History

The census of Quirinius

Luke’s account of a census under Quirinius at the time of Jesus’ birth conflicts with the historical record and Roman administrative practice.

The premise

Luke 2:1-5 makes three historical claims in five verses. Augustus ordered 'all the world' to be registered. This registration took place while Quirinius was governing Syria. And people went 'each to his own town', which for Joseph meant Bethlehem, the city of David. Luke also dates the conception of John the Baptist to the reign of Herod the Great (Luke 1:5), and Matthew places Jesus' birth in Herod's reign as well.

The problem is that most scholars date Herod's death to 4 BCE, and the only census of Judea connected with Quirinius is the one he conducted in 6 CE, after Rome deposed Herod's son Archelaus and annexed Judea. That is a gap of roughly a decade.

The methodological divide here is sharp. Apologists typically start from the conclusion that the text is historical and then assemble scenarios showing that it is not impossible. Historians instead ask which explanation best accounts for all the data. A scenario is possible if nothing rules it out. It becomes plausible only if some evidence points toward it. It is probable only if it explains the evidence better than the alternatives do.

The primary texts
“In those days a decree went out from Caesar Augustus that all the world should be registered. This was the first registration when Quirinius was governor of Syria. And all went to be registered, each to his own town. And Joseph also went up from Galilee, from the town of Nazareth, to Judea, to the city of David, which is called Bethlehem, because he was of the house and lineage of David, to be registered with Mary, his betrothed, who was with child.”Luke 2:1-5
“And when they had performed everything according to the Law of the Lord, they returned into Galilee, to their own town of Nazareth.”Luke 2:39
“In the days of Herod, king of Judea, there was a priest named Zechariah, of the division of Abijah.”Luke 1:5
“For before these days Theudas rose up, claiming to be somebody, and a number of men, about four hundred, joined him. He was killed, and all who followed him were dispersed and came to nothing. After him Judas the Galilean rose up in the days of the census and drew away some of the people after him. He also perished, and all who followed him were scattered.”Acts 5:36-37
The four historical problems

Luke's narrative conflicts with Roman administrative history on four distinct fronts.

  • The client kingdom problem. Herod was a client king. Within his kingdom, taxation and registration were his responsibility, not a Roman legate's. Josephus narrates Herod's reign in great detail and never mentions Rome imposing a census on him. The 6 CE census has an obvious rationale: Judea had just become a Roman province and needed a tax base for direct administration.
  • Quirinius's career. Before 6 CE, Quirinius was elsewhere. He was consul in 12 BCE, governed Crete and Cyrene, and campaigned against the Homonadenses in Anatolia (modern Turkey). Every proposed date for that war places him in Anatolia, not Syria. He became legate of Syria in 6 CE.
  • The purpose of censuses. Roman censuses recorded where people actually lived for tax purposes. Requiring travel to an ancestral town roughly a thousand years removed would defeat the census's purpose. The edict of Vibius Maximus (P. Lond. 904, 104 CE) orders people absent from their Egyptian districts (nomes) to return to their own ephestia (hearths)—where they live—not ancestral towns. Some modern translations and popular graphics mistranslate 'nomes' as 'homes' or add language about family residence that is absent in the Greek.
  • The Titulus Tiburtinus. Apologists cite this fragmentary inscription as evidence that Quirinius served as legate of Syria twice. But the inscription is anonymous, and its mention of military exploits involving a king does not match Quirinius's known campaign against the Homonadenses. Most classicists reject the identification with Quirinius, favoring Piso or Saturninus.
The knockout blow: Luke, Acts 5, and Josephus

Luke himself shows how he handles chronological material. In Acts 5, Gamaliel gives a speech set in the early 30s CE. He mentions Theudas and then, 'after him', Judas the Galilean 'in the days of the census'. Compare Josephus, Antiquities 20.97-102:

  1. Josephus tells of Theudas, a prophet who led a crowd to the Jordan and was beheaded under the procurator Fadus (mid-40s CE).
  2. A few lines later, he notes that the sons of Judas the Galilean, 'who stirred up the people when Quirinius assessed Judea', were crucified under Tiberius Alexander (which occurred decades after 6 CE).

Josephus presents this as a flashback: Theudas (mid-40s CE) is mentioned first, and then Josephus backtracks to Judas (6 CE). Luke reproduces Josephus's pairing of these two figures but puts them in the wrong chronological order, treating the flashback as a sequence. No other known ancient text pairs these two figures. Steve Mason and others have argued at length that Luke-Acts drew on Josephus here and compressed the narrative, mistaking his flashback for a timeline.

Acts 5:37 also says 'the census', not a later or second census. Luke therefore knows of one census associated with Judas—the census of Quirinius in 6 CE. If the same author was this careless with chronology in Acts, the most economical explanation for Luke 2:2 is the same: he knew of Quirinius's census and placed it in Herod's reign.

The apologetic defenses

Apologists have offered several ways to bring Luke's census into Herod's lifetime. Each is addressed below.

'Quirinius was only governing, not the legate'

The participle hēgemoneuontos literally means 'while he was governing'. Apologists argue it need not mean he held the supreme office of legate, so he could have had some lesser role earlier. This is a fair point about the Greek verb, which is generic.

The problem: The flexibility in the verb gains nothing without data. No evidence places Quirinius in Syria or Judea in any capacity before 6 CE. The argument makes a role possible but never reaches plausible.

'Prōtē means "before"'

The argument is that prōtē can be a comparative taking a genitive, so the verse would mean 'this registration took place before Quirinius governed Syria'.

The problem: Prōtē is feminine and agrees with the feminine apographē. To read it as 'before', it would have to attach to the genitive absolute Kyrēniou clause, separated from it by the verb egeneto. The Baylor Handbook on the Greek Text of Luke lists three possible readings, all taking prōtē with apographē, and judges 'this was the first census' the most likely. The 'before' reading is not among the likely options.

'An earlier census, started by Saturninus and finished by Quirinius'

One reconstruction has a census begun under Saturninus, continued under Varus until 4 BCE, and completed under Quirinius between 4 BCE and 1 CE.

The problem: Varus was legate when Herod died and put down the revolt that followed, so Quirinius cannot have held that office until after Herod's death at the earliest. No source records Quirinius as legate after Varus. Matthew's 'two years or under' also implies Jesus was born roughly two years before Herod's death (6 or 5 BCE), when Quirinius was campaigning in Anatolia.

'Josephus misdated the census'

John Rhoads (JETS, 2011) argued that Josephus linked the census to Archelaus's deposition when it was actually triggered by Herod's death.

The problem: The chronology of Syria's governors and Judea's rulers rests on a web of converging data. Rhoads's argument would require revising all of it, and the theory has been rejected by most scholars. Even if it were right, it would not make the census Herod's own.

'Joseph's father lived in Bethlehem'

Some apologists suggest Joseph was still part of his father's household in Bethlehem and went there for that reason.

The problem: No data supports this, and it contradicts the text. Luke 2:4 says Joseph went because he was of the house and lineage of David, which was about a thousand years in the past. Luke 2:39 calls Nazareth their 'own town', yet Luke 2:3 says each went to 'his own town'.

The verdict

The apologetic question is not whether the traditional reading is possible. The question is what is probable given all the evidence.

It is virtually certain that Quirinius conducted a census of Judea in 6 CE, and very likely that Luke 2:2 refers to it. It is very likely that this contradicts Luke 1:5 and Matthew 2. It is highly probable that no empire-wide census took place under Herod and that Rome would not have imposed one on his client kingdom. It is highly probable that Roman censuses did not require people to travel to ancestral towns.

Each apologetic scenario is possible, but several depend on unattested figures or events. None reaches probable, because none explains the evidence better than the simpler account.

The bottom line: The simplest explanation accounts for Luke 2, the contradiction with Luke 1 and Matthew, the Roman practice, and the chronological error in Acts 5. Luke or his source knew of the census of 6 CE and misplaced it, likely to align Jesus's birth with Bethlehem in accordance with Micah 5:2. While some evangelical scholars still defend the account, the majority position among critical scholars is that it is unhistorical.