I came by an interesting discussion here http://forum.wordreference.com/showthread.php?t=615365

3 interesting opinions from there

1>."The famous linguist Charles Ferguson wrote once (I forgot where and don't remember the words exactly): If I meet an Arab at a party and want to know from where he is, I don't ask "Where do you come from" but "In which region of the Arabic world is a dialect spoken that is nearest to Standard or Classical Arabic". " 

2>"Here are some examples of archaic characteristics found in Najdi, bedouin, and Asiri Arabic:

(1) The "true" passive voice is still employed, whereas in most other dialects it's been replaced by the pseudo-passive forms "infa3al" and "itfa3al".

(2) نون النسوة is still employed (this is also present in southern Iraq).

(3) The construction "ما أنا بفاعل". For example, Muhammad's words to Gabriel "ما أنا بقارئ" can still be heard in Najd, though the pronounciation would be something like "Manib Gari".

(4) The imperative form for "weak roots" (الأفعال الجوفاء) is still rendered the same way as in CA: قم شف رح جب در نم, which in most other Arabic dialects would be قوم شوف روح دور نوم, etc. Najdis would also say ارم امش ابد, which is the correct CA form. Other dialects would say ارمي امشي ابدا

(5) The diminutive form is still alive and constructed in the same way as in CA in order to color meaning. In most Arabic dialects, diminutive words are lexicalized and not normally generated.

(6) The "tanween" is still used to mark indefiniteness: جاني واحدن, معه سيارتن قديمة - though because there is no longer any declensional marking in Arabic, the tanween is always "kasr".

(7) The plural present form is still the same as in CA for 1st, 2nd, and 3rd person: نروح يروحون تروحون. In most other dialects, the "n" is omitted.

(8) There is still gender agreement between the adjective and the word that it modifies: قهوة عربية, whereas it's common in other dialects to say قهوة عربي

(9) There is still gender disagreement with numbers just as in CA: أربعة أشهر, ثلاث بنات

(10) Contractions such "kaman" and "mesh/mosh" have not been lexicalized in Najd, unlike elsewhere. This is one of the most important features that help Najdi Arabic retain much of the "spirit" of Classical Arabic. For example, in Najd, you don't hear expressions like "أنا كمان" (i.e. أنا كما أنّ, which makes no sense in CA), or "مش رايح" (i.e. ما هو شيء رايح, again makes no sense). Rather, you would hear "مانيب رايح" (i.e. ما أنا برايح)...."

(There are more points, go to the link to read the rest and the responses)

3>"But if you believe like I do, that there has ALWAYS been a separation between an “ideal” classical (which no one ever spoke on a daily basis) and a/the dialects, then the question makes no sense. There is no “closer than” among the dialects and the classical because they were never one….they were always separated and distinct….they were always a dual manifestation of the same language.

In fact, I take the extreme view (of course, I don’t think it’s extreme…..I think it’s as plain as the nose on the face) that the derivation occurred in the opposite direction….c’est a dire, the classical was derived as an ideal FROM the spoken dialects – fusHa, that is to say, is a derivative and invented language – the language of poets and high culture -
an ideal never to be mastered.

Another way to put it: On that very first day when “Arabic” was originally spoken as an acknowledged separate language from an “Ur” Semitic parent, what was spoken was not fusHa but was already a “dialect”…..or to put it another way, spoken came first, fusHa came later."

I would like to know your thoughts and opinions