1. Until now I have spoken in general about the root categories, the verb patterns, and their actions:

And now I shall speak of them in detail. First of all, you must know that every verb in the Hebrew language — whether it is complete or not complete — has three root letters. The grammarians have already agreed to measure every root according to the pattern pa‘al, and they called the first letter of every root pe, the second ‘ayin, and the third lamed.

And any root in which one of its three root letters is missing or weak is called “defective in pe,” or “weak in pe,” or in ‘ayin, or in lamed.

And in the Second Treatise, in the Sixth Principle, section 6, I shall explain the difference between the defective roots and the weak roots:

2. Therefore, remember this: wherever you find written “the pe of the verb,” or “the ‘ayin of the verb,” or “the lamed of the verb,” whether in verbs or in nouns, it means the first letter, the second letter, or the third letter respectively. I shall speak further about this in the discussion of noun patterns at the beginning of the Third Treatise.

3. And know that every verb called “complete” is one in which none of the letters aleph, he, vav, or yod occurs, and which also has no nun at the beginning, and whose last two letters are not identical; for in such a case none of the three root letters will ever be missing. For this reason it is called “complete.”

And a verb that has a nun at the beginning is called from the category of the defective roots, as will be explained in its category; and a verb that has one of the letters aleph, he, vav, or yod is called from the category of the weak roots. And I shall now explain the manner of their weakness.

4. The aleph may be weak at the beginning or at the end of verbs. For example, at the beginning, from akhal one says okhal, yokhal, tokhal, and at the end, as in qara, qarata, and so on. However, it does not appear in the middle of a verb, as in sha’ag, sha’al, and other similar forms.

The he becomes weak only at the end of a verb, such as galah, asah, and the like. But at the beginning or in the middle it does not become weak, as in harag, bahal, and the like.

The vav becomes weak only in the middle, such as kum, shuv, and the like; but at the beginning or at the end a root vav is not found.

The yod becomes weak only at the beginning, such as from yada, one says eda, yeda; eshev, yeshev.

The tsere vowel indicates the presence of a yod; and according to the opinion of some, a weak yod does not occur in the middle of verbs, and at the end it never occurs as a root letter according to everyone.

5. And a verb that has two identical root letters at the end is called from the category of doubled roots, such as sabab and tamam.

And only a few are found in which the first two letters are identical, and these follow the pattern of the complete verbs. For example, from dedeh one says eddaddeh (Isaiah 38:15); and from shashah, “veshishitem et ha-eifah” (Ezekiel 45:13). There are almost no others among verbs, but they are found more frequently among nouns, such as “darakh kokhav mi-Ya‘akov” (Numbers 24:17), “u-mashoaḥ ba-basar” (Jeremiah 22:14), “be-vavat ‘eino” (Zechariah 2:12), and the like — though they are few.

6. And verbs in which the first and third root letters are identical are not found at all among the complete verbs. The word ye‘al‘u dam (Job 39:30) does not have the root ‘ala‘, as I shall explain in the book HaHarkavah. But among roots that are not complete, such cases are found, such as hayah and natan, and a few others. They are found more frequently among nouns, especially with the letter shin, such as shemesh, shoresh, shalosh.