and pronounce the additional letters themselves. Thus they say that the pattern of istakhraja is “istaf‘ala.” Prosodists use combinations of these letters to represent metrical patterns, but consider only the metre, without distinguishing root letters from additional letters. Thus, when one of them was asked about the pattern of naktal, he answered, “naf‘al,” and the morphologists laughed at him. According to them, its pattern is naftal, since its original form is naktālu, whose underlying form is naktayilu. The yā carried a vowel and was preceded by a fathah, so it became an alif; then, when the verb became jussive, the alif was dropped. We give a similar explanation for na‘tadi and naftadi, from i‘tāda, ya‘tādu (“to be accustomed”), and iqtāda al-ba‘īra, yaqtāduhu (“to lead a camel”). Something similar applies to naqtayilu. When they omitted the alif, which is called the lām of the word, its pattern became … (1)

And they were made “eight.” It may carry a vowel, in which case it is hamzah, or carry no vowel. It counts as two letters under the first convention and one under the second. Alif is grouped with wāw and yā because they are the weak letters. This is why it was placed at the end of the alphabet. They pronounced the initial sound of each letter’s name, except alif: since they could not pronounce it at the beginning of an utterance, they placed lām before it and said “lā.” The one at the beginning is the vowelled hamzah, since hamzah occurs at the beginning of its name. Some people pronounce it “lām alif,” but the correct pronunciation is “lā.” A fuller discussion of this belongs elsewhere.

The point here is that knowledge requires authenticated transmission and soundly established reasoning. Weak reports, especially fabricated ones, cannot be relied upon. Likewise, unsound theories and false rational arguments rooted in ignorance cannot serve as evidence.