Sixty-six chapters spanning at least two centuries of history, and the question of whether they had one author is the single most consequential dating argument in the Hebrew Bible.
What the book is
Isaiah opens in the eighth century BCE, in Jerusalem, under the kings of Judah named in its first verse, with Assyria as the threat. It closes in a world where Babylon has already fallen, the exiles are returning, and the practical problems are rebuilding and community discipline. Somewhere between those two situations the book changes address.
How it is put together
Since the late eighteenth century, scholarship has generally divided the book into three parts. Chapters 1 to 39 are set in the Assyrian crisis of the eighth century and contain most of the material attributed to the prophet Isaiah son of Amoz. Chapters 40 to 55 address an audience already in Babylonian exile, with Cyrus of Persia on the horizon. Chapters 56 to 66 speak to a community back in the land, dealing with a rebuilt but disappointing present.
Chapters 36 to 39 are narrative rather than oracle, and they run closely parallel to 2 Kings 18 to 20. Which direction the borrowing runs is itself a question, and it is a useful one to sit with: two biblical books preserving nearly the same text is a documentary fact in its own right.
The dating question
At 44:28 and 45:1 the text names Cyrus, calls him shepherd and anointed, and says he will order Jerusalem rebuilt and the temple founded. Cyrus took Babylon in 539 BCE, roughly a century and a half after the eighth-century setting of the book’s opening.
If chapters 40 to 55 were written by the eighth-century prophet, that is a named prediction across a hundred and fifty years, and it is one of the strongest claims of predictive prophecy anywhere in the Bible. If those chapters were written during the exile by an anonymous author or authors, the naming of Cyrus is contemporary reference rather than prediction, and the burden shifts entirely.
The critical case does not rest on the Cyrus verses alone. It rests on the change of implied audience — chapters 40 onward address people who are already in exile as their present condition — together with differences in vocabulary, style and theological emphasis, and the absence of the eighth-century Assyrian setting that dominates the earlier chapters. The traditional case argues that a single prophet could be given a vision of a later situation, that the stylistic differences are within one writer’s range, and that the ancient tradition knows only one Isaiah.
One piece of evidence is regularly cited on both sides and deserves care. The Great Isaiah Scroll from Qumran, copied around the second century BCE, contains the whole book continuously, with no break or marker at chapter 40. That is decisive proof that the book circulated as a single scroll by that date. It is not evidence about authorship, because a scroll from the second century BCE cannot tell you who wrote what four hundred years earlier — it tells you the book had already been assembled. Both of those statements are true, and conflating them is the most common error made about this manuscript.
Where the external record touches it
Isaiah 36 to 37 describes Sennacherib’s campaign against Judah and the siege of Jerusalem under Hezekiah. Sennacherib left his own account, inscribed on clay prisms, one of which is in the British Museum. It records the campaign, names Hezekiah, describes shutting him up in Jerusalem “like a bird in a cage”, and lists the tribute taken. It does not record taking the city.
That is a rare and valuable situation: two accounts of one campaign, written by the two sides, both surviving, disagreeing about what happened and agreeing on enough that the disagreement is legible. It is the subject of a dossier on this site.
The later chapters intersect the Persian record. Cyrus’s own inscription on a clay cylinder describes his entry into Babylon and his policy toward displaced peoples and their sanctuaries, and it is routinely set beside Isaiah 44 to 45 and Ezra 1. What the cylinder actually says, and what it is often reported to say, are not the same thing, and separating them is one of the more useful things a reader can learn to do.
Where to start
Read chapter 39, then chapter 40, one after the other, and notice what changes between them: the addressee, the assumed situation, the tone. Whatever conclusion you come to about authorship, that transition is the thing the whole argument is about, and it is visible on a first reading.
Then read how we date a text, which sets out the reasoning above in general form.