The civil calendar counts straight from 1 BC to AD 1 with no year zero, because the monk who designed the count worked before zero reached Latin arithmetic.
Documented
The monk, the computus, and the absent cipher
Documented: Dionysius Exiguus devised the Anno Domini count for Easter tables, anchoring it to the Incarnation by his own estimate, and numbered years ordinally in Latin, primus, secundus, first, second, with no concept of a zeroth year. Bede's Ecclesiastical History (731) spread the count through Latin Europe; historians using it inherited the zeroless seam, writing 1 BC followed by 1 AD. Astronomers quietly repaired it centuries later: Cassini introduced a year 0 in 1740, and the ISO 8601 standard and astronomical numbering now use 0 for 1 BC and negatives before it.
Assessment: Fully documented. The seam is not a mistake anyone hid; it is a fossil of the arithmetic available in 525. Two calendars still run side by side, the historical one without zero, the astronomical one with, and every database spanning antiquity has to say which it means.
Stories
The millennium that came early
The folk version is the argument that erupts every hundred years: did the twentieth century end in 1999 or 2000? Pedants say 2000, because there was no year zero; the planet partied in 1999 anyway, and the purists were left holding a technically correct banner at an empty second celebration.
Assessment: Documented as recurring public quarrel, 1900 and 2000 both generated the same editorials. The pedants are right on the ordinal math; the crowd is right that round digits, not ordinals, are what humans celebrate. The zeroless calendar loses every time, gracefully.
Interesting claims
Dionysius missed the date too
The interesting claim compounds the gap: Dionysius also misplaced the anchor. The Gospel of Matthew ties the nativity to Herod the Great, who died in 4 BC, so on the count's own terms the reference year lands several years before year one. The AD system is thus off by both a missing zero and a misplaced start, and the world runs on it regardless, which makes it the most successful approximation in history.
Assessment: Documented in standard chronologies, the 4 BC terminus for Herod is secure from Josephus. The mismatch means the calendar counts from an event it cannot quite locate, using a zero it does not have. Neither flaw matters operationally, which is precisely the lesson: systems outlive their founding errors.
Sources and further reading
Dionysius Exiguus, the Easter tables and Argumenta (c. 525), the count's origin
Bede, The Ecclesiastical History of the English People (731), the count's spread
Cassini, Tables astronomiques (1740), the introduction of year 0
ISO 8601 and astronomical year numbering, the modern repair
Josephus, Antiquities of the Jews, the 4 BC terminus for Herod
Worth asking
A counting system that skips zero rules the world's dates. What other inherited defaults are you still filing under obvious?