded is that classical
Arabic has its origins even before the second century CE and has remained fixed
even before the advent of Islam.
Just because some
language is older, it does not automatically lead us to the conclusion that the
orthography, lexicography and grammar of that language was "fixed"
unless there is firm evidence. This is true for Syriac as we shall see.
5.1 SYRIAC ORTHOGRAPHY
& GRAMMAR
The first systematic
treatment of Syriac orthography and grammar was done by Jacob of Edessa [d. 708
CE]. His work is known only from a brief mention by Bar Hebraeus and some
fragments discovered in Oxford and London.[86]
We are told in the
first fragment that Jacob had been asked by a correspondent to invent vowel
letters for Syriac. He replies that others had already considered this. A Greek
grammarian had, he states, declared that the Greek alphabet at one time
consisted of seventeen letters only; it had gradually been expanded to
twenty-four with the introduction of vowel letters. But the desire to do the
same for Syriac had been restrained by the fear of "destroying all the
Scriptures that had been written up to that time in this imperfect script of the
Syriac language". Jacob, torn between a wish to accede to his friend's
request and the caution of his predecessors, resorted to a compromise. He
invented vowel letters to be written, like the letters of the alphabet, upon
the line. But they were to be applied only in this treatise to illustrate
morphological forms in Syriac; they are not intended to be a permanent feature
of the language. Jacob decided that the vowel letters are to be added only to
illustrate the sense and the arrangement of these morphological rules, so that
they will demonstrate the variation and the pronunciation of the sounds. They
are not added in order to perfect or improve the script. Jacob enlarges on the
inadequacy of the system of writing, declaring that Syriac can be read correctly
only by guessing or by learning the traditional enunciation or after laborious
study.[87] He says:
And, as I have already
said, one cannot read anything accurately without those three things which have
already been mentioned above - either, I repeat, by guess-work because of
appropriateness (to the passage) and the sense required by the reading of the
context in which it occurs; or from tradition handed down by others who were
well-acquainted in the past with that context and its variant readings and could
pronounce its sounds accurately and have handed down (this ability) to others -
not on account of accuracy in reading the letters, for the letters have no
quality of accuracy, but because they themselves have received the tradition
from others; or by dint of great labour as one passes swiftly and, as it were,
flies in reciting these passages, with the various marks of the points
assisting and indicating the various meanings, so that those who receive the
traditional (method of reading) do not understand the passage from the letters
but from the enunciation of the sounds by the lips of the person transmitting
the tradition.[88]
In his well-known
letter to George of Sarug, Jacob rebukes the copyists of his time for their
ignorance and carelessness:
I prohibit all those,
who may copy the books, which I have translated or composed, from changing of
their own accord anything of those which they have undertaken to copy, either
in writings, or in the points, whatsoever they may find; if even a manifest
error be found, for everyman is liable to error, ourselves, the scribe, who
hath received the book from us, those who compare copy with copy, the eye of
the reader which errs and deviates from correct vision....
With respect to the
position of the points also, every man takes authority to himself to place them
as he pleases... which is nothing but deformity, odiousness and ugliness that a
natural and living body should be deprived of those members which have been
given to it by nature.... Again there is something absurd and ugly when the
face or the head of a man is found to have three ears or three eyes or any
other redundant member.... But this is the beauty of nature that there be in it
neither superfluity nor defect. Every one of the member should be made fit for
the place, which has been prepared and rendered convenient for it by nature....
Similarly it may be
seen with respect to the placing of the points, which are distinguishing and
explanatory of the various things which are placed in this Mesopotamian, or
Edessene, or, to speak more distinctly, Syriac Book; not in abundance or
superfluity, nor where a member has no need to be distinguished from another
which is similar to it in the letters, is it right that points should be
placed; nor that they should be thick, and like to hands and feet in each of
which there are six digits. Neither should they be deficient or fewer than the
portion by which a member may, if possible, be distinguished from the others,
which are like it; because that as superfluity, it has been observed, is not
becoming, so also deficiency is not becoming. It is right that they be also put
in places, which are convenient for them, and not where there is a vacant
place, whether it be suitable or unsuitable.[89]
Indeed, he remarks,[90]
all the West (Syrians)
and others living in other districts do not even speak this Edessene language
correctly, since they do not understand the principle
which governs the
system of writing.
Even some of the
Edessenes, those, that is, who speak this Mesopotamian language (most)
correctly, cannot read correctly not only those foreign sounds from outside
their own language which they use but even these native sounds.
They certainly, Jacob
continues, could not pronounce or read words adopted into Syriac from Hebrew,
Greek, Byzantine or Persian. This is due to the lack of letters to represent
all the sounds used.
As far as the Syriac
orthography after the advent of Islam is concerned, Jacob of Edessa's
correspondences give ample evidence of its poor state, not to mention the
incorrect speaking and understanding of the language. The Syriac script during
the time of Jacob and after the advent of Islam was clearly scriptio defectiva
as opposed to the assumptions of Mingana and Luxenberg that it was scriptio plena.
Now if the Aramaeans were not even able to read and write their own language
correctly even after the advent of Islam, how can we know what they allegedly
wrote that finally became the "Qur'an", was indeed what they actually
meant?
As far as the earliest
stages of Syriac grammar is concerned, our information about it is scarce. Most
of what we know about the terminology of the early Syriac grammarians (e.g.,
Jacob of Edessa) is derived from the reports in the later writers, such as Bar
Hebraeus [d. 1286 CE], and it is likely that their terminology reflects Arabic
influence rather than a genuinely indigenous development. Versteegh says:
Unfortunately, we do
not know very much about the earliest attempts of the Syrians in grammar and
reading, in the sixth and seventh centuries. Most of what we know is derived
from later Syriac writings, which were, however, composed at a time when the Syriac
tradition had undergone the influence of Arabic grammar, and had taken over
most of the conceptual and terminological apparatus of that tradition. As a
matter of fact, both Syriac and Hebrew grammatical terminology, as we know it,
are a calque of the Arabic terminology.[91]
Unfortunately for
Mingana and Luxenberg, their assumptions of Syriac orthographical and
grammatical influences on Arabic fall flat on their faces.
Let us now turn our
attention to Syriac lexicography.
5.2 SYRIAC LEXICOGRAPHY
Hunain b. Ishaq [d. 873
CE] composed the first proper Syriac dictionary based on the alphabetical
order. He laid the foundations of the Syriac lexicography.[92] At the end of
the 9th century, Zacharias of Merv supplemented it by adding further entries,
but it was a confused work. Isho‘ Bar ‘Ali, Hunain's pupil, wrote a new
dictionary based on it and is available in print.[93] In it, Syriac words are
followed by Arabic equivalents or definitions. Occasionally, further Syriac
explanations are added. The fullest and most famous Syriac dictionary was that
of Bar Bahlul (10th century).[94] It can be described as a sort of encyclopedia
and is a Syriac-Arabic lexicon with both Syriac and Arabic used for the
explanation of a word. As in much Arabic lexicography, authorities are
mentioned for words included.
As for Mingana and
Luxenberg, their hypothesis is based on an assumption that the Aramaeans were
well advanced in their lexicography while the Arabs had a difficult time in
understanding their own language, especially the Qur'an. We know that this is
not true. The first Syriac dictionary appeared well after the advent of the
first Arabic dictionary of al-Khalil b. Ahmed (i.e., Kitab al-‘Ayn) among
others (Fig. 5).[95]
Figure 5: A
chronological chart of Arabic lexicography according to the dictionary
arrangement used.
By the time Hunain
composed his first dictionary, there was already intense lexicographical
activity among the Arabs. What is even more interesting is that the Syriac
dictionaries of ‘Isho Bar ‘Ali and Bar Bahlul uses Arabic to explain
equivalents and definitions as shown in Fig. 6.
(a)
(b)
Figure 6: (a) The
lexicon of Isho‘ Bar Ali showing the meaning of Syriac words in Arabic and (b)
the lexicon of Bar Bahlul showing the means of Syriac words in both Arabic and
Syriac.[96]
Taking into
consideration that with the rise of Islam the use of Syriac eclipsed in the
Middle East, it is surprising to see Aramaeans using Arabic to explain the
meanings of the words in their lexicon. If Luxenberg's tacit assumption of the
superiority of Syriac lexicography was indeed correct then we would expect the
Aramaeans to use Syriac to explain their words. On the contrary, we find that
both Isho‘ Bar ‘Ali and Bar Bahlul make use of Arabic extensively to explain
the words in Syriac. All this suggests that by the time Hunain b. Ishaq
arrived, Arabic lexicography was already well-developed and far more
sophisticated than the Syriac lexicography. As quipped by Renan, the chief
literary characteristic of Syriac is mediocrity, and the chief claim to fame of
the Syrians is that they passed on the lore of the Greeks to the Arabs.[97]
Haywood points out that:
Thus we see in Syriac
lexicography signs of indebtedness to the Arabs, even though there is little
evidence of direct imitation. As in Hebrew, grammar and grammatical terminology
show similar influence.[98]
It is surprising that
Luxenberg uses the lexicons of Payne Smith and Brockelmann, which are primarily
based on the lexicons of ‘Isho Bar ‘Ali and Bar Bahlul,[99] the latter two
compiled more than 250 years after the advent of Islam and about 100 years
after the production of first Arabic lexicon, to show the alleged Syro-Aramaic
reading of the Qur'an.
7. Conclusions
In the last thirty
years or so, many revisionistic theories have been proposed as to how the
Qur'an/Islam came about. According to these various revisionistic schools of
thought, Islam was originally a Jewish sect (pace Hagarism); the Qur'an was
contemporaneous with the sira (pace Wansbrough); Islam arose in the Negev
desert somehow allegedly validating Wansbrough's hypothesis (pace Nevo); the
Qur'an came after the sira and hadith (pace Rubin); the Qur'an was an Iraqi
product and predates the sira (pace Hawting) and, recently, the Qur'an is a
product of Syriac Christianity (pace Luxenberg). It seems that these
revisionistic schools often follow methodologies that do not agree with each
other (whether in whole or in part) and none of them seem to agree on any one
particular scenario, be it historical, social, cultural, political, economic or
religious. Something that appears to be more fundamental in their analysis is
that the revisionists are willing to formulate any theory to lend
verisimilitude to their opinions concerning the Qur'an/Islam, no matter how
much it contradicts all of the available well-established evidence, documentary
or otherwise. In this regard we discover that Luxenberg is no different. This
is precisely what has been point out by Robert Hoyland recently. While
discussing the corpus of documentary evidence for early Islam and its almost
total absence in the work of Christoph Luxenberg among others, he says:
Firstly, we do have a
number of bodies of evidence - especially non-Muslim sources, papyri,
inscriptions and archaeological excavations - that can serve as a useful external
referent and whose riches are only just beginning to be exploited in a
systematic manner. Secondly, the historical memory of the Muslim community is
more robust than some have claimed. For example, many of the deities, kings and
tribes of the pre-Islamic Arabs that are depicted by ninth-century Muslim
historians also feature in the epigraphic record, as do many of the rulers and
governors of the early Islamic state. This makes it difficult to see how
historical scenarios that require for their acceptance a total discontinuity in
the historical memory of the Muslim community - such as that Muhammad did not
exist, the Qur'an was not written in Arabic, Mecca was originally in a
different place etc. - can really be justified. Many of these scenarios rely on
absence of evidence, but it seems a shame to make such a recourse when there
are so many very vocal forms of material evidence still waiting to be
studied.[100]
Similar conclusions
were also reached by Federico Corriente, who said:
Since neither the sources
of the history of Early Islam nor trustworthy reports on the socio-linguistic
situations of Arabia in those days appear to support Luxenberg's claims,
cleverly interwoven as the warp of his hypothesis of the usefulness of an
alternative Syro-Aramaic reading of the dark passages of the Qur'an, one
wonders about the correctness of the particular solutions offered by him in
each case, only to find that a few of them may be accepted as improvements to
the traditional interpretations, while some are allowable but unnecessary as
sheer interpretative alternatives, and some might be outright rejected because
of misapprehensions and misinformation.[101]
In this article, we
have examined the claims of Mingana and Luxenberg concerning the origins of
various aspects of the Arabic script. It was shown that the Arabic script
originated from the Nabataean script as opposed to Luxenberg's syro-aramäische
script. The diacritical marks in Arabic were already known before the advent of
Islam and it is most likely that they came from the Nabataean script. As for
the vowel marks in Arabic, Abu al-Aswad al-Du'ali was the first one to
introduce them. It is highly unlikely that they could have originated from
Syriac because the orthographical needs of Syriac were to a great extent
different from that of Arabic. Furthermore, the terminology for vocalization in
Syriac script is known only from later authors such as Bar Hebraeus and shows
the influence of Arabic orthography. Consequently, the seven ahruf of the
Qur'an representing the seven vowel signs of Syriac is rather far-fetched. The
analysis of isnad bundles show that the tradition of the revelation of the
Qur'an in seven ahruf was already known in the first century of hijra and they
can in no way represent the Syriac vocalization of Jacob of Edessa.
The Qur'an was
palaeographically a frozen document in the first century of hijra as seen in
the manuscripts as well as the qira'at in which they were written; MS. Arabe
328a and MS. Or. 2165 being two good examples. Furthermore, examining the
geographical characteristics of the inscriptional evidence, shows that the bulk
of the pre-Islamic Syriac inscriptions are confined to the Edessa region in
modern south Turkey, the language which Luxenberg says the Qur'an was partially
written in. It is certainly a long way from the hijaz region and in particular
Makkah. The closest to the pre-Islamic Arabic inscriptions in terms of
geography are the Nabataean inscriptions. In fact, much of the socio-historic
context of the origins of Islam can be explained by considering the Nabataean
milieu. The Nabataean milieu successfully addresses the origins of Arabic
script, dotting and pre-Islamic existence classical Arabic. The proximity of
the Arabic script with the Nabataean script makes Luxenberg's hypothesis of
syro-aramäische reading of the Qur'an unfeasible. A Syro-Aramaic milieu would
find it impossible to account the aspects of the origins of Arabic script, its
dotting, origins of classical Arabic and the socio-historic context relating to
the rise of Islam.
There exist other valid
explanations for the presence of Aramaic words in the Qur'an. The Nabataeans
were Arabs who wrote using the Nabataean Aramaic script. The evidence from the
Raqush inscription [267 CE], the earliest dated Arabic text, with Aramaic
archaisms shows that the Arabs were familiar with Aramaic. Moreover, a stela
from Tayma in northern Saudi Arabia inscribed in Aramaic suggesting that
Aramaic had been very important in the hijaz as early as the fifth/fourth
centuries BC. Thus, it would be surprising if there were not quite a number of
Aramaic words in the hijazi dialect of Arabic. However, this does not make
Arabic a mixed language.
To support his
hypothesis for the alleged Syro-Aramaic reading of the Qur'an, Luxenberg laments
the late origins of Islamic literature. He says:
According to Islamic
tradition, the Koran dates back to the 7th century, while the first examples of
Arabic literature in the full sense of the phrase are found only two centuries
later, at the time of the ´Biography of the Prophet´; that is, of the life of
Mohammed as written by Ibn Hisham, who died in 828. We may thus establish that
post-Koranic Arabic literature developed by degrees, in the period following
the work of al-Khalil bin Ahmad, who died in 786, the founder of Arabic
lexicography (kitab al-ayn), and of Sibawwayh, who died in 796, to whom the
grammar of classical Arabic is due. Now, if we assume that the composition of
the Koran was brought to an end in the year of the Prophet Mohammed´s death, in
632, we find before us an interval of 150 years, during which there is no trace
of Arabic literature worthy of note.
It does not occur to
him that the Syriac sources he uses to prove the alleged Syro-Aramaic reading
of Qur'an are even later than the Islamic sources. As we have seen, he is
unfamiliar with the origins of both Arabic and Syriac source material. To give
some more examples, the earliest Arabic literature that comes to us is in the
form of hadith collections. An example is the Sahifah of Hammam bin Munabbih,
[d. 110 AH /719 CE], a Yemenite follower and a disciple of the companion Abu
Hurrayrah, [d. 58 AH / 677 CE], from whom Hammam wrote this Sahifah, which
comprises 138 hadith and is believed to have been written around the mid-first
AH/seventh century. This is available as a printed edition.[102] The hadith
collections of Ibn Jurayj [d. 150 AH] and Ma‘mar b. Rashad [d. 153 AH], many of
them transmitted by ‘Abd al-Razzaq in his Musannaf, are also available in
print.[103] Motzki has traced the material in the Musannaf of ‘Abd al-Razzaq to
the first century of hijra.[104]
Apart from Luxenberg's
lack of understanding regarding the development of Syriac and Arabic
orthographies, grammars and lexicographies, his work makes no attempt to anchor
his arguments in any believable historical context, as we have already seen
earlier. It is not clear who these Christians of pre-Islamic Makkah were who
used the alleged Qur'anic aramäische-arabische Mischsprache and how these
writings produced the Arabic Qur'an. What kind of time scales were involved in
the transformation? What were their religious beliefs and what made them change
their(!) religion into Islam?[105]
To conclude with a
quote from Walid Saleh, essentially, Luxenberg is arguing that Islam is the
result of a philological comedy (or tragedy) of errors. One is reminded here of
Emperor Julian's quip against the Christians and the Christians' apt response; paraphrasing
it, one could say that the Muslims read their scripture and misunderstood it;
had they understood it, they would be Christians.
And Allah knows best!
Acknowledgements
One of the authors
(MSMS) would like to thank Dr. R. Hoyland for fruitful discussions on the
Syriac and Greek inscriptions from pre-Islamic times. Dr. Hoyland is not
associated with Islamic Awareness.
Bookmark and Share
References & Notes
[1] A. Mingana,
"Syriac Influences On The Style Of The Kur'an", Bulletin Of The John
Rylands Library Manchester, 1927, Volume II, pp. 77–98; Also see A. Mingana,
"An Ancient Syriac Translation Of The Kur'an Exhibiting New Verses And
Variants", Bulletin Of The John Rylands Library Manchester, 1925, Volume
IX, pp. 188-235.
[2] A. Jeffery, The Foreign
Vocabulary Of The Qur'an, 1938, Oriental Institute: Baroda (India).
[3] C. Luxenberg, Die
syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der
Koransprache, 2000, Das Arabische Book: Berlin.
[4] A. Mingana,
"The Transmission Of The Qur'an", Journal of The Manchester Egyptian
and Oriental Society, 1916, p. 45.
[5] Rev. A. Mingana
& A. S. Lewis (eds.), Leaves From Three Ancient Qur'âns Possibly
Pre-‘Othmânic With A List Of Their Variants, 1914, Cambridge: At The University
Press, p. xxxi.
[6] ibid., p. xxx.
[7] C. Luxenberg, Die
syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der
Koransprache, op. cit., p. 299. In his own words:
Gegenüber der
bisherigen Annahme eines in Mekka gesprochenen arabischen Dialekts hat die
vorliegende Studie ergeben, daß es sich hierbei eher um eine
aramäisch-arabische Mischsprache gehandelt haben muß, sofern die arabische
Tradition die Koransprache mit der der Qurays, der Bewohner von Mekka
identifiziert.... Dies würde die Annahme nahelegen, daß Mekka ursprünglich eine
aramäische Ansiedlung war.
No comments:
Post a Comment