so that it may provide reassuring precedent!’ He smiled.”

I say:

Rather, al-Farrā’ and others had already advanced this explanation. Al-Farrā’ holds the same standing among the Kufan grammarians as Sībawayh does among the Basrans. Ismāʿīl, however, relied on Basran grammar, and al-Mubarrid was especially close to him.

This explanation may be clarified as follows: the singular is dhā (‘this’). Had they treated it like other nouns, they would have formed the dual as dhawāni, not dhāni, just as they said ʿasawāni (‘two staffs’), rajawāni (‘two sides’), and similar forms of triliteral nouns. Hā is a particle used to draw attention. Where they had omitted the final root consonant, they said abawāni (‘two fathers’), so forming the dual restored the original form. In other instances they said … (1) and yadāni (‘two hands’).

As for dhā, they did not say dhawāni. Rather, they proceeded as they did with dhū and dhāt, meaning ‘possessor’: they said huwa dhū ʿilmin (‘he possesses knowledge’) and humā dhawā ʿilmin (‘they both possess knowledge’), just as Allah said:

“Both having spreading branches.”

For the demonstrative, they said dhāni (‘these two’, masculine) and tāni (‘these two’, feminine), just as Allah said:

“So these are two proofs from your Lord.”

Dhā meaning ‘possessor’ is a declinable noun: its inflection changes according to whether it is nominative, accusative, or genitive, yielding dhū, dhā, and dhī. Demonstrative forms, relative pronouns, and personal pronouns, however, are indeclinable;