The Crown Arrived Later

In 1130 Roger II was crowned at Palermo, and the Kingdom of Sicily is usually dated from that coronation. The Norman conquest of the island had been finished by 1091.

Those are not rival dates for one occurrence. One records a crown and the other records territorial control, and they sit thirty-nine years apart.

Counted from the coronation, the kingdom lasted a little under seven centuries. Anyone describing eight hundred years of Sicilian history is counting from the conquest, or from Norman holdings older than that. Both figures are arithmetically correct, and they are measuring different objects.

The received version compresses this into a founding year. The record keeps the event in view, and the discrepancy has been carried forward for centuries by people who were not making anything up. One source chose the kingdom. Another chose the territory. Each took its own arithmetic with it.

The crown arrived after the island #

Sicily makes the problem legible because its dates are unusually clean. A coronation is a named political act. A completed conquest is a different named political act. Neither can be corrected into the other, and no further research will merge them.

Then the same state breaks again. The Sicilian Vespers of 1282 split the kingdom proclaimed in 1130 between an island court at Palermo under the Crown of Aragon and a mainland court at Naples. Both used forms of the royal style of Sicily. Ferdinand’s Kingdom of the Two Sicilies reunited the administration in 1816, more than five centuries later.

The name survived the split. The state did not survive it in one piece.

A table giving Sicily one start, one end and one duration has therefore already decided, silently, whether the object being measured is a crown, an island, a title, a court, or a later administrative union.

England looks simpler only because its record-keeping settled early on a particular event. The tradition descending from the Anglo-Saxon Chronicle treats Æthelstan’s recognition at Eamont Bridge on 12 July 927 as the founding moment, and the Acts of Union 1707 supply the end. The arithmetic between those endpoints is sound. The bare figure is still incomplete, because the endpoints were chosen.

A statute can make the beginning conspicuous #

The Kingdom of Ireland is the opposite case, and the contrast is instructive because nothing about it is retrospective.

The Crown of Ireland Act passed the Parliament of Ireland on 18 June 1542 and declared Henry VIII King of Ireland. England had governed the island as a lordship under a twelfth-century papal grant; the statute replaced that basis with a domestic one after Henry’s excommunication. The end is equally legible: the Acts of Union 1800 merged Ireland into the United Kingdom effective on 1 January 1801.

No founding line was crowned in 1542 and no royal family was extinguished in 1801. The separate kingdom began and ended through instruments, and the instruments say so.

That is the case for weighting an incorporation date over a company’s older economic lineage. It is not a general rule. A formation instrument answers when the present legal person began. First revenue, a control change, brand adoption and predecessor operations answer other questions, and calling any one of them the company’s age without naming the event turns a convention into an apparent property of nature.

Duration inherits both boundary choices #

The Kingdom of Kongo is recorded as c. 1390 to 1914, which puts a tentative founding at one end and a legal abolition at the other. Its territorial power had effectively ended with defeat at Mbwila in 1665 and the civil war that followed. The 1914 date records the abolition of a titular monarchy two and a half centuries after that.

Neither ending is wrong. They describe different things, and the different things lasted different lengths of time: territorial power, a succession claim, a legal title.

A business unit produces the same picture when the operating activity stops, the brand continues, and the legal entity is dissolved years later. The duration changes because the object under measurement has changed, and nothing on the chart records the substitution.

An index has exactly this problem. First publication, mandate inception, a benchmark reset and a control change can all be genuine starts, and they do not produce the same return history. Recalculating each version will not reveal which age is secretly true. It reveals what each claim decided to count.

Legal ends are stronger evidence than economic ones #

The transfer stops at the record, and it is worth saying where.

Many kingdoms have no authoritative incorporation instrument at all. Kongo’s c. 1390 opening is an approximation before anyone starts arguing, while its 1914 abolition is a legal act with a document behind it. A modern issuer almost always has a formation instrument, even where its economic lineage runs earlier. The legal start is more inspectable, which is a genuine advantage, and it still settles nothing about commercial continuity.

That asymmetry decides which conflicts can be closed from the file. A date conflict caused by a data-entry error resolves to one documented event once the record is corrected. Sicily does not resolve. It yields a coronation in 1130 and a completed conquest in 1091, and both stay.

Ireland yields a statute in 1542 and a union in 1801. Kongo yields a loss of territory in 1665 and an abolition in 1914. The plurality is the evidence rather than noise in it.

The 1130 coronation is the date most often given for Sicily. The 1091 conquest is the date that changes its duration.

The first gives a kingdom a birthday. The second decides what the birthday has been made to mean.