[In the following excerpt originally published in 1594, Tasso discusses Orlando furioso in terms of the Aristotelian concept of epic unit.]

[The poet] must see to it that his fable (by fable I mean the form of the poem that can be defined as the weaving or composition of its events)—he must see to it, I say, that the fable he wishes to fashion is entire, or, as we may put it, whole, that it is of an appropriate magnitude, and that it is one. (p. 62)

The fable is to be whole or entire because it is to be perfect, and nothing can be perfect that is not entire. Perfection and integrity will be found in the fable if it possesses a beginning, a middle, and an end. The beginning is that which does not necessarily come after something else but has other things after it. The end is that which comes after all other things and has nothing after it. The middle is placed between the two, following some things and followed by others. But to depart a little from the brevity of definitions, I call a fable entire that contains in itself everything necessary to its intelligibility, sets forth the causes and origin of the deed it undertakes to treat, and leads by due means to an end that leaves nothing inadequately concluded or resolved. Thus Homer, we see, has done in the Odyssey: first with the journey of Telemachus to Nestor and Menelaus, and then with the tale Ulysses tells Alcinous, he perfectly clarifies the state of affairs and what happened after Ulysses left Troy; and Virgil does the same with Aeneas' tale to Dido. Although the poet snatches the hearer into the midst of things as if they were already known, none the less he then proceeds to inform him little by little of what happened earlier. But Orlando Innamorato and Orlando furioso are not entire, but faulty in reporting what they involve: the Furioso lacks a beginning, the Innamorato an ending. Still, no artistic defect but death was at fault in Boiardo, and in Ariosto not ignorance but his choosing to finish what his predecessor had started. It is quite unnecessay to prove that the Innamorato is imperfect. So too Orlando furioso is obviously not a whole; for whether we take Ruggiero's love or the war between Charlemagne and Agramante as its main action, it lacks a beginning, since it does not tell when or how Ruggiero fell in love with Bradamante, or when or how the Africans began the war against the French, except perhaps with a bare reference in one or two lines. And readers would often have to grope in the dark for these stories if they could not learn what is necessary from the Innamorato. But, as I say, we must not consider Orlando Innamorato and Furioso two distinct works, but a single poem begun by one poet and completed by the other along the same lines, though with a better interweaving and colour; considered so, the poem is a whole, lacking nothing for the intelligibility of its stories. (pp. 62-3)

The supporters of unity, making a shield of the authority of Aristotle and the majesty of ancient Greek and Latin poets, and not lacking weapons provided by reason, have against them the habit of the present era, the universal agreement of ladies, gentlemen, and courts, and apparently experience as well, the infallible test of truth. For Ariosto, leaving the tracks of ancient writers and the rules of Aristotle, has encompassed within his poem many diverse actions; and he is read again and again by all ages and both sexes, is known in all languages, liked by everyone, praised by all, his fame ever alive and renewed, the glorious talk of men's tongues. (p. 66)

Let us grant what cannot be denied, that delight is the end of poetry; so too I grant what experience demonstrates, that Orlando furioso delights our contemporaries more than [Trissino's] Italia Liberata or even the Iliad and the Odyssey. But I do deny—and this is basic and all-important to our thesis—that multiplicity of action is more apt to delight than unity; for the opposite can be proved on the authority of Aristotle with the argument he adduces in the Problems. Although Orlando furioso, which contains several fables, gives more delight than any other Tuscan poem or even the poems of Homer, this is not because of unity or multiplicity of fable, but for two reasons that detract nothing from our argument. One is that the Furioso treats of love, chivalry, adventure, enchantment, in short of inventions more charming and more adapted to our ears; the other that Ariosto excels many other poets in propriety of manners and decorum of character. Both reasons are accidental, unrelated to multiplicity or unity of fable, and are not so involved with the one as to be unsuitable to the other. We should not therefore conclude that multiplicity delights more than unity. (pp. 76-7)