Luke vs. Matthew on the Year of Christ's Birth by Richard Carrier, Ph.D. (2006)
It is widely acknowledged that Quirinius became "governor of Syria" in 6 C.E., only then conducting a census of Judaea, and that Herod the Great died in 4 B.C.E., ten years before. Since Matthew indisputably claims Jesus was born while Herod the Great was still alive, while Luke indisputably claims Jesus was born when Quirinius was governor of Syria during a census of Judaea, Luke and Matthew are clearly in contradiction regarding when Jesus was born. They disagree by at least ten years, which entails one of them has made a historical error (or both have).
Because this contradiction is so clear and certain and strongly backed by evidence, inerrancy proponents have invented a dizzying array of attempts to remove this contradiction by reconciling the details in Matthew and Luke. I have examined and researched these efforts in thorough detail, perhaps more than anyone. I have found all of them unsuccessful, even ludicrously so. Consequently, the primary importance of this contradiction is that it is one of the clearest and most irrefutable examples of historical error in the Bible, which is perhaps why it has generated so many desperate attempts to wiggle out of it.
And that is the second reason this error is so important: short of the vast and diverse "Bethlehem Star" literature, there is probably no other biblical error for which so many false, groundless, or implausible arguments have been contrived to "invent" or "revise" the historical facts of the ancient Roman world. As a teacher and a scholar, I find all this disinformation and wanton invention about the period I study quite appalling. And because there are so many such contrivances, relating to technical details of the social and political history of Rome that are difficult if not impossible for the average layman to investigate, duty demands that some impassioned expert do all the necessary research and make it available to the common reader. Otherwise, all those false claims might simply be believed and eventually become common assumptions.
I am a published author with considerable qualifications in the study of ancient history and languages (see inset), which should give my findings some weight of authority. I thoroughly document all the arguments and evidence relating to this discrepancy in an extensive article at the Secular Web, The Date of the Nativity in Luke (5th ed., 2006). I will provide here only a summary of the major conclusions of that article, along with some of the most relevant evidence.
Invariably people write to me proposing some "new" argument, without having bothered to read the complete article above, only to learn that their "new" argument is in fact "old" and already refuted there. I have received countless such emails to date, over several years, and have yet to hear of any fact or argument I haven't already addressed. I advise all challengers to read that original article in its entirety before proposing to challenge the fact that Luke contradicts Matthew on the year of Christ's birth.
The present summary is no substitute for that, and is only provided here to conveniently summarize the current status of this biblical error for readers not intent on gainsaying what they don't like to hear.
BACKGROUND OF THE PROBLEM
The Gospel of Luke says (2:1-6) Jesus was born during a census, which the historian Josephus records took place after Herod the Great died, and after his successor, Archelaus, was deposed after a ten year reign (Antiquities of the Jews 17.342-55, 18.1-2, 18.26). But the Gospel of Matthew says (2:1-3) Jesus was born when Herod the Great was still alive, possibly two years before he died (2:7-16), and before Archelaus even took office (2:19-22). On a plain reading of the Bible, this is a contradiction. Someone erred.
Efforts to challenge this conclusion fall into three general categories: either Luke meant something other than his words actually say, or Josephus failed to mention some "other" time that Quirinius was governor and a census made of Judaea, or Josephus erred in dating relevant events. A fourth option, of arguing Matthew meant something other than what he said, is blocked by the absolute clarity with which he said it. There is no rational way to argue that Matthew was referring to the political situation anywhere near 6 C.E. Not only does Matthew's narrative make this clear, but the physical evidence from coins of the region leave little room for disputing that Herod ceased his reign in 4 B.C.E., Archelaus then succeeded him, then ceased his own reign ten years later in 5 C.E., and Roman control of Judaea began in the year 6 (the year the earliest Roman coins struck for Judaea begin).1 These facts are also recorded by Josephus (who is generally quite reliable on matters of public chronology) and partly corroborated by another historian, Cassius Dio (e.g. Roman History 55.27.6).
Consequently, to this day no one has attempted to argue that Matthew was describing an event of 6 C.E. Instead, all efforts are directed to arguing that Luke was describing an event five to twelve years earlier than we think, either by trying to harmonize Luke with Josephus, or by attacking the accuracy or alleged omissions of Josephus. However, all such attempts have required inventing or distorting socio-political facts of the period, or twisting Greek grammar or vocabulary beyond anything that would be recognizable to an ancient reader. Allowing either tactic would permit us to prove that no text ever written in human history has ever been in error about anything. Such an irrational consequence convicts both tactics as unsound.
ATTEMPTED SOLUTION #1 - Looking in Luke for a Different Date
Some observe that Luke says John the Baptist was born during the reign of "Herod the King" (1:5) and appears to have Jesus born less than a year later (1:22-24, 1:31-36, 1:80, 2:1, 2:40-42), which appears to agree with Matthew. However, Matthew does not mention or date the birth of John, and despite the impression given by English translations, Luke is unclear how much time actually passed between his birth and that of Jesus. More importantly, Archelaus was also called Herod (even on his own coins) and even Josephus calls him a king (Antiquities of the Jews 18.93). Unlike Matthew, Luke provides no detail indicating either he or his source meant anything other than Herod Archelaus when dating the birth of John. Therefore, unless we assume Luke is contradicting himself, we can't assume he dated either the birth of John or Jesus to the time of Herod the Great. So there is no case to be made from 1:5 that Luke agreed with Matthew.
Likewise, when Luke dates the start of John's ministry to 28 C.E. (3:1) and then over twenty verses later says Jesus began his own ministry at "about thirty" (3:23) some assume the two ministries began the same year, which would place the birth of Jesus at "about" 3 B.C.E. which for a "rough" estimate is close enough to fit Matthew. But Luke never says the two ministries began the same year, and for various reasons it's unlikely they did. Luke clearly didn't know the year Jesus started his ministry, since he didn't know how old he was, despite claiming to know exactly when he was born. Since "about" thirty can be off by at least four years (26-34), and since Luke allows some time to pass between the start of John's ministry and the baptism of Jesus, and since scholars agree Jesus could have begun and ended his ministry anytime between 28 and 33 C.E., we are left with a window between 7 B.C.E. to 7 C.E. for his birth, far too wide to pin down. So there is no good case to be made from Luke 3:1-23 that he agreed with Matthew.2
ATTEMPTED SOLUTION #2 - Inventing Another Governorship for Quirinius
Even fishing a different date out of Luke would leave a contradiction within Luke, since the only chronological detail about Jesus that Luke is absolutely clear on is that he was born during "the first census when Quirinius was governing Syria" (2:2). All evidence confirms that Quirinius first became governor of Syria in 6 C.E. and the first Roman census of Judaea occurred at that time, and Luke clearly says this was a Roman census (2:1-6). So Luke didn't leave much room to maneuver. To reconcile Luke with Matthew, one must invent two facts nowhere in evidence: some other Syrian governorship for Quirinius and some other census affecting Judaea, both before Herod the Great died.
But trying to invent an earlier Syrian governorship for Quirinius is a lost cause. Not only is there no evidence of it, and not only does it go against a plain reading of all the evidence we do have, but it's essentially impossible. No one ever governed the same province twice in the whole of Roman history. So the claim that Quirinius was the sole known exception is so extraordinary it certainly can't be maintained without evidence. Such an astonishing and unique honor could not have been omitted by Josephus or Tacitus (Annals 3.48), yet both describe his career without any mention of it. Historical evidence also confirms other men governed Syria between 12 and 3 B.C.E., so Quirinius could not have been governor then, and he was not qualified to hold that office before the year 12.3
Stymied by all these facts, inerrantists have resorted to everything from fabricating evidence of dual governorships or other fictional offices Quirinius is supposed to have held, to changing the year of Herod's death. None of this is even remotely reasonable, and most of it is based on the fantasies of amateurs or the abandoned conjectures of long dead historians. First, the alleged physical evidence:
The Lapis Tiburtinus
This is a headless (and thus nameless) inscription that the Vatican has taken the liberty to "restore" with the name of Quirinius. It is then "interpreted" as saying he governed Syria twice. But the actual inscription does not say anyone governed Syria twice, nor does it belong to Quirinius. Scholars now believe it belongs to Lucius Calpurnius Piso, since it fits what we know of his career very well, while no basis exists for claiming it belongs to Quirinius. And even if it did, it doesn't say anything about governing Syria twice. It says the honoree "received the governorship of Asia and then again of Syria."4
The Lapis Venetus
This is an inscription that really does mention Quirinius. It is the epitaph of Aemilius Secundus and reports he helped Quirinius conduct a census when the latter was governing Syria. So this inscription confirms that a census was taken of Syria when Quirinius was governor. It does not give a date, either for the census or the inscription itself. But there is no reason to believe this is a reference to any other census under Quirinius except the only one we know of, that of 6 C.E. Several inerrantists have simply "invented" early dates for this inscription, and then used these fabricated dates to claim this inscription as proof there was an earlier census under Quirinius. It proves no such thing.5
The Antioch Stones
These are two stones commemorating the offices of Gaius Julius Caristanius Caesiano, both mentioning that he held the deputy management of a city duumvirate held by Quirinius. The date is unknown but probably before the year 1. Conjuring various fantasies, inerrantists finagle this city office into evidence of an earlier governorship of Syria, but no rational argument can produce that conclusion. First, a duumvirate is a city office and has nothing to do with a provincial governorship. Second, this duumvirate was held in Galatia, not Syria.6
The Vardaman Coins
Jerry Vardaman claimed to have discovered microscopic letters literally covering ancient coins and inscriptions conveying all sorts of strange new facts, which he used to completely rewrite history. One of these amazing new "discoveries" was evidence Jesus was born in 11 or 12 B.C.E. Needless to say, Vardaman's claim constitutes fringe quackery that has gained no respect in the academic community. I inspected one of these coins myself under a magnifying glass and a digital microscope at the British Museum and found none of these amazing microscopic letters. Case closed.7
Since none of this evidence supports an earlier governorship for Quirinius, and all other evidence makes such a thing virtually impossible, only two strategies remain for the inerrantists: either Quirinius held some other "special command" in Syria and wasn't governor per se, or Quirinius held an unrecorded "dual-governorship" with some other governor. Neither of these proposals makes any sense in the context of Roman politics or historiography.
First, the "sub-command" thesis. Luke's choice of vocabulary is somewhat imprecise, using a word that can refer to many different positions of command. Seizing on this, inerrantists argue that Luke meant "when Quirinius was holding a command in Syria," and not "when Quirinius was governing Syria." But stretching the word like this requires ignoring the grammar. Luke says "of Syria," not "in Syria," and thus he could not have been referring to some command in Syria but only a command of Syria. Even if we ignore Luke's grammar, the only real "command" anyone can find for Quirinius is a war he fought in Galatia, probably between 6 and 1 B.C.E. But there is no logical way Luke would refer to a census in Syria by referencing a war in Galatia, and no one would ever write or read "governing Syria" as meaning "fighting a war in Galatia." Unless Luke was a profoundly stupid man, or erred in his historical facts, he would have named the actual governor of Syria who oversaw a census in Judaea, not some unrelated officer in a faraway province.8
A completely different tactic, to get around the problem that all the governors of Syria between 12 and 3 B.C.E. are already known, is to claim Quirinius was holding a dual governorship with one of those other governors. Not only does this still require claiming Quirinius governed the same province twice, an oddity never before recorded in the history of Roman politics, but it also requires completely inventing the idea of a "dual governorship." Since there is no evidence in all of Roman history of any province assigned two governors at the same time, this is another extraordinary claim that requires evidence to be believed. Since there isn't any, inerrantists invent some.
Josephus is cited as saying "Saturninus and Volumnius were in charge of Syria" (Antiquities of the Jews 16.280) which is said to "prove" Syria was special enough to be assigned two governors. But Volumnius was not a governor. He was a procurator (Josephus, Jewish War 1.538), an office held only by men of fundamentally inferior rank, who were not even qualified to hold the office of a provincial governor. Conversely, a Roman who had achieved senatorial, and even consular rank--like Quirinius--would never deign to accept such a humiliating office as procurator. Socially, this would be as unbelievable as a United States president taking a job as shift manager at a local McDonald's. In Roman society, this would be so remarkable and unprecedented that, again, Josephus and Tacitus would not have omitted it from their accounts of his career. The same error is made using an inscription reporting that two "deputies," Rutilius Gallicus and Sentius Caecilianus, were assigned to the province of Africa. But this inscription clearly states that one of them was a praetor, and thus not of consular rank. So again, what we have here are not two governors, but a governor and his subordinate. Neither example supports even the conjecture that Quirinius could have held a dual-governorship, much less that he ever did so--or that anyone ever did. It would also make no logical sense for Luke to name a governor's subordinate rather than the actual governor of Syria.9
So there is no basis for that claim, either. Since Josephus records and thus confirms an actual census under Quirinius in 6 C.E. when Quirinius was, in fact, governor "of Syria," all exactly as Luke says, there is no plausible case to be made that Luke had any other event in mind. All the evidence we have corroborates this conclusion, and none supports any alternative or renders any even remotely plausible.
ATTEMPTED SOLUTION #3 - Inventing a New Date for Herod's Death
Since there is no reasonable way to get Quirinius to be governor of Syria anytime before 3 B.C.E., the natural last-ditch resort is to argue that Herod didn't really die in 4 B.C.E. Since there is no clear evidence who was governing Syria after 3 B.C.E., or where Quirinius was in those years, inerrantists fantasize that their imaginary "earlier governorship" of Quirinius fell around then and simply failed to be mentioned. This still doesn't avoid all the problems noted before--from a total lack of evidence to the extraordinary implausibility of a second governorship. It also requires rewriting history.
Josephus already says Varus, not Quirinius, was governing Syria when Herod died (Jewish War 1.9-10), and despite attempts to argue otherwise, Josephus is very clear and precise in his chronology for these events and cites several first-hand sources for them, while the manuscript tradition for the relevant details is completely sound, so there is no plausible case to be made that he is mistaken.10 Likewise, as mentioned earlier, evidence from coins corroborates all of this, including the reigns of Herod's successors, Philip, Antipas and Archelaus. The reign of Archelaus is further corroborated by Cassius Dio (55.27.6).
Josephus also mentions a lunar eclipse soon before Herod's death, and astronomers note there was such an eclipse in 5 B.C.E. and 1 B.C.E. Inerrantists therefore want Herod to have died in or shortly after 1 B.C.E. However, not only is all evidence against such a notion, but the Jewish Scroll of Fasting records the calendar day of Herod's death, and it preceded that of the eclipse of 1 B.C.E, but not that of 5 B.C.E. Since Josephus says his death followed (not preceeded) an eclipse, the eclipse Josephus mentions was probably that of the year 5. In the end, there is simply no evidence Herod died later than the year 4, and no plausible case to be made that he did.
ATTEMPTED SOLUTION #4 - Inventing Another Census
There is no reasonable case to be made that Quirinius ever "governed Syria" before 6 C.E. Yet inventing such a fantasy contrary to all evidence and precedent is not even enough to eliminate the contradiction between Matthew and Luke. A census of Judaea before 6 C.E. must also be invented contrary to all evidence and precedent. This is because Luke's description establishes three facts: the census affected Judaea (2:4-5), it was conducted under the administration of Syria (2:2), and it was specifically a Roman census (2:1). These three facts rule out every attempt to argue that Luke meant some other census or event than the one recorded for 6 C.E.
Before 6 C.E. Judaea was a nominally free kingdom, not a Roman province. Having sided with Augustus in the civil war that established him as emperor, Judaea was granted a favorable treaty assuring relative independence. This is proven by the coin evidence that Judaea continued to be governed by its own kings and rulers, not Roman officials, until 6 C.E., and extensively confirmed by Josephus and Cassius Dio. Though such "allied kingdoms" were kept under a tight leash and informally controlled and meddled with, all evidence regarding the legal and political practices of Roman emperors in the first century and before confirms that these states were not subject to direct Roman administration, taxation or levies. That was, in fact, the very point of not annexing them as provinces: not only to reward friendly states (and thus encourage other states to be friendly), but to avoid the headache and expense of taking over a region that was already pacified, subservient, and paying sufficient dues.
Therefore, it is historically impossible that a Roman census was conducted under a Roman provincial governor when Judaea was still an allied kingdom. But it was typical and logical that immediately upon annexing a new territory a census would be taken of it. This was necessary to begin direct taxation and levies. So when Josephus describes Archelaus being removed from office, then Judaea being annexed to Syria and placed under the Roman command of Quirinius and his prefect Coponius, and then a census being conducted for the specific purpose of taking account of what Archelaus had left them, this description makes complete historical sense. In contrast, no other hypothesized "census" scenario makes any historical sense at all.
As Josephus reports, and as all logic and precedent entail, Judaea was not being directly taxed by Rome nor administered by Romans before the year 6, and therefore there would be no purpose for Augustus to order a census there (Luke 2:1). Since forcing such a census on an allied kingdom in violation of its honor and its treaties would be such an astonishing and devastating insult contrary to all known precedent, there is no way it wouldn't have been noticed by historians like Josephus, nor any reason the Romans would undertake such a pointless and dangerous task. They would have nothing to gain by it, and plenty to lose, and Augustus was not so reckless as to think otherwise.11
Nevertheless, inerrantists must have the impossible in order to save their bible from error. So these are the arguments they have attempted:
Luke Meant "Before Quirinius"?
The word translated "first" in Luke 2:2 can in certain contexts mean "before." But for various reasons such a meaning would not be grammatically correct in this case. Luke can only have meant, and all his readers would only have understood his sentence to mean, the first Augustan census that happened under Quirinius. And that is how all translators correctly interpret it.12
Romans Subjected Allied Kingdoms to a Census?
Since the very idea of Romans conducting a census of an allied kingdom is wholly implausible and unprecedented, inerrantists have tried to invent evidence of it. The same Lapis Venetus discussed above is cited as "evidence" that the free state of Apamea was subject to an official Roman census. However, neither that census nor the inscription itself is dated, and as noted above there is no reason to believe the inscription refers to any other census than that of 6 C.E. Regardless, since no city named "Apamea" was free after 12 B.C.E. and Quirinius was not of consular rank before that year, it is impossible that the Apamea referred to in this inscription was an independent kingdom at the time. The only other "evidence" offered is a census revolt put down by legions in Cappadocia in 36 C.E. (Tacitus, Annals 6.41). Since Cappadocia was annexed as a Roman province in 17 C.E. (Annals 2.42, 2.56) this census was clearly not of an allied kingdom.13
Jesus Born During Census of 8 B.C.E.?
There is a modern myth that the Romans regularly conducted a census of their empire every 14 years. This is not true. There was little coordination between censuses of Roman citizens and censuses of provincial inhabitants, and rarely any fixed period of years between censuses, not even for citizens. Though Egypt continued to maintain a 14-year census cycle that the Romans inherited from the previous Ptolemaic government, this was not extended to or consistently coordinated with any other province. Other provinces were assessed when they could be, often at various different times from each other.14
Nevertheless, this myth of a 14-year cycle is often used to support a claim that it would have been the census of 8 B.C.E. when Jesus was born, during the governorship of Saturninus rather than Quirinius. Why? Because some claim Tertullian said Jesus was born during the census of Saturninus (Against Marcion 4.19). But Tertullian doesn't say that. He says "censuses were conducted in Judaea by Sentius Saturninus" that confirm Jesus had brothers. Since these brothers had to have been born after Jesus, Tertullian cannot be referring to any census during which Jesus was born. There was in fact another Sentius Saturninus who was governor of Syria from 19 to 21 C.E. (the son of the former Sentius Saturninus) and it's almost certain this is whom Tertullian means.15
So there is no support for linking Luke's census with any census of 8 B.C.E. Though we know there was a universal census conducted in that year, it was only of Roman citizens, not the provinces. It therefore could not have been the census Luke describes, which clearly affected non-citizen inhabitants of Judaea. Indeed, Judaea was not even a Roman territory in the year 8, nor was Quirinius governing Syria then.
Since all those arguments fail, the last resort is to claim that it wasn't really a Roman census but a census conducted by Herod the Great. The immediate problem with this is that Luke does not say any such thing. He is quite clear that he means a census ordered by Augustus, not Herod, and carried out under Quirinius, specifically in connection with Syria, not under Herod independently. Judaea was annexed to Syria under Quirinius in 6 C.E. and immediately subject to a census. Obviously that's the census Luke means.
Nevertheless, the desperate plea is made that Luke "really meant" an unknown Herodian census. Apart from resting on no evidence at all, this claim is implausible for a number of reasons. Herod had no need of conducting a census, for the tribute owed him and that he owed Rome was a fixed annual sum. It didn't matter how many people were paying. Yet a census entails a vast outlay of expenses to cover administration and recordkeeping, and ties up a considerable amount of manpower. It also entails a significant inconvenience to the population, as even Luke's description makes clear. The Jews also had a tradition of cultural and moral hostility to a peacetime census. For example, 2 Samuel 24:1-17 and 1 Chronicles 21:1-17 depict the very idea of a peacetime census as sinful and Satanic, and when the Romans finally started subjecting Judaeans to a census in 6 C.E., many violently rebelled (Acts 5:37 and Josephus, Antiquities of the Jews 18.1-8 & 20.102 and Jewish War 2.433-34 & 7.252-54). For all these reasons, Herod had nothing to gain and plenty to lose by conducting his own census. To go ahead and do it anyway would have been so remarkable and unprecedented it could not have been omitted by historians like Josephus. Yet somehow no one noticed this remarkable census. And since Herod's involvement would be the most remarkable thing about it, it's inexplicable why Luke never mentions this, but only links the census to Roman decrees, Roman magistrates, and Roman provinces.16
Stymied again, inerrantists resort to the last ditch effort of claiming Luke didn't really mean a "census" but an "oath-taking." And since according to Josephus (Antiquities of the Jews 16.136, 17.34-43, 17.89) Herod commanded his subjects in Judaea to swear an oath of loyalty to Rome in or around 8 B.C.E., "obviously" that's what Luke meant. This is an indefensible thesis. Luke says "census" not "oath." Nor does he describe a situation where Herod is commanding people to take oaths, but of Augustus commanding people to be assessed. Moreover, the only possible rationale for Joseph's travel (Luke 2:3-5) is a tribal census register or the possession of taxable property in Bethlehem subject to a census. In contrast, an oath could be sworn anywhere and did not require traveling, nor is there evidence mass oaths involved precise counting. And, again, Quirinius was still not governing Syria in 8 B.C.E.
Some claim Luke meant an oath Augustus had made to him every five years, but this oath was only made "by the consuls and priests" of Rome, not even all citizens much less any provincials (Res Gestae 9). So this would never have involved Herod or Judaea. Some claim an annual oath was sworn by all the people subject to Rome on the anniversary of the emperor's accession, but even if that were the case, Judaea was not subject to Rome until 6 C.E., and an annual oath could not have involved constantly returning to one's ancestral city every year, or surely such an amazing inconvenience would be mentioned in the histories of the period. Others claim the event during which Jesus was born was when "the people of Rome" proclaimed Augustus "Father of the Nation" in 2 B.C.E., though that again requires reinventing the date of Herod's death, inventing an impossible second governorship for Quirinius, and ignoring what Luke actually says. And after all that, this event was only a vote made by Roman citizens anyway. So this would never have involved Judaeans.17
When everything above is considered, there is simply no way Luke Luke 2:1-6 could have meant or ever been read as referring to any national oath.
There is no reasonable way to get Matthew and Luke to agree with each other on the year Jesus was born. Luke clearly dates his birth in 6 C.E. and Matthew clearly dates it before 4 B.C.E. (possibly as early as 8 or 6 B.C.E.). Everyone concedes Matthew 's narrative cannot be fudged to fit 6 C.E. And all attempts to force Luke to fit Matthew require groundless assertions contrary to all evidence and precedent, and always require declaring that in one way or another Luke didn't mean what he said. Not one of these proposed "solutions" rests on any evidence other than complete fabrications or distortions.
Ample evidence supports the conclusion that Luke meant no other year than 6 C.E., and no clear case can be made that Luke had any other year in mind. There is no way Quirinius could have governed Syria in any earlier year, nor could he have co-ruled Syria or been holding any other office there that Luke would refer to. There is no evidence that Augustus ever did or even would order a census of a Judaean kingdom before its annexation to Syria in 6 C.E. And Luke can neither have meant nor been describing a national "oath." Neither the Lapis Tiburtinus, nor the Lapis Venetus, nor the Antioch Stones even remotely imply any other conclusion. The Vardaman coins are definitively bogus. No reasonable case can be made that Herod the Great was still alive after 4 B.C.E. It's grammatically impossible to read Luke 2:2 as saying "before Quirinius governed Syria." And there is no rationale for assuming a census of Roman citizens in 8 B.C.E. would ever have affected the lives of any Judaean, and no evidence that Herod ever did or even would order a census of his own people, nor is it at all reasonable to interpret Luke as referring to such a thing.
There is no escaping the conclusion. Matthew contradicts Luke on a question of historical fact, and this entails either Matthew or Luke reports something historically false. The Bible is in error.
1 See Burnett's Roman Provincial Coinage (1992), including the supplemental volume with corrections.
3 Marcus Titius from 12 to 9 B.C.E., Sentius Saturninus from 9 to 6 B.C.E., and Quintilius Varus from 6 to 3 B.C.E., each serving a typical three year term, and both historians and inscriptions confirm Quirinius did not achieve consular rank until 12 B.C.E.. Under Roman law and principle, such a rank was a political prerequisite for holding a provincial proconsulship. For previous governors: Josephus, Antiquities of the Jews 16.270-81, 16.344, 17.6-7, 17.24, 17.57, 17.89-133, 17.221-23, 17.250-98; Josephus, Jewish War 1.577, 1.617-39, 2.66-80; Strabo 16.1.28; Velleius 2.117.2; Tacitus, Histories 5.9.2. Coin evidence also exists for their terms, though without precise dates. The consulship of Quirinius is reported in surviving consular lists etched in stone (see the entry for him in Pauly, Wissowa, and Kroll, Realencyclopï¿½die der klassischen Altertumwissenschaft) and by Cassius Dio 54.28.2 For more on the problems of inventing a new governorship for Quirinius see "Was Quirinius Twice Governor?" in my full Secular Web article.
7 For my published reports on the Vardaman debacle, see Richard Carrier, "Pseudohistory in Jerry Vardaman's Magic Coins: The Nonsense of Micrographic Letters," Skeptical Inquirer 26.2 (March-April 2002): pp. 39-41, 61; and Richard Carrier, "More on Vardaman's Microletters," Skeptical Inquirer 26.4 (July-August 2002): pp. 60-61. For summary details, see "Vardaman's Magic 'Coin'" in my full Secular Web article.
9 For more see "Was Quirinius Sharing Command with a Previous Governor?" in my full Secular Web article.
10 Several wholly implausible arguments for rewriting the chronology in Josephus have been offered by Jack Finegan in his Handbook of Biblical Chronology (1998, revised edition). Finegan's errors consist of faulty math, implausible and unsupported conjectures, and relying on incompetent manuscript analysis. These gaffes are barely worthy of attention. Nevertheless, I address them in detail under "Was Herod Alive in 2 B.C.?" in my full Secular Web article.
11 For more on the basic problems of inventing an earlier census see "Was There a Roman Census in Judaea Before Quirinius?" in my full Secular Web article.
Richard Carrier Support
Help Richard Carrier fulfill his destiny as The Quizat Schaderacht of Reason (or just make a token of your appreciation for his scholarship) by sending him a little something from his support page: