‘Now in those days a decree went out from Caesar Augustus to register all the empire for taxes. This was the first registration, taken when Quirinius was governor of Syria. Everyone went to his own town to be registered. So Joseph also went up from the town of Nazareth in Galilee to Judea, to the city of David called Bethlehem, because he was of the house and family line of David.’

This passage presents a number of historical challenges.

A. What are the problems?

According to Helen Bond:

‘Luke’s account contains a number of inaccuracies: the census didn’t include “all the world” (Luke 2:1), it didn’t involve Galilee, and people would have registered at their usual homes (not those of their ancestors). Furthermore, if historical, Luke’s account would contradict Matthew’s clear indication that Jesus was born shortly before the death of Herod I (that is, around a decade earlier, in 6 B.C.E.). Attempts to suggest that Luke’s reference to the “first registration” (Luke 2:2) was an earlier census carried out by Quirinius have not been successful.’

Teasing out some of the concerns raised by critics:

1. At this time, Judea was not part of the Roman Empire, but was a client kingdom under Herod the Great.  It would therefore not have been included in such a census.

2. There is no external evidence for a universal census under Caesar Augustus.  Moroever, at the time that Augustus issued this degree, Judea was not under direct Roman control, but was a client kingdom ruled by Herod the Great. It is claimed that it would not therefore have been included in any Roman census.

Helen Bond thinks that Luke is mistaken when he says that the census was taken of the ‘entire Roman world’, stating that it was taken only in Judea, Samaria, and Idumea (not Galilee).  But, according to the Lexham Geographic Commentary:

‘The solution might simply be that the registration that brought Mary and Joseph to Bethlehem was a local census limited to the land of Israel, yet part of a larger effort to gather statistics for the entire empire.’

Brown (Birth of the Messiah) suggests the same thing.

3. The available evidence suggests that people were not required to return to their ancestral homes at the time of a census; rather, people registered where they lived.  But the belief that Nazareth was Joseph’s original home relies on a certain (though admittedly common) translation of Lk 2:39.  But Bethlehem may well have been Joseph’s actual home.  Even if it wasn’t, it is possible that he owned some property there (so Marshall; and this would have been sufficient reason for him to travel there for the purpose of the census).  But it is difficult, in the light of Lk 2:24, to see Joseph as a man of material substance.  Or, Bethlehem remained Joseph’s legal or family home: Keener says that:

‘pottery samples suggest a recent migration of people from the Bethlehem area to Nazareth around this period, so Joseph and many other settlers in Galilee may have hailed from Judea.’ (IVPBBC, 2nd ed.)

Kruger cites a papyrus which throws some light on the question:

‘One manuscript, known as P. Lond. 3.904, provides this fascinating description of a Roman census: “It is necessary that all persons who are not resident at home for one reason or another at this time return to their homeplaces in order to undergo the usual registration formalities and to attend to the cultivation of the land which is their concern.”’

Kruger concludes:

‘Here we see, contrary to the critics of Luke 2:2, that there were times when an individual had to register in his hometown—namely, when he owned property in that town and was temporarily living elsewhere. This squares well with the Gospel of Luke, where it appears Joseph was only living in Nazareth temporarily. He was originally from Bethlehem, where he likely owned a family plot (as one of David’s descendants). Under this scenario, Luke’s description of Joseph returning to his hometown of Bethlehem proves plausible.’

4. Joseph would not have been required to take Mary with him; only the male head of the family was required to register.  This difficulty is resolved if the suggested resolution to point 2 above is accepted.

Marshall says that while this objection would apply to Roman citizens in Italy, in Syria women aged 12 and over were subject to poll tax.

5. Luke seems to have his dates wrong.  Josephus says that Quirinius took a census in AD 6/7 AD.  Quirinius was governor of Syria from 6 to 12 AD, whereas Jesus is known to have been born before the death of Herod the Great in 4 BC (cf. Lk 1:5; Mt 2:1,15f). The governor of Syria at that time was C. Sentius Saturninus (9–6 BC) or Quinctilius Varus (6–4 BC).

Evans, Edwards and others judge that the census mentioned in Acts 5:36f is that which took place under Quirinius n AD 6-7.  But this cannot be reconciled with what Luke himself writes in the present passage.

John H. Rhoads has argued that it is Josephus, and not Luke, who is mistaken:

‘the account which Josephus tells of the census conducted by Quirinius, and the corresponding revolt by Judas the Galilean, is actually a mistaken duplication, broadly speaking, of events which occurred much earlier.’

B. What are the interpretative options?

1. Some think that Luke was simply mistaken.

A sceptic such as Richard Carrier has no hesitation:

‘There is no way to rescue the Gospels of Matthew and Luke from contradicting each other on this one point of historical fact. The contradiction is plain and irrefutable, and stands as proof of the fallibility of the Bible, as well as the falsehood of at least one of the two New Testament accounts of the birth of Jesus.’

Roman Catholic scholar Luke Timothy Johnson writes:

‘Luke’s attempt at synchronism is not entirely successful, as endless technical discussions have made clear. Herod died in 4 B.C.E., and Augustus was emperor from 27 B.C.E. to 14 C.E. So far, so good. But Quirinius was governor in Syria from 6–7 C.E., and the gap can’t be filled. Luke simply has the facts wrong.’  But it is ‘unlikely…that Luke mistakenly connected Jesus’ birth to the AD 6 census, since in Acts 5:36 he refers to this census in its proper context.’ (Lexham Geographic Commentary on the Gospels)

Eric Franklin (The Oxford Bible Commentary) thinks that Luke’s account at this point ‘raises many virtually insurmountable problems’.

‘We have no evidence for an empire-wide census under Augustus and the likelihood of this including the land of a client king such as was Herod the Great is remote.’

After a brief survey of the problems, John T. Carroll concludes:

‘It is impossible to salvage historical accuracy for Luke’s report.’

For Carroll, the important thing is not the historical value of this part of the narrative, but its function within the story (i.e. getting Joseph and Mary to Bethlehem, in order for Jesus to be born there).

Jonathan Pearce devotes an entire chapter (ch. 6) seeking to demonstrate that according to Matthew, Jesus was born before 4BC (having been born before the death of Herod the Great), creating a ten-year discrepancy with Luke, who has Jesus born around 6AD (at the time of the census under Quirinius).  Mark Edward takes a similar view.

Brown (Birth of the Messiah) clearly thinks that Luke is mistaken, and that alternative interpretations are motivated by the need to protect the Evangelist’s accuracy.

With regard to the apparent lack of evidence of an empire-wide census under Augustus,

‘We know that Augustus instituted three regional censuses around the empire (Hayles 1973, 120; Fitzmyer 1981, 400). Other censuses during or near this period were instituted in Egypt, Syria, Gaul, and Spain (Hayles 1973, 127-9). In this light it is not unlikely that Augustus instituted a census for Palestine (Schürmann 1969, 100).’ (Holman Apologetics Commentary)

From this evidence, it would seem that:

‘Luke’s talk of an empire-wide edict probably reflects the ongoing census process of this era (Plummer 1896, 48; Hoehner 1977, 15). In other words, the eventual effect of the periodic regional censuses could be the registration of essentially the entire Roman territory, which to Luke’s eyes would seem like registration of “the whole inhabited world.”’ (op. cit.)

Pate (40 Questions About the Historical Jesus), similarly, thinks that it is uncertain that the text means that Augustus took an census of the entire empire.  The language could simply mean that various parts of the empire were subject to various censuses during the time of Augustus.  The Greek (suggests Pate) could mean that Augustus decreed that a census, that had been taking place in some parts of the empire, should now be extended to all parts, including clients states such as Judea.

It is not certain that Luke in 2:1 means that Augustus took one enormous census of the whole empire. The language is general and could simply mean that the various parts of the empire were subject to various censuses during the time of Augustus. The Greek says that Caesar decreed that “all of the Roman world be enrolled.” Both the present tense of apographō (“I enroll”) and the use of pas (“all”) suggest that Luke intended to say that Caesar Augustus decreed that the enrollment, which had been previously been going on in some parts of the empire, should now be extended to all parts, including client states like Judea.

Given Luke’s avowed commitment to historical accuracy, his careful use of historical markers elsewhere, and his access to at least one member of Jesus’ family (Mary his mother or James his brother), we should sceptical about scepticism in this regard.

2. Others think that two narratives – that of Herod’s death in 4 BC and that of Rome’s annexation of Judea in AD 6 (leading to the census of Quirinius) – have been conflated.  This, it is claimed, would be consistent with ancient, rather than modern, standards of historiography are applied.  This is the preferred view of S. Young in DJG 2nd ed. (art. ‘Birth of Jesus’).

So also Edwards, who concludes after a lengthy discussion:

‘Given available evidence related to the matter, it appears that the reference to Quirinius in Luke 2:2 is a conflation of the census of Quirinius with the death of Herod the Great. The two events were easy to conflate, for the death of Herod and the census of Quirinius were both epic events, and both incited massive protests that were violently suppressed by the Roman army. A full resolution of the historical problems related to the dating of the census of Quirinius seems impossible on the basis of current historical knowledge.’

However, the conflation of two well-remembered events that occurred a decade apart does seem rather a stretch.

3. Still others think that the ‘Herod’ referred to in Lk 1:5 is not Herod the Great but Archelaus, who was also known as Herod the Ethnarch, who reigned in AD 5-6.

4. Another school of thought proposes that Luke is referring to an earlier census that took place under Quirinius.

This theory was championed by Ramsay, on the basis of inscriptional evidence that (according to Ramsay) indicates that Quirinius served as governor of Syria from 11/10 bc to 8/7 bc, as well as in the later, well-attested, period.  But the inscription in question – the Lapis Tiburtinus – is fragmentary and bears no names.  Therefore, Ramsay’s theory cannot be established as fact.

McDowell, Evidence That Demands a Verdict, 71, states:

‘It was at one time conceded that Luke had entirely missed the boat in the events he portrayed as surrounding the birth of Jesus, Lk 2:1-3. Critics argued that there was no census, that Quirinius was not the governor of Syria at that time and that everyone did not have to return to his ancestral home.

‘First of all, archaeological discoveries show that the Romans had a regular enrollment of taxpayers and also held censuses every 14 years. This procedure was indeed begun under Augustus and the first took place in either 23-22 BC or in 9-8 BC. The latter would be the one to which Luke refers. Second, we find evidence that Quirinius was governor of Syria around 7 BC. This assumption is based on an inscription found in Antioch ascribing to Quirinius this post. As a result of this finding, it is now supposed that he was governor twice – once in 7 BC and the other time in 6 AD (the date ascribed by Josephus). ‘Last, in regard to the practice of enrollment, a papyrus found in Egypt gives directions for the conduct of a census. It reads, “Because of the approaching census it is necessary that all those residing for any cause away from their homes should at once prepare to return to their own governments in order that they may complete the family registration of the enrollment and that the tilled lands may retain those belonging to them.’

Three inscriptions are often cited as providing supporting evidence for this.  But most commentators (believing as well as sceptical) find this theory unconvincing.  It is not intrinsically improbable,

‘that Quirinius was involved with a census during the last years of Nero. Toward the end of his reign Herod fell out of favor with Rome (c. 8/7 B.C.). This was followed by his sons engaging in an intense struggle for the throne at a time when Herod was extremely ill. All of these factors would allow for the Roman government to take a census in his land in order to assess the situation before his death. Although it is difficult to pinpoint the exact year of the census, it was probably sometime between 6 and 4 B.C.’ (DJG, 1st ed.)

According to Bock (Holman Apologetics Commentary),

‘If prōtō is best rendered “first,” then Luke was calling this the “first” census that occurred during Quirinius’s governorship. This could indicate that Luke knew of two governorships for Quirinius and two censuses associated with him. The fact that in Acts 5:37 Luke referred to a census best identified as the one that occurred under Quirinius in AD 6 strengthens this possibility. Thus if Luke meant to differentiate between two separate censuses taken by Quirinius, it is best to assume that the census he discussed in chapter 2 of his Gospel is not to be confused with the better-known Quirinius census of ad 6.’

Keener (IVPBBC) inclines towards this view.

Gleason Archer (Encyclopedia of Bible Difficulties) notes that Luke refers to this as the ‘first’ census, strongly implying that there was at least one more:

‘Luke was therefore well aware of that second census, taken by Quirinius again in A.D. 7, which Josephus alludes to in the passage cited above. We know this because Luke (who lived much closer to the time than Josephus did) also quotes Gamaliel as alluding to the insurrection of Judas of Galilee “in the days of the census taking” (Acts 5:37). The Romans tended to conduct a census every fourteen years, and so this comes out right for a first census in 7 B.C. and a second in A.D. 7.’

Michael Kruger notes that Justin Martyr (middle of 2nd century) refers to this census, but calls Quirinius ‘epitropos‘ (procurator), rather than ‘governor’.  Interestingly, Luke doesn’t refer to Quirinius as a ‘governor’, but as a hegemon – another word for procurator.  Tertullian (AD 160–240) says that the census took place under Saturninus.  So why did Luke mention Quirinius, and not Saturninus? Probably, because the former probably implemented the census under the latter, and became well-known for the later census (AD6), by which time he had become governor.

Jonathan Pearce (The Nativity: A Critical Examination) expends most of his energy in ch. 7 of that work to exposing the weaknesses in the theory that Quirinius was governor at two separate periods.  Unfortunately, he does not even mention other, arguably more plausible explanations (such as point 7, below).

5. As a variation on the above, it has been suggested that Quirinius did indeed instigate the census during the time of King Herod, but was only able to complete it when he became governor of Syria some years later.  (It is known that censuses could take years to complete – a census that was begun in Gaul around 6 BC took around 40 years to complete.)  Luke’s Greek may support this view: ‘This census became important/prominent when Quirinius was governor of Syria’.  As a futher variation on this theme, it has been suggested that the census was instigated under an earlier governor (Varus) and completed later, under Quirinius.

According to the Lexham Geographic Commentary:

‘It is unlikely, however, that Luke mistakenly connected Jesus’ birth to the AD 6 census since in Acts 5:37 he refers to this census in its proper context. As for why Luke mentions Quirinius before he was officially governor of Syria, it may be that Quirinius completed a census begun by his predecessor.’

6. The 19th-century commentator Albert Barnes refers to another possible solution, according to which the title ‘governor of Syria’ is anticipatory; he later became governor of Syria, and is so titled here to identify him as the one who later became governor:

‘The passage here means, “This was the first census of Cyrenius, governor of Syria.” It is called the first to distinguish it from one afterward taken by Cyrenius, Acts 5:37. It is said to be the census taken by Cyrenius, governor of Syria; not that he was then governor, but that it was taken by him who was afterward familiarly known as governor. Cyrenius, governor of Syria, was the name by which the man was known when Luke wrote his gospel, and it was not improper to say that the taxing was made by Cyrenius, the governor of Syria, though he might not have been actually governor for many years afterward. Thus Herodian says that “to Marcus the emperor were born several daughters and two sons,” though several of those children were born to him before he was emperor. Thus it is not improper to say that General Washington saved Braddock’s army, or was engaged in the old French war, though he was not actually made general till many years afterward. According to this Augustus sent Cyrenius, an active, enterprising man, to take the census. At that time he was a Roman senator. Afterward he was made governor of the same country, and received the title which Luke gives him.’

7. David J. Armitage has argued that Lk 2:1-5 constitutes a digression, in which Luke points forward to events that occurred several years after the birth of Jesus.

Armitage suggests that:

‘Luke 2:1-5 does actually refer to the AD 6 census as described by Josephus, and that Luke introduces it as part of a brief digression—what we might call a ‘flash-forward’—in which he describes a return visit by Mary and Joseph from Nazareth to Bethlehem some years after Jesus was born there. Mentioning this return visit, which could have involved registration of property that Joseph still owned in Bethlehem (his original hometown), would presumably serve to emphasise the official connection of the family of ‘Jesus of Nazareth’ with Bethlehem, the town of David.’

The text would then read something like this:

‘The child grew and was strengthened in spirit, and he was in the wilderness until the day of his public appearance to Israel. As it happens, it was during that time that a decree went out from Caesar Augustus to register all the Roman world (this was the first registration, when Quirinius was governor of Syria), and everyone went – each into their own town – to be registered. Joseph also went up: out of Galilee, away from the town of Nazareth, into Judea, to David’s town (which is called Bethlehem) because he was from the house and family of David; he went to be registered with Mary (she who was his betrothed when she was pregnant).’

‘Now, it transpired that the days were completed for her to give birth when they were in that place, and she gave birth to her firstborn son and wrapped him in cloths and laid him in a feeding trough, because there was insufficient space for them in their lodging place.’

The argument, as summarised by Andrew Wilson, consists of the following steps:

‘1. Luke is clear in 1:5 that the timeframe for the birth narratives is set “in the days of Herod the King.” That is when Zechariah meets the angel, which is quickly followed by Elizabeth’s pregnancy, which overlaps with Mary’s (1:26, 36). So Luke knew perfectly well that Jesus was born during the reign of Herod the Great, and thus by 4 BC.

2. The phrase “in those days”, in Luke, refers back (as usually in Luke) to the preceding verse, which refers to the time John the Baptist was growing up in the desert (1:80). The census is therefore placed while John is maturing, rather than when John is a newborn.

3. Ἐγένετο δὲ (“and it happened”) marks a transition from narrative background (the growing up of John) to a specific narrative sequence that occurs against that background (the census). This is how the phrase often functions in Luke-Acts.

4. Assuming for a moment that Luke’s first readers knew the census under Quirinius had taken place in AD 6, and that Herod the Great had died in 4 BC, the reference to the census would make clear to them that a narrative digression was taking place from 2:2.

5. Luke says that Joseph went back εἰς τὴν ἑαυτοῦ πόλιν (“to his home town”), which this reading takes at face value; he was not merely returning to a town that his ancestors had come from.

6. The participial phrase τῇ ἐμνηστευμένῃ αὐτῷ, οὔσῃ ἐγκύῳ (“the one betrothed to him, being pregnant”) identifies Mary as the person we met in chapter 1, rather than affirming that she was betrothed and pregnant at the time of the census. The grammar is ambiguous here, and could go either way.

7. The Ἐγένετο δὲ of 2:6 indicates a return to the main narrative, following the digression. This, again, fits with the way the phrase is used elsewhere in Luke-Acts.

8. The clarification that Jesus was born in Bethlehem (2:6) is intended to emphasise place rather than time.

9. As many have pointed out, often with an iconoclastic grin, the famous κατάλυμα (“inn”) was probably a room within a private house, rather than a commercial inn.’

8.  Others think that the word prōtē in Lk 2:2 should be translated either ‘earlier’ or ‘before’, rather than ‘first’.

This is the view taken by a number of competent scholars, including F.F. Bruce, Nicholas Perrin, N.T. Wright, Craig Blomberg, John Nolland, Marvin Pate and David Garland.

Garland translates: ‘This registration was before Quirinius governed Syria.’

Fizmyer: This option is a ‘last ditch solution to save the historicity involved’.

To the contrary, there are sound reasons for adopting this interpretation.  Garland points out that:

(a) The translation ‘first’ would be odd, because no second or subsequent census is mentioned.

(b) This was not, in fact, the ‘first’ census that had been imposed on the Jews.  Herod, in his attempt to Romanise his territory, would had carried out several.

(c) It make sense for Luke to allude to the later census under Quirinius, because it was more memorable than the ongoing process of census and taxation carried out by Herod.  Indeed, the later census had caused a rebellion, according to Josephus.

(d) All of this accords well with Luke’s mention of ‘the census’ and related revolt in Acts 5:37.

(e) Time keeping in the ancient world was not by means of clocks and calendars, but by reference to significant events.  The mention of the census of Quirinius provides a chronological signpost for the birth of Jesus.

(f) In contrast to Judas the Galilean, who led a revolt against the census (which we take to be the census under Quirinius, Acts 5:37), Jesus’ family complied peacefully with the demands of the census.

(g) Luke’s mention of the later, more infamous, census underscored the contrast between the proud earthly empire of Caesar and the reign of peace which God’s Messiah will usher in.

F.F. Bruce states that

‘The Greek of Luke 2:2 can be translated: “This enrollment (census) was before that made when Quirinius was governor of Syria.”  Because of the construction of the sentence (why say that this was the first census, when there was no other to compare it with?), this is not an unlikely reading. In this case there is no problem, since that census of A.D. 6 is well known to historians.’

Wright (Who Was Jesus?, p89 and other publications) concludes:

‘My guess is that Luke knew a tradition in which Jesus was born during some sort of census, and that Luke knew as well as we do that it couldn’t have been the one conducted under Quirinius, because by then Jesus was about ten years old. That is why he wrote that the census was the one before that conducted by Quirinius.’

This is also the view of Blomberg (The Historical Reliability of the New Testament), who writes:

‘For a while it appeared that new archeological evidence would support a joint rule of some kind between Quirinius and another Roman appointee at an earlier date, but this has not materialized. I am now more inclined to suggest a straightforward alternative translation: “This census took place before Quirinius was governing Syria” (NIV mg). Although prōtos elsewhere in Luke always means “first,” the second most common meaning of the word is “before,” and the entire Greek clause is notoriously ambiguous because Luke did not use any articles to help make his meaning more precise. The most literal translation that is still intelligible in English is, “This census was first/ before Quirinius governing Syria” (hautē apographē prōtē egeneto hēgemoneuontos tēs Surias Kurēniou). The text certainly can mean, “This census was the first while Quirinius was governing Syria,” but one would normally expect an article before apographē and again before prōtē if that were Luke’s intention. But we could translate, “This census was before [one] when Quirinius was governor.” The census in AD 6 under Quirinius was particularly infamous because it provoked the failed rebellion by Judas the Galilean. So it would be natural for a biographer or historian to refer to an earlier census with reference to the later, much better-remembered one.’

Nolland states:

‘The governorship of Quirinius was an important turning point in Judean history, marking as it did the annexation of Judea, which was made profoundly visible by the census registration with which Quirinius governorship began. That registration was the registration, (cf. Ac 5:37) and it is natural that Luke should distinguish from it a preliminary registration in the time of Herod the Great…This seems better than forcing an earlier governorship on Quirinius and more likely than the contradiction in the Lukan infancy narratives created by an identification of the census here as that of A.D. 6.’ (WBC)

Garland thinks that the census that drove Joseph and Mary to Bethlehem was instigated by Herod.  Luke mentioned the later census of Quirinius as a notable historical marker.  (We might add that, from the vantage point of Luke’s date of writing, the time difference between the earlier census and the later, better-known census would not be particularly significant in a culture without general use of calendars).

Marshall allows this as a possibility, but concludes:

‘No solution is free from difficulty, and the problem can hardly be solved without the discovery of fresh evidence.’

Franklin thinks that this, though it is the best attempt at harmonization,

‘is not…a natural reading of the Greek and has about it something of the air of desperation.’

So also Edwards:

‘The near-universal denotation of prōtē (and clearly its connotation in v. 2) is “first in sequence” rather than “before” (which would require proteros).’

Richard Carrier considers it to be a grammatically impossible reading, Of course, I am not competent to assess what is, and what is not, ‘a natural reading of the Greek’, but I notice that a number of those who are competent are happy to accept this reading.

Marshall:

‘The form of the sentence is…odd, since it is hard to see why πρῶτος was introduced without any object of comparison, and it may be that πρῶτος should be understood as a comparative with the meaning ‘before’. Luke does write loose sentences on occasion, and this may well be an example of such.’

It can be plausibly argued that Bethlehem was, in fact, Joseph’s own home, and that he went to Nazareth in order to be betrothed to Mary.  This would explain why Matthew makes no mention of Nazareth.

Pate inclines to the view just mentioned, and summarises Witherington’s conclusion, that it is probable that:

‘Luke is referring to a census under Quirinius that took place prior to the famous one in AD 6–7. If so, we have no clear record outside Luke of such an action by Quirinius, though it is not impossible that it took place. Herod’s power was on the wane at the time of Jesus’ birth, and a census in preparation for the change of power could well have been forced on Herod since he had fallen into some disfavor with Augustus near the end of his life. We know also that Quirinius had been made consul in 12 BC and a person of his rank serving in the East frequently had far-reaching authority and duties. It is thus not improbable that, acting as Caesar’s agent, he had Herod take a census.’

Mark Edward, after surveying other attempted explanations, and judging them to be unsatisfactory, fails to mention this proposal.

In connection with this interpretation, we might profitably return to Acts 5:37, where there is a simple reference to ‘the census’.  This suggests that it was a major event, one that had lodged itself in the mind of the people for many years.  This fact itself would help to explain why Luke (in his Gospel, chapter 2) might refer to the census that brought Joseph and Mary from Nazareth to Bethlehem as occuring before this better-known census.

Jason Engwer adds that this later census clearly took place in turbulent times.  No such turbulance is apparent in Luke 2 –

‘Mary, Joseph, the shepherds, etc. interact with people, travel, and so on without any mention of the unrest surrounding the 6 A.D. census or any suggestion that the individuals involved in the opening chapters of Luke were concerned about such matters.’

We have, then, some evidence (but not proof) that the census of Luke 2 and that of Acts 5 are separate and equally historical events, or, according to Engwer, part of the same, prolonged, event.

9. Some scholars express a degree of agnosticism on this issue

These include some of those mentioned above.

After a brief review of the available options, Stein concludes:

‘It must be confessed that there is no easy explanation at the present time for this historical problem of the census date, but some new evidence might in the future vindicate the historical accuracy of Luke on this point.’

Conclusion

The accounts of Jesus’ birth offered by Matthew and Luke have much in common:-

Jesus was born in Bethlehem            Matt 2:1                   Luke 2:2

In time of Herod (d. 4 BC)                  Matt 2:1                   Luke 1:5

Mother: Mary                                      Matt 1:18                 Luke 1:26

Father: Joseph (named the child)      Matt 1:18                 Luke 1:26

But not the biological father               Matt 1:16, 20, 22     Luke 1:34; 3:23

Brought up in Nazareth in Galilee      Matt 2:22-23           Luke 2:39

From the line of David                        Matt 1:1                  Luke 1:32

We have good grounds for regarding Luke as a reliable historian.  His demonstrable accuracy in the Acts of the Apostles led archaeologist Sir William Ramsay from scepticism to respect in this regard.  It is clear enough that

‘Luke was not confusing this with the one held in A.D. 6 because that was just after the deposition of Herod’s son Archelaus, whereas the context of the birth narrative of Jesus was in the days of Herod the Great.’ (DJG)

See this.