Tafsir of Ash-Shams 91:4

"And the night when it covers it": meaning the sun, by shrouding its light. The attribution is as previously mentioned. It has also been said that it refers to the earth, and another view is that it refers to the world.

The imperfect tense (yaghshāhā) is used here instead of the perfect tense (ghashāhā), as was the case with the preceding verses (e.g., if it were said: "when it covered it"). Abu Hayyan stated that this was done to observe the rhyme scheme. He did not say ghashāhā (it covered it) because that would require the omission of one of the two objects due to the verb's transitivity to both of them; for it is said: "I covered you with such-and-such," as al-Raghib mentioned. Thus it is said.

Some eminent scholars said the imperfect tense was used to draw attention to the equality of all times in His—the Exalted—presence. Al-Khafaji said that it is better to say that "the night" here refers to the darkness that occurs after the absence of light, not the original non-existence or the primordial darkness, for this is more manifest in signifying His power. Since it is future-oriented relative to what preceded it, the expression had to be changed to indicate the intended meaning.

Al-Zamakhshari found the issue of the grammatical status of the accusative case (nasb) following "when" (idhā) difficult. He argued that if all the "waws" (the conjunctions) following the first one are taken as conjunctive, then one must contend with conjunction upon the two objects of two different agents—such as conjoining "the day," for instance, to "the sun," which is governed by the verb of oath—and conjoining the adverb, i.e., idhā in "when it reveals it," to its counterpart in "when it follows it," which is governed by the verb of oath. If, however, they are taken as oaths, it would necessitate the gathering of multiple oaths upon a single response, which al-Khalil and Sibawayh considered reprehensible.

He answered by choosing the first option and denying the necessity of that outcome. He stated that the oath-particle "waw" is such that the expression of the verb (of oath) is entirely dropped; thus, it has a status different from the "ba," where the verb is sometimes expressed and sometimes omitted. Therefore, the "waw" stands in place of the verb of oath, and its "ba" fills their role; the subsequent "waws" and conjunctions are deputies for this "waw," so they act as governors of the genitive and accusative. Thus, the conjunction is of the category of conjoining upon the two objects of a single agent. This is like saying, "I struck Zayd [with] Amr, and [I struck] Bakr [with] Khalid"; you place the case in the nominative via the "waw" and the accusative due to its standing in the place of "I struck," which is their governor. End quote.

You know that the first of the conjunctive "waws" here does not have anything for it to govern in the accusative. Perhaps he meant that it acts as such if there is an object in the accusative, or that it is a governor based on the meaning of "By the sun and its brightness," i.e., "By the sun and its light when it rises." Furthermore, no one has ever held that conjunctive letters are governors. Also, the problem is predicated on the prohibition of conjoining upon the objects of two different agents absolutely. Even if one were to permit it absolutely, or under the condition that the conjoined item be in the genitive—as a group of scholars held, such as in your saying, "In the house is Zayd, and in the room is Amr"—there would be no difficulty.

Moreover, his argument is predicated on accepting this reprehensibility and the impossibility of escaping the gathering (of oaths) by estimating a response for each of the oaths. Even if one did not accept this, or if one did and estimated a response for each, there would remain no difficulty. It is also predicated on the assumption that idhā is exclusively an adverb of time, which is denied, as it is possible that it has been divested of its adverbial nature; in which case, it becomes a substitute (badal) for what follows the "waw," as was said regarding the verse: "And after tomorrow—oh, my sorrow for tomorrow / When my companions depart and I am not departing." Here, idhā is a substitute for "tomorrow."

Even if one concedes that it is an adverb, it is permissible to estimate that each prefixed noun it attaches to has a "was" (a state of being) associated with it; one could estimate: "and the moon when it follows it," "and the clearing of the day when it reveals it," and "the covering of the night when it covers it." Or, it may be made to attach to an omitted element that acts as a circumstantial qualifier (hal) inferred from what follows it; i.e., "I swear by the moon, [existing] when it follows it, and by the night, [existing] when it reveals it," as some have claimed. But this also has points of contention.

Al-Zamakhshari is also countered by such verses as His saying: "And the night when it retreats" and "And the morning when it breathes," for the "waw" there is conjunctive, and the explicit verb of oath has already preceded, as stated by Sheikh Ibn al-Hajib. Furthermore, the verified position, as some researchers have stated, is that the adverb is not governed by the verb of oath, because the meaning would be corrupted—since temporal restriction is not intended, neither in the past nor the future. Rather, it is governed by an omitted noun such as "the greatness," for swearing by a thing is an act of magnifying it. It is as if he said: "I swear by the greatness of such-and-such time." The objection raised against this—that His swearing by a thing is a metaphor to show its greatness and demonstrate its honor, and thus it is permissible to restrict it considering part of the intended meaning (i.e., the demonstration)—does not hold. Moreover, if the oath is an act of magnifying, then estimating it is redundant; if one were to concede it, the metaphor would be either taba'iyyah (in the word) or tamthiliyyah (in the structure). In any case, there is nothing there that would serve as a grammatical attachment, and the estimation is [only] so that something might be attached to it to emphasize what is intended; thus, there is no linguistic [necessity].