Why Bible chronologies disagree

Open three books of Bible chronology and you will get three answers for the same event. One will place the Creation in 4004 BC. Another will say 5500 BC. The Jewish civil calendar counts from 3761 BC. A fourth writer will decline to give a year at all.

None of these people is being careless. They are working from different manuscripts, different assumptions about how genealogies work, and different views about what the numbers in Genesis were ever meant to do. The gap between them is not a rounding error. Between the highest and lowest figures for the Creation there is more than fifteen hundred years.

This matters for a timeline. Every entry before about 900 BC carries a date that somebody had to reason their way to. Knowing how that reasoning works is the difference between reading a date and trusting it blindly.

Two questions that get confused

Almost every argument about Bible chronology is really two arguments wearing one coat.

The first is a question about text: what numbers does the Bible actually contain, and which manuscript should be believed where they differ? That is a question about copies, and it has evidence attached to it.

The second is a question about genre: were those numbers written down to be added up? That is a question about what kind of writing Genesis is, and reasonable people who agree on every manuscript reading still land in different places on it.

Keeping the two apart makes the whole subject much easier to follow.

Three Bibles, three sets of numbers

The genealogies of Genesis 5 and Genesis 11 are the spine of every early chronology. They are built to be added: each man is given an age at the birth of his named son, so the years run end to end from Adam to Abraham.

The problem is that three ancient versions of those chapters survive, and they do not carry the same ages.

The pattern is consistent enough to look deliberate. In the Septuagint, many of the patriarchs father their named son a full hundred years later than they do in the Hebrew. The Samaritan text shortens some of the same lives. The Septuagint also carries an extra generation, a second Cainan, between Arphaxad and Shelah — and Luke, writing in Greek, has that name in his genealogy of Christ where the Hebrew of Genesis does not.

So a chronologist has to choose a text before adding a single number. Ussher chose the Hebrew, which is why his figure is one of the shorter ones. Much of the early Christian world used the Septuagint, which is why the older Byzantine reckoning put the Creation near 5500 BC.

Where 4004 BC came from

James Ussher was Archbishop of Armagh, and his Annals of the Old Testament appeared in 1650. He worked from the Hebrew text, added the genealogies, tied the later kings to Persian and Babylonian records he could date, and arrived at a Creation in 4004 BC. He went further than that and named the evening before Sunday 23 October.

Ussher was a serious scholar, and by the standards of his century the work was careful. What made his figure famous was not the argument but the printing. From the early eighteenth century his dates were set in the margins of English Bibles alongside the text, and generations of readers grew up seeing 4004 BC on the same page as Genesis 1. It came to look like scripture. It never was.

The dates that do not move

Not everything is uncertain. From roughly the ninth century BC onwards, biblical events can be pinned to records kept by other nations, and those records can be dated on their own terms.

The anchor is an eclipse. An Assyrian list of year-names records a solar eclipse in the month of Simanu during the term of the official Bur-Sagale, and that eclipse can be identified astronomically as 15 June 763 BC. With that one fixed point the whole Assyrian year-list can be counted forwards and backwards, and every king mentioned in it lands on a real date.

From there the joins to scripture are direct:

• 853 BC. Shalmaneser III's account of the battle of Qarqar names Ahab of Israel among the kings who fought him.
• 841 BC. The Black Obelisk shows Jehu bringing tribute to the same Assyrian king.
• 722 BC. Samaria falls and the northern kingdom ends.
• 701 BC. Sennacherib besieges the cities of Judah and shuts Hezekiah up in Jerusalem, as both the Bible and the Assyrian annals describe.
• 16 March 597 BC. The Babylonian Chronicle dates the capture of Jerusalem to the day.
• 587 or 586 BC. The temple of Solomon is burned. The one-year spread here is a genuine open question about how a regnal year was counted, not a gap in the evidence.

This is why the later half of Old Testament history is dated far more confidently than the earlier half. It is not that the record improves. It is that other nations start writing things down that can be checked.

The 480 years that will not sit still

One verse decides where the Exodus goes. First Kings 6:1 says Solomon began the temple in the four hundred and eightieth year after Israel came out of Egypt. Solomon's fourth year can be placed near 967 BC, so the arithmetic gives an Exodus around 1446 BC.

Two things complicate it. The Septuagint of the same verse reads four hundred and forty, not four hundred and eighty. And a great deal of the archaeological argument points instead to the thirteenth century, closer to 1250 BC, in the reign of Ramesses II.

Both dates have long-standing, serious defenders, and the choice between them shifts the entire period of the judges by two centuries. A timeline has to pick one and say which. Pretending the question is settled would be the only real mistake.

What a genealogy is for

The deepest disagreement is not about manuscripts at all.

Hebrew genealogies do not always list every generation. The word rendered "son of" can mean grandson, or descendant. Matthew's genealogy of Christ is arranged in three neat groups of fourteen, and to make the third group fit he passes over three kings of Judah whom Chronicles names in order. Nobody accuses Matthew of error. He is doing something other than counting.

If the lists in Genesis work the same way — and many careful readers think they do — then adding the ages gives you the sum of the named generations, which is a smaller number than the elapsed time. On that reading, the genealogy is a chain of custody rather than a stopwatch: it shows that the covenant passed unbroken from Adam to Abraham, without claiming to record how many years that took.

Latter-day Saint readers have a further reason to hold these figures loosely. Modern revelation is remarkably quiet about the arithmetic. Doctrine and Covenants 77 speaks of the earth's seven thousand years of temporal existence, but it is answering a question about the book of Revelation, not publishing a calendar. Abraham 3 makes time itself relative to the observer. And the Church has never adopted an official age for the earth.

How this site handles it

Three rules, applied everywhere.

Say which reckoning is being used. Early entries follow the conventional dates used in Latter-day Saint study materials, so the timeline lines up with the manuals and lesson books most readers already have.

Mark the ones that are traditional. A date before Abraham is a scholarly convention, not a measurement. Entries in that range say so.

Show the disagreement rather than hide it. Where the published chronologies conflict, the entry names the alternatives instead of quietly choosing one.