Let me preface my remarks with an admonition against misinterpretation.
I have very little knowledge of Islam or Quran, and there is no way
I will comment on either. I am writing as a professional mathematician,
with interest in a mathematical question. Check me out onhttp://cs.anu.edu.au/~bdm/
" if you care.
My instant reaction when my attention was drawn to this thread
was "I've seen it before". Consider this (paraphrased):
"The number of words is 7*25, and the total numerical value is
7*14809. The vocabulary has 7*7*2 words, 7*79 letters, 7*7*6 vowels
and 7*37 consonants. These occur in 7*19 forms, which list has
numerical value 7*12809; 7*16 forms occurring just once."
It continues on like this to a total of more than 150 "features"
involving the number 7. This refers to a single paragraph of a
Greek text of the Gospel of Mark (Mark 16:9-20). I took it from
a book "The last twelve verses of Mark, their genuineness
established", by Ivan Panin. Panin was a Canadian of Russian birth
who devoted most of his life to discovering such features. I have
a fair library of books by him or following him. They tend to be
distributed by Christian organisations calling themselves "Revivalist",
"Pentecostal" or similar, but are rarely mentioned by mainstream
Christian churches. For reasons I could never understand, another
major distributor is the British Israelites---a group who believe
that the Brits are one of the "lost tribes of Israel".
About 20 years ago I spent a lot of time studying this stuff.
I even started to write a book about it, but never finished. My
conclusion was clear, and I still hold it in very firm confidence:
This material has NO mathematical/statistical basis.
It is not so easy to convince people of this conclusion, because
when you read Panin's work you can't help but be struck by how
amazing it sounds. It really is very impressive. It is also true!
Panin did not tell lies about his arithmetic; accidental errors
aside, all his "features" are really there. I checked many.
So why did I dismiss it mathematically? Allow me an analogy.
[Don't you just hate analogies? tough :-)] My local phone book
has about 100,000 entries. Let's go through and find the numerical
value of every one. We find that about 1/7th, say 14,000, are
multiples of 7. Now we look at those 14,000 and select names with
Polish origin. Thanks to many immigrants from Poland to Australia
in the past, we find quite a few, say 1000 (a guess). Now I write
a little pamphlet "Amazing Discovery in Canberra! Was Polish
Immigration Divinely Ordained?" and in there I list 1000 names of
Canberra residents of Polish origin and the numerical value of each.
Incredibly, every single one is a multiple of 7!! I throw in some
Biblical (or Nostradamus) quotes of dubious relevance, and there
you have it.
You can see the nature of my lie-free deception: people who read my
pamphlet don't know what I left out.
Possibly you don't accept the validity of the analogy yet, so let me
explain that with a genuine example. One of Panin's most famous
"discoveries" was a pattern of 7 in the first sentence of Genesis,
Hebrew text of course. From that one short sentence, he listed
20 or 30 things involving the number 7. To analyse whether it was
statistically significant, I made a big table of the things that
Panin was willing to use as a "feature" in his writings, and I closed
it under orthogonality with respect to substructures. [Haha, I wrote
that on purpose to show off. What it means is:] For example, if in one
place Panin counted "total numerical value of the nouns" and in another
place "total number of letters in the first and last words", I would
add "total numerical value of the first and last words" and "total
number of letters in the nouns". However I put a tight limit on the
complexity, for example the "total numerical value of the first letters
of each noun appearing in an odd position in the sentence" was OUT.
The idea was to just include the sort of "feature" that Panin himself
would be happy to present.
Having made my big table (I forget exactly, something like 300 entries)
I calculated them all and determined what proportion of them were
multiples of 7. The result was only slightly more than 1/7 --- so
slight that about 15% of random sentences would score as much. If you
consider that Panin had the freedom to choose which sentence to present,
and which number (7 or another) to use, you can see that there wasn't
much to be impressed about after all.
A problem faced by anyone without the time for doing calculations is
that it is not intuitively clear that the number of potential "features"
is so huge. I mentioned 300 for a single sentence, but for something
larger the list could run into many thousands. What happens is that
a "feature" has many possible items "letter, word, sentence, value,
count, vowel, consonant, first, last, noun, verb, unique, etc etc"
that can be used in a huge number of combinations. Some of them are
going to be multiples of 7 (or 19 or whatever) just by chance. That
still leaves a lot, so with good chance you can find a smaller set of
these "good" features that seem to be connected to each other. Then
you present just those features alone, and it appears amazing.
Panin had the additional tool of picking and choosing amongst variants
of the text where there was choice. He had a wonderful excuse: the
combination of variants showing "the" pattern must be the valid one!
He published a Numeric Greek New Testament "exact to every single letter"
by this method. I have a copy.
I have not read about the similar "features" claimed for the Quran
except in this group, but nothing I have seen suggests it is different.
I'd be very interested in reading it though; is there a source in English?
The bottom line is this: find other reasons for believing in these
patterns if you wish, but be warned that no mathematician or
statistician will support you on technical grounds.
(Australian National University; not connected to RIT.)