A Worldwide Census under Quirinius?


There is an attempt by each writer of the Synoptic Gospels to either re-correct or present their own version of what they perceive to be the “accurate” story of Jesus’ birth, and this has caused disagreements between the gospel records. The discrepancies between the different gospel records were so numerous and glaring that it did not escape the attention of Origen:

Scripture was not even free from factual error. So careful a student of the text as Origen could not help being aware of discrepancies between the different gospel records. The normal procedure of the early Church scholar was to explain these away by elaborate attempts to harmonise the conflicting accounts. That was a game which Origen could play when he wanted to as ingeniously as anyone else. But he did not believe that it could solve the problem entirely. Moreover, the factual truth of some recorded incidents was open to serious doubt on other grounds also, on grounds for example of the intrinsic improbability…Thus the discrepancies between the different records and the historical implausibility of some of the incidents described are such that they might well undermine our whole faith in the trustworthiness of the gospels.1

Luke’s account of the setting of Christ’s birth has often been criticized by those who have recognized the errors made by the author of this gospel. Our focus is on the “worldwide census” attributed to Quirinius, recorded in Luke 2:1-5 as follows:

    Luke 2

    1 And it came to pass in those days, that there went out a decree from Caesar Augustus, that all the world should be taxed.

    2 (And this taxing was first made when Quirinius was governor of Syria.)

    3 And all went to be taxed, every one into his own city.

    4 And Joseph also went up from Galilee, out of the city of Nazareth, into Judaea, unto the city of David, which is called Bethlehem; (because he was of the house and lineage of David:)

    5 To be taxed with Mary his espoused wife, being great with child.

In this instance, the narrative of Luke’s gospel firmly roots the story of Christ’s birth in the context of the worldwide administration of Roman government and demonstrates how God uses unwitting and unwilling men to bring about His purposes. The amazing thing about this narrative is that there was no such census in the days of Herod! Denis McBride, in his commentary on Luke, writes:

Scholars have unresolved questions about the historical accuracy of the events surrounding Luke’s birth narrative: there is no historical record of a universal census at the time specified by Luke; the Roman custom of taxation was based on the individual’s place of residence, not his place of ancestry; Quirinius was governor of Syria during the years A.D. 6-9, some ten years after the birth of Jesus. Ingenius attempts have been made to resolve the historical difficulties, but the important question which can be answered concerns Luke’s intention, which is clear: the census places the birth of Jesus within the framework of world history, it also situates the birth in Joseph’s ancestral city, Bethlehem, the place prophecised for the beginning of the Messiah.2

Despite the above observation, Christian scholars have attempted to “reconcile” this error with historical and documented evidence. Our purpose now is to establish whether these forms of “reconcilations” are acceptable and to further scrutinise the seriousness of this error by Luke, insha’allah.

“Worldwide Census”? What “Worldwide Census”?

Helms comments upon this factual error made by Luke and the bizzare nature of this story:

Though Luke 1:5 dates the birth of Jesus in the “days of Herod, King of Judaea,” who died in 4 B.C., he wants the journey from Galilee to Bethlehem to have occured in response to a census called when “Quirinius was governor of Syria.” As historians know, “the one and only census conducted while Quirinius was legate in Syria affected only Judaea, not Galilee, and took place in A.D. 6-7, a good ten years after the death of Herod the Great.” In his anxiety to relate the Galilean upbringing with the supposed Bethlehem birth, Luke confused his facts. Indeed, Luke’s anxiety has involved him in some real absurdities, like the needless ninety mile journey of a woman in her last days of pregnancy – for it was the Davidic Joseph who supposedly had to be registered in the ancestral village, not the Levitical Mary. Worse yet, Luke has been forced to contrive a universal dislocation for a simple tax registration: who could imagine the efficient Romans requiring millions in the empire to journey scores of hundreds of miles to the villages of millennium-old ancestors merely to sign a tax from! Needless to say, no such event ever happened in the history of the Roman empire, but Micah 5:2 must be fulfilled.3

The historian Robin Lane Fox adds more details exposing the obvious errors in Luke and writes that:

The error, so far, might seem rather marginal. The third Gospel has confused a local census in Judaea with a worldwide decree from Augustus; it has tried to date the story by an obscure Quirinius, whereas elsewhere, like Matthew’s, its story takes place under Herod the Great. In fact, the trouble goes very much deeper. There is a contradiction in Luke’s story: if Quirinius was governor, the Roman census is credible but Herod is a mistake. There is also a contradiction with Matthew’s story: if Quirinius or the Roman census is correct, Herod was not king and Matthew’s stories of the Wise Men, the Massacre of the Innocents and the Flight into Egypt are all chronologically impossible. If Herod was king, there could have been no census according to Caeser Augustus. Even if there had been such a census, the third Gospel’s view of it runs into further problems.

Its decree from Caeser required registration (apographe, in Greek). Exactly this word is used for a tax census in contemporary documents which have survived from Egypt under Roman rule. An emperor would not be imagined to have registered his Jewish subjects for any other purpose. He was certainly not planning conscription: Jews were exempt from military service in the Roman army. Tax, certainly, caused the census, but the practices of Roman taxation do not agree with the Gospel’s narrative. Correctly, it begins by explaining that ‘all went to be taxed, every one into his own city’ (Luke 2:3). Joseph’s ‘own city’ is defined in the Gospel by his supposed ancestry, not by his residence and ownership of property. In the Gospel’s view, Joseph was descended from David, and so he went to Bethlehem, the ‘city of David’, a proper birth-place for a future Messiah. However, Roman censuses cared nothing for remote genealogies, let alone for false ones: they were based on ownership of property by the living, not the dead. As the Gospel has already stated at the time of the Annunciation (Luke 1:26), Joseph and Mary were people of Nazareth in Galilee, the home town which later rejected its prophet, Jesus. A Roman census would not have taken Joseph to Bethlehem where he and Mary owned nothing and were therefore assumed to have needed to lodge as visitors at an inn. There was a sound reason for the Romans’ type of registration. The census was their base for at least two types of tax: a poll tax and a tax on property of various kinds. There was not even a legal need for Mary to go and register with her betrothed husband. We know from the evidence of Roman tax censuses from Egypt, still surviving on papyrus, that one householder could make the return for everyone in his care. Mary might have chosen to go anyway, in order to give Joseph support, but it was not a necessary journey for a wife who was terminally pregnant.

Above all, it was not a journey which a Galilaean, a man of Nazareth, would have been required to make. In AD 6 Galilee, unlike Judaea, had remained under its independent ruler and would not have been bound by a Roman census or taxing. This ruler’s existence is known from Josephus, other histories and his own coins: as a Galilaean, Joseph of Nazareth was exempt from the entire business.4

Christian fundamentalists have, of course, made numerous desperate attempts to “resolve” the glaring historical error in the above referred passage,their numerous efforts have not had much of an impact, this is conceded by the New American Bible:

1 [1-2] Although universal registrations of Roman citizens are attested in 28 B.C., 8 B.C., and A.D. 14 and enrollments in individual provinces of those who are not Roman citizens are also attested, such a universal census of the Roman world under Caesar Augustus is unknown outside the New Testament. Moreover, there are notorious historical problems connected with Luke’s dating the census when Quirinius was governor of Syria, and the various attempts to resolve the difficulties have proved unsuccessful.5

If Luke was being “inspired” by the Holy Ghost, then why did the Holy Ghost allow him to make such a glaring factual error? Was the Holy Ghost not concerned that Luke would write down this factual error and in years to come many people would believe this factual error to be an eternal truth?

The simple fact of the matter is that Luke was neither inspired by God or the Holy Ghost, nor did he ever claim to be inspired. Rather, he states clearly that the purpose of his writing was to compose a more “orderly account” of the past events. He categorically stated that because others had written accounts of Jesus?s life in the past, therefore he will do the same, but with a difference – Luke will attempt to write a more “orderly account” in contrast to the writings of his predecessors.

Moreover, Luke never stated that the other gospel writings were “scripture”. He indicates that he has had several predecessors in writing a narrative of the life of Jesus. He also concedes that his gospel is based upon oral traditions in circulation as the stories have been passed down by “eyewittnesses and ministers of the word.” Furthermore, Luke also made use of written sources. His reason for writing was because he deemed the attempts of his predeccessors to be inadequate and insufficient and thus believed he could do a better job. Donald Guthrie states the following in a footnote within his introduction to the New Testament:

Luke?s preface is illuminating in regard to his own approach to his task. He claims to have made a comprehensive and accurate survey over a considerable period, which throws a good deal of light on his seriousness of purpose. Moreover, Luke admits that others had previously attempted the same task, but his words imply that he found them unsatisfactory…6

These do not sound like the words of one who believes God is inspiring him to write “divine scripture.”

Attempted Explanations With Irreconcilable Fiction

Christian scholars Lee M. McDonald and Porter attempted to explain the historical error in Luke with regard to the census. They cite from the sources of ancient authorities, trying their best to “solve” or to “explain” away Luke?s error, but in vain:

Luke 2:1-2 says that Quirinius was governor of Syria when Jesus was born, but Tacitus (Ann. 6.41) states that Quirinius began as governor only after Archelaus was expelled from office in A.D. 6. Josephus (Ant. 18; War 2.117; 7.253) claims that Quirinius ordered a census for tax purposes in A.D. 6-7 that caused a rebellion. If Tacitus and Josephus are correct, then by Luke’s accounting, the birth of Jesus comes too late and Luke is out of step with other historians of the time. Geldenhuys, however, argued that Quirinius had a dual reign and also a dual enrollment or registration as Luke 2:2 indicates, the first being in the first decade B.C. and the second in the first decade A.D., which is probably also referred to in Acts 5:37 and in Josephus, Ant. 20.97-105. Josephus. however, does not mention that Quirinius served as magistrate in Syria on more than one occasion, and this has some importance since he refers to Quirinius’s rule in Ant. 17.353 and 18.1-4. A marble slab from Tivoli (Tibur) dated sometime after A.D. 14 and currently in the Vatican Museum, mentions an unnamed official who served as legate of Syria on two occasions, but there is nothing in the inscription that specifically ties it to Quirinius. A second inscription, found in Pisidian Antioch in 1912, is dedicated to G. Caristanius Fronto, a colonist of Antioch who served as a prefect for two magistrates. It reads, “prefect of P. Sulpicius Quirinius, chief magistrate (duumvir), prefect of M. Servilius.” A third inscription, though similar, specifies that Fronto is now the prefect of a third magistrate as well. From these inscriptions, however, it is impossible to prove conclusively a dual reign of Quirinius. Publius Sulpicius Ouirinius could have become consul in Syria in 12 B.C. and died in A.D. 21 (see Tacitus, Ann. 3.48; Strabo, Geog. 12.6.5), but there is much conjecture in this. More difficult to explain is the reason for calling a census for the taxation of the people by Augustus when Herod the Great was king and he himself had his own taxes and tax collectors. The client king paid tributes to Rome but was free to collect his own taxes. In examples where Rome levied taxes directly upon the people, significant changes had occurred. For instance, after Herod the Great died and his ruthless son Archelaus was deposed by the emperor in A.D. 6. Judea was no longer a Herodian tetrarchy but a Roman province under the direction of Syria and directly taxed. As a result, there was widespread rebellion in the land, and Judas the Galilean led a revolt against Rome (Josephus. Ant. 18.1-6). Does Luke confuse this time with the time of the birth of Jesus? He certainly uses the taking of a census to indicate why Joseph and Mary arrived in the town of Bethlehem, the city of the promise of a shepherd king.7

Despite their explanations of Luke’s statement with the historical sources, they are forced to admit that:

[t]here is apparently no simple answer available to this difficult issue of correlating Luke’s story of the birth of Jesus with available historical data, and questions remain about Luke’s understanding of the historical context of the birth of Jesus.8

Here we should bear in mind that these scholars, who are certainly no “liberals”, would have loved to declare that Luke did not make an error. Even if there was the remotest “explanation” available, no matter how far-fetched, they would have readily adopted it to save Luke. Yet despite their intense desire to side with Luke, they have no other alternative but to finally concede that there is really no “explanation” that can simply “explain away” the error in Luke. This should give us an indication as to the gravity of the error made by Luke.

Some apologists argue that the practice of going to an ancestral city is credible based upon an appeal to Egyptian censuses. Raymond E. Brown on the other hand has the following to say regarding this claim:

In Roman censuses there is no clear evidence of a practise of going to an ancestral city to be enrolled; the oft-cited examples from Egypt are not the same as what Luke describes…9

He goes on to say

…of the house and lineage of David. Here, as in 1:27, Luke makes reference to Joseph as a Davidid, a lineage affirmed elsewhere in the two genealogies of Jesus and in Matt 1:20. Is there a difference between “house” (oikos) and “lineage” (patria)? Some have taken oikos to mean that Joseph had a home in Bethlehem, or patria to mean that he had property there. Still another sugges tion is that he was returning to his home in Bethlehem (“his own city” of vs. 3) after having gone to Nazareth to claim Mary his bride who lived there. These suggestions run against the reference to Nazareth as “their own city” in 2:39 and against the indication in 2:7 that Joseph had no place to stay in Bethlehem. (Thus we have here no real parallel to the kata Qikian censuses in Egypt where people were registered in the area where their home [oikia] or property was found; Luke refers to a census by ancestry.) It is highly dubious that one can press Davidic lineage to mean direct royal lineage, as Bornhauser, Kindheitsge schichte, 99, has done when he argues that, while all the Davidids did not have to go to Bethlehem, Joseph did because he was a royal scion. There is nothing in Luke’s narrative to suggest that.10

Guignebert writes that:

We will not unduly stress the peculiarity of the mode of census-taking implied by our text, but it is to be noted that it is a very strange proceeding. The moving about of men and families which this reckless decree must have caused throughout the whole of the Empire, is almost beyond imagination, and one cannot help wondering what advantage there could be for the Roman state in this return, for a single day, of so many scattered individuals, not to the places of their birth, but to the original homes of their ancestors. For it is to be remembered that those of royal descent were not the only ones affected by this fantastic ordinance, and many a poor man must have been hard put to it to discover the cradle of his race. The suspicion, or, rather, the conviction, is borne in upon us at first sight that the editor of Luke has simply been looking for some means of bringing Joseph and Mary to Bethlehem, in order to have Jesus born there. A hagiographer of his type never bothers much about common sense in inventing the circumstances he requires. In this case, no notice is taken of the fact that ?the city of David? was not the city of Mary, and that there seems to have been no necessity for her to have made such a journey on the eve of her confinement. It is all outside the plane of reality.11

In other words, the author or the editor of Luke had to invent the story of the census to serve a theological purpose, that being the birth of Jesus in Bethlehem. To such storytellers, historical and factual truth hardly matters since the main purpose is to relate a good story to captivate the readers.

Prof. Bart Ehrman, who is presently one of the leading New Testament textual critic and authority in the world, says the following regarding the census story of Luke:

In addition to the difficulties raised by a detailed comparison of the two birth narratives found in the New Testament, serious historical problems are raised by the familiar stories found in Luke alone. Contrary to what Luke indicates, historians have long known from several ancient inscriptions, the Roman historian Tacitus, and the Jewish historian Josephus that Quirinius was not the governor of Syria until 6 C.E, fully ten years after Herod the Great died. If Jesus was born during the reign of Herod, then Quirinius was not the Syrian governor. We also have no record of a worldwide census under Augustus, or under any emperor at any time. Moreover, a census in which everyone was to return to their ancestral home would have been more than a bureaucratic nightmare; it would have been well nigh impossible. In Luke, Joseph is said to retum to Bethlehem because his ancestor David came from there; but David lived a thousand years before Joseph. Can it be possible that everyone in the empire was to return to the place their ancestors lived a thousand years earlier? If such a census were required in our day, where would you go? Imagine the massive migrations involved. Then imagine that no other ancient author considered it important enough to mention, even in passing!12

This should certainly give us a pause and think over the matter. How could an event on such a scale not be noticed by any historian or writer other than Luke? Did they “forget” about it? This seems to be an unlikely explanation. It is more likely and probable that Luke invented this “historical fact” in order to serve a theological purpose to support his viewpoint.

The late Christian scholar Raymond Brown, who is a recognized, authoritative and world-renowned Christian scholar, states that

Luke begins his story with a reference to a census of the whole world ordered by Augustus, conducted by Quirinius, and affecting Joseph, a Galilean inhabitant of Nazareth, so that he had to go to his ancestral city. This supplied the occasion for the birth of Jesus in Bethlehem about the time of the reign of Herod, king of Judea (1:5). As I point out at length in Appendix VII, this information is dubious on almost every score, despite the elaborate attempts by scholars to defend Lucan accuracy. By way of lesser difficulties we have no evidence of one census under Augustus that covered the whole Empire, nor of a census requirement that people be registered in their ancestral cities. While these difficulties can be explained away, we cannot resolve satisfactorily the major objections, namely, the one and only census conducted while Quirinius was legate in Syria affected only Judea, not Galilee, and took place in A.D. 6-7, a good ten years after the death of Herod the Great. Since elsewhere Luke shows himself inaccurate about the dating of events surrounding this census, the evidence favors the theory that the use of the census to explain the presence of Joseph and Mary at Bethlehem is a Lucan device based on a confused memory. Luke may have had a tradition that associated the birth of Jesus with the end of a Herodian reign (a dating confirmed by Matthew) and a time of political trouble. But Luke seems not to have known that, some eighty years before he wrote, there were two such troubled endings of Herodian reigns, namely, the end of the reign of Herod the Great in 4 B.C., when Jews protested against the giving of Judea to Archelaus, and the end of the reign of Archelaus in A.D. 6, when Jews revolted against the census imposed by Quirinius. Consequently he has given a composite scene as a setting for Jesus’ birth.13

Raymond Brown, who was himself a believeing and practising Christian, took no pleasure in declaring there was a serious factual error in Luke. Whenever it is possible he tries to reconcile minor discrepancies within Luke, however, with all his learning even he was unable to “explain away” the census error within Luke nor declare it completely error free.

Hence the consensus of Bible scholars, which includes practising and believeing Christian scholars, is that Luke is not error free at least on this instance. Despite the numerous “explanations” offerred by some Christian apologists, these have failed to win wide acceptance and approval by mainstream scholarship. Some conservative Christian scholars, such as Bruce Metzger, write that

It should be added, however, that Luke?s statement that the census was conducted while Quirinius was governor of Syria (Luke 2:2) still remains in conflict with what Tacitus and Josephus report concerning the sequence of governors of Syria. In such cases the cautious historian will await acquisition of further information which may resolve the discrepancy.14

It should become obvious to the readers that if the Gospels were not religious literature and if they were not deemed “inspired” but considered the same as any secular literature, then conservative scholars such as Bruce Metzger would not have given a dime or such a benefit of a doubt to them. They realize well that there is a serious problem with Luke on this matter, but because it is their belief that Luke was “inspired”, hence they cannot outright admit and face reality that Luke made an error. They want to “wait” until a time a better explanation, or “evidence”, surfaces so it could be used to vindicate Luke. Other conservative scholars, such as John Drane, argue that because Luke was careful and accurate over other matters, therefore it must follow that he could not have erred over the matter of the census under Quirinius and would have had good reasons to write what he did! It is only the presupposition of Christian missionaries and apologists that the New Testament is God’s Word which precludes them from even seriously considering such an explanation ? that Luke made an error. Take away that presupposition, and it becomes easy to see how such a mistake could have arisen.

Raymond E. Brown, in his Birth of the Messiah studies the problem of the census in detail, exposing the factual error made by Luke together with refuting many excuses cooked up by a few Christian apologists in order to “explain away” the obvious error in Luke. In Appendix VII he writes that

At that time an edict went out from Caesar Augustus that a census should be taken of the whole world. (This was the first census under Quirinius as governor of Syria.) And so all went to be inscribed in the census, each to his own city. Joseph also went up from Galilee, from the city of Nazareth, into Judea to the city of David which is called Bethlehem, because he was of the house and lineage of David, to have himself inscribed in the census. (Luke 2: 1-5). If this notice about the census stood by itself, there would be some problems about the extent and the manner of registration; but the chronology would cause no difficulty. Augustus reigned from 44/42 B.C. to A.D. 14; Publius Sulpicius Quirinius became governor or legate of Syria in A.D. 6 and conducted a census of Judea (not of Galilee) in A.D. 6-7. The last mentioned date would then be implicitly fixed for Luke as the year of the birth of Jesus. But the chronological information in ch. 2 of Luke does not stand by itself; and when we compare it with other chronological information that he has given us in 1:5 and 3:1,23, there seems to be an irreconcilable conflict. In 1:5 Luke tells us that the annunciation of the birth of JBap took place “in the days of Herod, king of Judea.” According to our best information Herod died in March/April 4 B.C. (Note on Matt 2:1). According to Luke 1:36 Mary’s pregnancy began some six months after Elizabeth’s; and so Jesus would have been born about fifteen or sixteen months after the annunciation of JBap’s birth. This would fix the year of his birth at no later than 3 B.C. and would bring Luke into an approximate correspondence with the information in Matt 2 that Jesus was born in the last years of Herod’s reign.1 However, such a date is ten years before Quirinius became governor of Syria and conducted the census. There are three basic approaches in dealing with this conflict. First, one may seek to reinterpret the Herod chronology of Luke 1 to agree with the Quirinius census dating (A.D. 6-7) of Luke 2. Second, one may seek to reinterpret the Quirinius census chronology of Luke 2 to agree with the Herod dating (4-3 B.c.) of Luke 1. Third, one may recognize that one or both of the Lucan datings are confused, and that there is neither a need nor a possibility of reconciling them. Basically this appendix will come to the conclusion that the third approach is the most plausible, but only after a thorough discussion of suggestions made in the first and second approaches.15

Raymond Brown then proceeds to discuss and demolish, one by one, many other”answers” and “explanations” cooked up by apologists:

The first approach may be dealt with briefly, for the chronology of Herod the Great’s reign is too fixed to be changed in order to match a date of A.D. 6-7. One ingenious suggestion, however, is that Luke did not mean Herod the Great but Archelaus, who is occasionally called Herod 2 and who ruled as king of Judea from 4 B.C. to A.D. 6. One could theorize that the annunciation of JBap’s birth took place toward the end of Archelaus’ reign (A.D. 5-6) and that Jesus was born after Archelaus had been deposed and the newly installed Quirinius began the census (A.D. 6-7). Another suggestion recognizes the likelihood mentioned in ? 1OD that, in joining the births of JBap and Jesus, Luke reflects theology rather than history. It is then proposed that the Herod (the Great) dating of 4-3 B.C. is correct for the birth of JBap, while the Quirinius census dating of A.D. 6-7 is correct for the birth of Jesus.3 These and other suggestions which date the birth of Jesus in A.D. 6-7 bring Luke into conflict with Matthew. More importantly they contradict the remaining items of Lucan chronological information mentioned in the first paragraph, namely, 3:1,23. The combination of those two verses pertaining to the beginning of Jesus’ ministry indicates that Jesus was about thirty years old in the fifteenth year of Tiberius (A.D. 27-28). This would agree with a birth in 4-3 B.C. at the time of Herod the Great, but hardly with a birth in A.D. 6-7. 4 This difficulty has caused most scholars who wish to preserve Lucan accuracy to prefer the second approach and to seek to redate the census that Luke mentions to an earlier period, harmonious with the reign of Herod the Great. Much of the rest of this appendix will be devoted to the difficulties encountered in the second approach.16

Let us now proceed to the refutation of the “second approach”:

Did Augustus ever issue an edict that the whole world, i.e., the Roman Empire, be enrolled in a census? Certainly not in the sense in which a modern reader might interpret the Lucan statement! In the reign of Augustus there was no single census covering the Empire; and granted the different legal statuses of provinces and client kingdoms, a sweeping universal edict seems most unlikely. But Luke may not have meant a single census. The long peace under Augustus made empire-wide policies possible, and Augustus was interested in censuses for various purposes. During his reign there were three enumerations of Roman citizens for statistical purposes (28 and 8 B.C.; A.D. 13-14). Taxation and military service were the main goals for census of non-citizens in the provinces; and we know of censuses held at different times in Gaul and in Egypt. Thus, what Luke may be telling us in an oversimplified statement is that the census conducted (in Judea) by Quirinius as governor of Syria was in obedience to Augustus’ policy of getting accurate population statistics for the whole Empire. Would a Roman census have sent people back to their tribal or ancestral homes to be enrolled, as Luke describes in the case of Joseph? We have no clear parallel for such a practice. Since enrollment was primarily for taxation purposes, the general Roman pattern was to register people where they lived or in the nearby principal city of a district (the city from which the tax would be collected). A papyrus (Lond. 904, 20f.) describes a census in Egypt in A.D. 104 wherein a temporary dweller, in order to be enrolled, had to go back to the area of his regular domicile where he had a house. (Sometimes this is referred to as a kata oikian census.) Obviously this ruling was motivated by tax considerations about property and agriculture; and it offers little support for sending Joseph from Nazareth where he permanentIy resided (2:39) to Bethlehem where clearly he had no property or wealth, according to 2:7. Nevertheless, one cannot rule out the possibility that, since the Romans often adapted their administration to local circumstances, a census conducted in Judea would respect the strong attachment of Jews to tribal and ancestral relationships. Even if Luke had little historical information about how the census of Quirinius had been conducted, he lived in the Roman Empire and may have undergone census enrollment himself. It is dangerous to assume that he described a process of registration that would have been patently opposed to everything that he and his readers knew. Yet, if we cannot confirm or deny the pattern of enrollment in ancestral cities which Luke depicts, his narrative seems to presume that the census of Quirinius affected Galileans. This is not factual for the census of A.D. 6-7, since at that time Galilee was not under Quirinius’ direct supervision but was a tetrarchy ruled by Herod Antipas. And so now we must raise directly the question as to whether there was an earlier census by Quirinius before Galilee and Judea were ruled separately.7 Was Quirinius governor (legate) of Syria during or shortly after the reign of Herod the Great, and thus around or before 4 B.C.? Drawing chiefly upon Josephus, we can put together the following chronology for the legates of Syria:

    23-13 B.C.: M. Agrippa
    ca. 10 B.C.: M. Titius
    9-6 B.C.: S. Sentius Saturninus
    6-4 B.C. or later: Quintilius (or Quinctilius) Varus
    1 B.C. to ca. A.D. 4: Gaius Caesar
    A.D. 4-5: L. Volusius Saturninus
    A.D. 4 to after 7: P. Sulpicius Quirinius

If Quirinius served as governor of Syria twice, once in A.D. 6 and once earlier, the two possible time slots for the earlier governorship would be before M. Titius (and thus before 10 B.C.) or between Quintilius Varus and Gaius Caesar (and thus between 4 and 1 B.C.). Either possibility could be reconciled with the Lucan information. However, from what we know of the relatively well-documented career of Quirinius, it is unlikely that he had an earlier governorship at either of those periods. He served as consul in 12 B.C. (Tacitus Annals III 48). He was in Asia Minor sometime after 12 and before 6 B.C. leading the legions in the war against the Homonadenses. He was in the Near East, specifically in Syria, as an advisor of Gaius Caesar for several years before A.D. 4. But there is no mention of Quirinius having been legate in the nearly twenty years of his career from 12 B.C. to A.D. 6. Josephus, who describes several times the beginning of Quirinius’ legateship in A.D. 6, gives no hint that Quirinius had served previously in that capacity.

Two inscriptions have been brought into the discussion to lend support to an earlier governorship by Quirinius. The first is the Lapis Tiburtinus, an inscription on a marble slab found in 1764 in the neighborhood of Tivoli (Tibur) and now in the Vatican Museum.10 This inscription or titulus, composed after A.D. 14, describes an unnamed person who was a major official victorious in war, and who twice served as legate, the second time serving as legate in Syria. The thesis that this is a reference to Quirinius is a pure guess with as many dissenters as adherents.11 The other inscription was found on a marble base in Antioch of Pisidia by W. M. Ramsay in 1912. The inscription is dedicated to G. Caristianus Fronto a colonist of Antioch who served “as prefect of P. Sulpicius Quirinius, the chief magistrate [duumvir], and as prefect of M. Servilius.” Quirinius is identified as a chief magistrate, while Servilius is not; but Ramsay argues that Quirinius and Servilius were of equal status, and indeed Quirinius was legate of Syria at the same time that Servilius was Iegate of Galatia during the Homonadensian war (before 6 B.C.). Obviously Ramsay’s theory goes considerably beyond what the inscription says. His thesis that Quirinius served as legate of Syria during the Homonadensian war is highly dubious: the war took place in the Taurus mountains, and for military operations there Galatia was a far more practical base than Syria. Thus neither inscription lessens the difficulty of proving that Quirinius had an earlier governorship in Syria during which he might have conducted a census. Indeed, even if without evidence one does posit this earlier governorship of Quirinius, how does one explain a Roman census of Palestine during the reign of Herod the Great? A client king who paid tribute to Rome, Herod had his own taxes and tax collectors; and there is no evidence that the Romans collected taxes based on a census within his realm…The known census of Quirinius in A.D. 6-7 was conducted precisely because Herod’s son Archelaus had been deposed, and Judea was now coming under direct Roman government and taxation. But let me bring to the reader’s attention four contrary indications that have been offered as support for the thesis that a Roman census was possible in Herod the Great’s realm. First, Augustus became very displeased with Herod in 8 B.C. and wrote, threatening to treat Herod no longer as a friend but as a subject…It has been suggested that this might have led to the imposition of Roman taxes and a Roman census. But, in point of fact, the evidence is that Augustus did not make good on his threat and withdraw his friendship.13 Seconod, Josephus…mentions an oath of allegiance to the emperor taken by the Jewish people ca. 7 B.C., under Herod the Great’s direction. But the claim that this involved an acceptance of direct Roman taxation and a Roman census is gratuitous. A variant of this suggestion has been proposed by Barnett, “Apographe”, who contends that Luke is not referring to a census for the purpose of taxation, but to a registration or enrollment (NOTE on “census” in 2:1) that Herod required for the purpose of the oath-taking. Of course, there is no evidence of such a registration, and Barnett has to suppose that Luke 2:2 should be translated: “This was an enrollment conducted before Quirinius was governor of Syria” – a translation that makes us wonder why Luke would have mentioned Quirinius at all. Third, Josephus twice makes reference to the fact that Herod the Great remitted to the people part of their taxes, and from that it is deduced that there must have been census records. In fact, however, it proves only the existence of records, not of a census, much less of a Roman census. Fourth, an instance of Roman taxation of a client kingdom has been pointed to in the narrative of Tacitus, Annals V1 41, where ca. A.D. 36 Prince Archelaus the Younger conducted a census according to the Roman custom among the Cietae who were his subjects in the Cilician section of Cappadocia. But the passage does not say that the Romans imposed this census on Archelaus or his subjects. Moreover, the imposition of the taxes was bitterly resented by the Cietae and led to a revolt. The imposition of a Roman census and Roman tax in the realm of Herod the Great should have produced exactly the same result; so the silence of Josephus, not only about such a census and tax under Herod, but also about a revolt or protest against Rome over such a practice, is an eloquent argument that there was no Roman census of Palestine before the census under Quirinius in A.D. 6-7 – an event, with its concomitant revolt, carefully described by Josephus. Let us consider for a moment Josephus’ description of the Quirinius census. Judea was reduced from the position of an Herodian tetrarchy and became a Roman province annexed to Syria in A.D. 6 (759 A.U.C.); and Josephus…reports that Quirinius, who had been dispatched by Augustus to become legate of Syria, visited Judea “in order to make an assessment of the property of the Jews and to liquidate the estate of Archelaus.” At first the Jews were shocked to hear of the registration of property, but most of them yielded at the intercession of the high priest. However, Judas the Galilean led a rebellion, becoming the founder of the nationalistic Zealot movement…Certainly the Josephus narrative would not lead us to suspect that the Jews were previously accustomed to Roman census and taxation! In speaking of “the first census under Quirinius”, surely Luke is referring to this well known (and even notorious) census of A.D. 6-7,16 a census to which he again refers in Acts 5:37.17

We note how Raymond Brown attempts to explain away and reconcile some minor problems within the Lucan narrative, but that he is unable to be as receptive as he would like to be towards the alleged census under Quirinius since the available data stands squarely against it. Raymond Brown then proceeds to make mention of the “final efforts” to save the account in Luke from discrepancy:

Before I pass a general judgment on the historicity of Luke 2:1-5, let me mention some final efforts to save Lucan accuracy. It has been proposed that the present reading of Luke 2:2 is a mistake or textual corruption and that we should read “under Saturninus as governor of Syria.” (I did not mention this as a textual variant in the NOTE on 2:2, for there is no manuscript evidence that supports it.) Since Saturninus 17 was governor in the period 9-6 B.C., this would at least synchronize the census with the reign of Herod the Great, although it still labors under the lack of evidence for a Roman census during that reign. The support for identifying Saturninus as the census taker is a passage in Tertullian, Adversus Marcion IV xix 10: “At that time there were censuses that had been taken in Judea under Augustus by Sentius Saturninus, in which they might have enquired about Jesus’ ancestry.” The fact that ‘Tertullian speaks of censuses (plural) is often overlooked, 18 and it is assumed that he was referring to the census in Luke 2:1-5. Some have thought that he had a superior text of 2:2 which read “Saturninus”; others have thought that he was correcting the “Quirinius” reading in Luke after consulting Roman census records. However, there is no real evidence that Tertullian had the Lucan census in mind. The passage occurs as part of an argument against the Docetists; and Tertullian wants to show that Luke 8:19-21 (“My mother and my brother are those who hear the word of God and do it”) does not mean Jesus denied having any human ancestry. Accordingly, Tertullian points to the census records as a possible way of proving such ancestry. Probably he is assuming that such records existed for Palestine, even as they existed in Tertullian’s own time. This is not just a guess about Tertullian’s reasoning, for we have evidence of his assumptions in relation to the death of Jesus. In the Apologeticum, v 2 and xxi 24, he tells us that Pilate must have reported the facts about the crucifixion to Tiberius who communicated them to the Senate, where there were debates about the divinity of Christ, in which Tiberius favored Christ! Another ingenious attempt to rescue Luke is to propose that the census he mentions was a two-step process-a census begun under Saturninus or Varus (thus close to Herod’s reign), but completed under Quirinius. However, Luke does not say that the census was completed under Quirinius, but that it took place (egeneto) under Quirinius. Stauffer 20 has recently revived a form of the two-step thesis, by distinguishing two terms: (a) the apographe (literally, “registration”) of taxable property and persons, entailing an appearance at the registry office; and (b) the apotimesis (literally, “evaluation”) or actual tax assessment on the basis of the registration. According to Stauffer, Luke is describing the apographe (using that word), which was the first step in the Palestinian census and took place under Saturninus ca. 7 B.C., when Herod was still alive. It was then that Jesus was born. Josephus describes the apotimesis or second step in the census which took place under Quirinius in A.D. 6-7. A fourteen-year process in a census is not unusual, Stauffer argues, since the census in Gaul took forty years. But, besides the difficulty of proving even a Roman apographe in Herod’s reign, Stauffer’s thesis faces major terminological obstacles. Luke speaks of an apographe under Quirinius, not of an apotimesis under Quirinius; and Josephus uses both terms to describe the census of Quirinius, so that neither ancient author supports Stauffer’s distinction. Another form of the two-step process is the proposal that there was a Jewish census conducted by the priests in Herod’s time, and that Luke has confused this with the Roman census ten to fifteen years later. But this is tantamount to a recognition that Luke was inaccurate.18

The conservative Christian scholar Donald Guthrie also refers to Stauffer?s defense in his introduction to the Gospels and Acts, conceding in a footnote that

Unfortunately, Stauffer cites no specific supporting evidence, although if his theory is true it would remove all the difficulties.19

Finally, the undeniable conclusion is that:

When all is evaluated, the weight of the evidence is strongly against the possibility of reconciling the information in Luke 1 and Luke 2. There is no serious reason to believe that there was a Roman census of Palestine under Quirinius during the reign of Herod the Great. (Indeed, as regards the non-biblical “evidence,” it is doubtful that anyone would have even thought about an earlier census if he were not trying to defend Lucan accuracy.) The information in Luke 1 may be correct: Jesus may indeed have been born during or at the end of the reign of Herod the Great. But Luke seems to be inaccurate in associating that birth with the one and only census of Judea (not of Galilee) conducted in A.D. 6-7 under Quirinius….20

Defenders of Luke adopt their position because they believe that Luke was “inspired” by the Holy Ghost and therefore could not have erred over any matter imaginable. Therefore they cannot bring themselves to accept the fact that at least on one occasion Luke was seriously wrong and erroneous. It is only the presupposition of Christian missionaries and apologists that the New Testament is God’s Word which precludes them from even seriously considering such an explanation. Take away that presupposition, and it becomes easy to see how such a mistake could have arisen.

Luke & the Association of Fiction

Now that it is clear that a “worldwide census” under Quirinius had never existed, we would like to know why the author of Luke had invented this fiction? The answer is because this association enables the explanation of why Joseph and Mary were in the city of Bethlehem when Jesus was born.

…even if Luke was inaccurate on this dating of the census of Quirinius and mistakenly thought that it could have been associated with the birth of Jesus, we must recognize that the association enabled Luke to explain why Joseph and Mary were in Bethlehem when the child was born. It also served admirably the interests of Lucan theology, giving the nativity a backdrop of world and Israelite history.21

This is further elucidated as follows:

It is easy to sympathize with the third Gospeller’s plight. He had not been present at the earliest days of Jesus’s mission, as he admits in his opening sentence. He was writing from what he had heard perhaps thirty years or so after Jesus’s death, although many scholars would date his books even later. He knew that Joseph and Mary were people of Nazareth, but there were Christians who said that Jesus had been born in Bethlehem, according to the scripture. Why would a man from Nazareth be visiting Bethlehem with his heavily pregnant wife? Somebody, perhaps the Gospeller himself, assumed that the cause was that universal culprit, personal tax. The census of Quirinius was a landmark in Jewish history, and so the Gospel attached the birth in Bethlehem to this well-known fact. The idea had its advantages: the Christian story could begin with Joseph and Mary meekly obeying the orders of Roman government. This origin showed the true nature of a religion which people in the Roman Empire had since misunderstood as a revolutionary movement. The decree, the Gospeller assumed, was not just the edict of a local governor: it was the worldwide decree of the Caeser himself, an exaggeration which was not out of keeping with his misuse of words like ‘all’ or ‘everywhere’ at other points in his books. He was writing for a highly placed Gentile, Theophilus, in the Roman world. It was good to begin with connections between his story and Roman government; a higher truth was served by an impossible fiction.22

In other words, the unknown authors of the Gospels were in the bad habit of creating fiction solely for the interest of sustaining their unique interpretations of Jesus’ birth. Why then should we trust anything within the Gospels when we know that their authors had freely invented and transposed “facts”, “figures” and stories for their personal reasons merely to “support” their respective theology?

Luke had invented the fiction of the census of Quirinius in order to “explain” why Joseph and Mary were in Bethlehem. Can we trust such writers who concoct and invent stories and facts in such a manner? How do we distinguish between the actual historical facts and pure fictions within the Gospels since they contain a mixture of both? Certainly they cannot be trusted blindly or with too much confidence knowing the fact their authors were not quite concerned with historical accuracy, their goals were to mold and invent stories for theological reasons.

That the gospels are a product of fact and fictions is admitted by many conservative scholars as well, such as the Evangelical scholar John Drane, who writes that

…they [New Testament gospels] are certainly carefully crafted narratives aiming to tell the story of Jesus’ life and teaching. As such, they are not to be judged by the standards of scientific enquiry, but according to the practises of story telling, in which the ‘truth’ of a narrative is to be judged as a whole on its own terms, rather than in relation to notions of truth and falsehood drawn from some other sphere of human endeavour. The early Christian communities clearly had no problem in accepting that within the gospel traditions there would be a subtile combination of factual and fictional elements. Had they not done so, they would certainly not have tolerated the existence of four gospels which, for all their similarities, are sufficiently different from one another as to defy all attempts at producing one harmonized, factual version of the life and teachings of Jesus from them. They knew that both artists and historians operate under similar constraints as they seek to balance fact with fictional elaboration, and that the telling of a good story…depends on the coherent combination of both these elements. While all four gospels contain factual fictive elements, the fourth gospel appears to have a greater prepondrance of the latter.23

The conservative Christian scholar, Rev. Arthur W. Wainwright, on the other hand wrote:

It is certainly obvious that Matthew and Luke have sometimes modified what they found in Mark, and Mark himself may have modified some of the traditions which he received. Nevertheless, there is a strong basis of truth in the Synoptic Gospels, although an element of human error may have entered into the transmission of the sayings and narratives.24

The Anchor Bible Dictionary also discusses a number of solutions proposed by apologists to explain Luke’s error, duly disposing them all as unconvincing:

Luke wrote (2:1-2), “It happened in those days that a decree went out from Caesar Augustus that the whole world be registered for a tax, the decree first went out while Cyrenius was governor of Syria.” Cyrenius is the Greek form of the name Quirinius (PW 4: 823), and the fact that he held a census during his term in office is confirmed by an inscription from Apamea (modern Aleppo) in Syria as well as by Josephus. Although it was already recognized in antiquity that there was a problem with Luke’s chronology – Tertullian knew that Sentius Saturninus was governor of Syria at the end of Herod’s life (Adv. Marc. 4.19)-a number of efforts have nevertheless been made by modern scholars to reconcile this statement with Luke’s belief that Herod was still alive when the census was held. The most forceful argument has been that Quirinius must have been governor twice and that he must have held two censuses. It is based upon the contention that Quirinius is the officer mentioned on an inscription from Tibur which contains part of the career of a senator in the reign of Augustus.

This inscription does not, however, preserve the name of the man it commemorates, and, indeed, it preserves only part of the career. The surviving section runs, “. . . king, which he brought under the control of Caesar Augustus and the Roman people, the senate decreed two days of thanksgiving to the immortal gods because of the deeds which he had successfully accomplished; as proconsul he obtained the province of Asia and, serving again as a legate of the divine Augustus with pro-praetorian power he obtained the province of Syria and Phoenicia . . .” (ILS 918). The Latin text of the phrase, “. . . serving again (iterum) as legate . . . ,” is legatus pr. pr. divi Augusti iterum Syriam et Phoenicen optinuit. Many scholars have wanted to translate this as “serving as legate of the divine Augustus with pro-praetorian power he obtained the province of Syria and Phoenicia again (iterum),” taking iterum with optinuit rather than with the words which proceeded it. This has enabled them to claim that this official governed Syria twice, and that he thus must be Quirinius. Aside from an obvious circularity, the most serious objection to this argument is that this is not the proper way to read the Latin. In normal Latin iterum is understood with the words that precede it. There is no reason to translate it any other way here, and the phrase should be taken as a reference to the fact that the man in question had held more than one province as a legate of Augustus (Syme 1973: 592-93).

A great number of other arguments have been adduced at one time or another to reconcile Luke’s narrative with the facts of Roman history. All of them fail to answer four other basic objections to the historicity of Luke’s statement (HJP? 399-427). These are:

1. There is no other evidence for an empire-wide census in the reign of Augustus.

2. In a Roman census Joseph would not have been required to travel to Bethlehem, and he would not have been required to bring Mary with him.

3. A Roman census could not have been carried out in Herod’s kingdom while Herod was alive.

4. Josephus refers to the census of Quirinius in a.d. 6/7 as something that was without precedent in the region.25

The Anchor Bible has no choice but to conclude that:

[i]n the face of these objections, it is impossible to defend Luke’s dating of the Nativity. The easiest explanation for his error is that he wished to provide a synchronism between the birth of Christ and a famous event and so picked upon the census of Quirinius, which caused a great stir throughout the region, as Josephus makes plain.26


According to Matthew, Jesus should be born (between 6 and 4 B.C.) during the reign of the great Herod, who died in 4 B.C.; the census which according to Luke, obliged the parents to travel to Bethlehem, was executed by Quirinius, who according to the precise indications of Flavius Josephus, the Jewish historian, became procurator of Syria in 7 A.D., i.e. 11 – 12 years later. Hence:

The scale of the Gospel’s error is now clear. The first census did occur under Quirinius, but it belonged in AD 6 when Herod the Great was long dead; it was a local census in Roman Judaea and there was no decree from Carsar Augustus to all the world; in AD 6 Joseph of Nazareth would not have registered at Bethlehem: as a Galilaean he was not under direct Roman rule and was exempt from Judaea’s registration; his wife had no legal need to leave home. Luke’s story is historically impossible and internally incoherent. It clashes with his own date for the Annunciation (which he places under Herod) and with Matthew’s long story of the Nativity which also presupposes Herod the Great as king. It is, therefore, false.27

The above conclusion of this historical discrepancy within the Gospel of Luke is not entirely inconsistent with the overall picture of Luke as a historian, namely that

Luke’s intention “to write an orderly account” (1:3) does not imply that he gives us exact history or chronology…Thus, if one wishes to use the statements in the Lucan Prologue to make prejudgments about the amount of historical precision one can expect in the infancy narrative, one must first interpret the Prologue in the light of Luke’s procedure in the body of Luke/Acts – a procedure that gives evidence of considerable freedom of composition, occasional historical inexactitudes, and a primary interest in the logical rather than the chronological.28

With the evidence we have shown above, it is clear that where the “worldwide census” of Quirinius is concerned, it is a historical error that inevitably deals a heavy blow to the case of the Bible “inerrancy” and its claim to be the ‘Word of God’. No doubt that the missionaries would have to grapple with this problem and attempt at a “repackage” in order to sell their “inerrant” Book to the Muslim masses. Fortunately, it will take an insurmountable effort to convince the educated Muslim of their propaganda, and an education is something that we all know the missionaries clearly lack!

And only God knows best. bismika-tombstone A Worldwide Census under Quirinius?


  1. M. F. Wiles, Chapter 14: Origen As Biblical Scholar, in P. R Ackroyd and C. F. Evans (eds), The Cambridge History of the Bible: From the Beginnings to Jerome, Vol. 1 (Cambridge University Press, 1970), p. 463 []
  2. Denis McBride, The Gospel of Luke: A Reflective Commentary (Dominican Publications Dublin, 1982), pp. 36-37 []
  3. Randel Helms, Gospel Fictions, pp. 59-60 []
  4. Robin Lane Fox, The Unauthorized Version: Truth and Fiction In The Bible (Penguin Books Ltd, 1991), pp. 30-31 []
  5. Luke 2, New American Bible, United States Conference of Catholic Bishops [Online Document] []
  6. Donald Guthrie, B.D., M. Th., New Testament Introduction. The Gospels and Acts (Inter-Varsity Press, 1966) p. 87 []
  7. Lee Martin Mc Donald and Stanley E. Porter, Early Christianity and Its Sacred Literature(Hendrickson Publishers, Inc., 2000), pp. 120-121 []
  8. ibid. []
  9. Raymond E. Brown, The Birth Of The Messiah (Cassell & Collier Macmillan Publishers Ltd., 1977), p. 396] []
  10. ibid. []
  11. Charles Guignebert, Jesus, ET by S. H. Hooke, (University Books, NY 1956), p. 99 []
  12. Bart D. Ehrman, The New Testament: A Historical Introduction to the Early Christian Writings (Oxford University Press, 2000), p. 109 []
  13. Raymond E. Brown, Op. Cit., pp. 412-413 []
  14. Bruce M. Metzger, The New Testament: Its Background, Growth and Content (Abingdon Press Nashville, 1985) p. 107 []
  15. Raymond E. Brown, Op. Cit., pp. 547-548 []
  16. ibid., pp. 548 []
  17. ibid., pp. 548-553 []
  18. ibid., pp. 553-554 []
  19. Donald Guthrie, op. cit., p. 167 []
  20. Raymond E. Brown, op. cit., p. 554 []
  21. ibid., pp. 554-555 []
  22. Robin Lane Fox, op. cit., p. 32 []
  23. John Drane, Introducing the New Testament (Lion Publishing Plc, 1999) pp. 210-211 []
  24. Rev. Arthur W. Wainwright, A Guide to the New Testament (The FPWorth Press, 1965), p. 30 []
  25. Freedman, David Noel, ed., The Anchor Bible Dictionary (New York:Doubleday) 1997, 1992 []
  26. ibid. []
  27. Robin Lane Fox, op. cit., pp. 30-31 []
  28. Raymond E. Brown, op. cit., p. 239 []

1 Comment

  1. You argued your case with such admirable skepticism that I thought you were an atheist: it wasn’t until the very end that I, a non-believer, realised that you were in fact a Muslim – it’s to your credit.

Leave a Reply

Your email address will not be published. Required fields are marked *