The Infancy Narratives in Matthew and Luke

by James Kiefer



It is often said that the Gospels (or more generally, all the books of the New Testament) were written long after the Crucifixion, and that therefore they cannot be historically reliable. I have written elsewhere on the dating of the Gospels, but it strikes me that a more direct approach is in order. For the most part, we are interested in the date at which a Gospel was written primarily as an aid in estimating the historical reliability of the narrative. I therefore propose to discuss directly the question: Where we can check them, how reliable are the Gospels as historical accounts? Ordinarily, in the biography of a public figure, one expects that the account of his adult life, after he became well known, will stick reasonably close to the facts, since there will be many people around who have observed the facts, but that the account of his early years will be somewhat embellished. Thus, when reading a biography of an American President, we tend to take with a grain of salt stories about George Washington and the cherry tree, or Abraham Lincoln as a youth walking ten miles to pay six cents to a customer he had accidentally overcharged, or twenty miles to borrow a law book, while at the same time expecting the book to be quite accurate about things like the date of the president's inauguration, his wife's name, the name of his vice-president, etc. Similarly, if we were to discover that the Infancy Narratives in the Gospels are quite inaccurate, we would not necessarily conclude that the entire account is worthless. However, if we find Matthew and Luke accurate in their accounts of the birth and early years of Jesus, accurate precisely where most biographies are least trustworthy, then we shall have good grounds for expecting accuracy in their accounts of Jesus' public career. For this reason, I shall be concentrating on the Infancy Narratives in the first two chapters of Matthew and Luke.



Let us consider the opening of the second chapter of Matthew:

+ Now when Jesus was born in Bethlehem of Judea in the days of + Herod the king, there came Magi from the East to Jerusalem + saying, "Where is he that is born King of the Jews? for we have + seen his star in the rising and have come to pay him homage."

We might begin testing the factual accuracy of Matthew's account by trying to match his "star in the rising" with some astronomical phenomenon of the time. The time in question is, of course, the last few years of the reign of King Herod. Herod probably died in March of 4 BC, though a few scholars have held out for 1 BC. (That Jesus should be born in 4 BC or earlier seems odd, but the practice of dating events from the birth of Christ was established more than five hundred years later, and the calendar reformers were a few years off.) Now, assuming that Herod died in 4 BC, do we have any record, aside from that in Matthew, of some kind of heavenly appearance that could be the star that the Magi spoke of? One possibility would be a comet or a nova, an exploding star. Halley's Comet, which appears every seventy-six years or thereabouts, was visible in 12 BC, but this is far too early. Some commentators have been so bold as to say that there were no comets or novae between 12 and 4 BC, and that we know that there weren't, because the Chinese kept very complete records of such things, and did not record any comets or novae for those years. Negative statements like that are always dangerous. In fact, a Mr. John Williams, in a catalogue published in London in 1871, lists as entries 52 and 53 two comets sighted by Chinese and Korean astronomers in 5 BC and early 4 BC respectively. What I take to be the same two sightings are discussed in an article in the December, 1977 Bulletin of the Royal Astronomical Society of Great Britain. According to the Bulletin, the date for the second sighting is a calendric impossibility (something analogous to February 29, 1967), an obvious slip of the pen, and may conceivably be the same as the first sighting. If they are two separate sightings, then this would agree very nicely with the fact that Matthew says that the Magi saw the star before they started their journey and again while they were on the road from Jerusalem to Bethlehem, but seems to say that they did not see it all the time in between.


Since novae and most comets are non-repeating phenomena, we must rely on ancient records for information about them. The motions of the planets are another matter. These can be calculated with great exactness for thousands of years into the past or future. The next hypothesis that we shall consider is accordingly far more detailed. It is that the star of Bethlehem was the triple conjunction of Jupiter and Saturn in the constellation Pisces in the year 7 BC. We first explain what a triple conjunction is, and then explain why this would have excited the wise men, and what they would have thought it meant. There is in the sky a circular belt of stars called the Zodiac completely surrounding the earth, and consisting of twelve constellations. The planets, as seen from the earth, appear to move around the Zodiac. Jupiter makes a complete circuit of the Zodiac about once every 12 years (that is, it moves about 30 degrees per year. Saturn makes a complete curcuit of the Zodiac about once every 30 years -- that is, it moves about 12 degrees a year. Thus every year Jupiter gains about 18 degrees on Saturn, and about every 20 years (actually closer to 19.87 years) it laps the track and passes Saturn from behind. When they are neck and neck, we say that they are in conjunction. A minority of these conjunctions (about one in six) are triple conjunctions, meaning that Jupiter passes Saturn, then falls behind, then passes again. This is rare enough to attract attention. In 7 BC, Jupiter and Saturn had a triple conjunction in the constellation Pisces, and it did attract attention. In fact, astronomers were looking forward to it at least ten years before it occurred. A cuneiform tablet (The Almanac of Sippur), written in 17 BC, gives the motions of the planets for the next few years, and clearly regards this triple conjunction as the major astronomical event in the immediate future. The times for all the events of the conjunction (the times when the two planets began to move backwards, the times when they resumed their forward motions, the three times at which they had the same longitude) have been calculated and recorded in loving detail. In 1981, Jupiter and Saturn had another triple conjunction (not in Pisces), the one hundredth conjunction and the fourteenth triple conjunction since that of 7 BC. The triple conjunction of 1464 is, as far as I can learn, the only triple conjunction in Pisces since 7 BC. Using round numbers, we may say that we have a conjunction of Jupiter and Saturn every 20 years, a triple conjunction on the average about once every 120 years, and a triple conjunction in Pisces on the average about once every 1440 years.


Matthew describes the wise men as seeing a star and deducing that it heralded the birth of a great king. Whatever else the wise men were, they were astrologers -- that is, they believed that the motions and appearances of the heavenly bodies foretold earthly events. Now a simple conjunction of Jupiter and Saturn, such as happened every twenty years, would not seem especially significant, but a triple conjunction would receive more attention. What would astrologers of that time take it to mean?

Jupiter was regarded as the royal planet, the sign of kings. Therefore the conjunction would have been taken to indicate a royal event, such as the birth of a king.

For Pisces, two interpretations are available. First, there was a system of assigning the houses of the Zodiac to parts of the Mediterranean world, with Pisces assigned to Syria and Palestine. Second, because of a slow wobbling of the earth on its axis, like a spinning top, the equinoctial point (the point in the Zodiac which is occupied by the sun on the first day of spring, when the days and nights are equally long) moves slowly through the Zodiac, spending about two thousand years in each house. It is now more or less in the process of moving from Pisces into Aquarius. Therefore, according to astrologers, the Age of Pisces is ending, and this is the dawning of the Age of Aquarius. I have been able to learn very little about this age, except that it is about to be a very good age, full of peace, prosperity, and universal happiness, making it quite a contrast with the preceding age, which has been somewhat of a disappointment. Two thousand years ago, I suspect that astrologers were looking forward with great eagerness to the dawning of the Age of Pisces, and assuring their listeners that it would be a golden age, full of peace, prosperity, and universal happiness, a contrast with the preceding ages, which had been somewhat of a disappointment. Pisces, therefore, would be a sign of the Golden Age to come.

Curiously, the same two interpretations are also available for Saturn. First, there was a system of assigning planets to countries or peoples which made Saturn the special protector of the Jewish people. The Roman historian Tacitus, without explanation, refers to Saturn as the god of the Jews. Jewish writers identified the planet Saturn with the archangel Michael, who was considered to be the guardian angel of the Jewish nation. Second, there was the belief that the god Saturn had ruled the earth in a past Golden Age, and would do so again some day. One day, or so people thought, Saturn would return and would abolish toil and slavery, poverty, disease, and every other ill. Men would live in harmony with one another, and even the wild beasts would no longer be dangerous. In celebration of that hope, the Romans celebrated every December a week-long feast, the Saturnalia, during which slaves reclined at table and were waited on by their masters. The Roman poet Vergil refers to that hope in his Fourth Eclogue (written about 40 BC), where he says:

Now there has come the last age of which the Cumaean Sybil sang; a great orderly line of centuries begins anew; now too the Virgin returns; the reign of Saturn returns; a new human generation descends from the high heavens. Upon the child now to be born, under whom the race of iron will cease and a golden race will spring up over the whole world, .... the herds will not be afraid of the mighty lions.... the serpent will perish, as will the deceptive poison herb, while the aromatic Assyrian shrub will spring up in every field. ... Every land will be fruitful; yet the earth will not feel the rake, nor the vine feel the pruning hook.

Thus, in either of two ways, the wise men could interpret the triple conjunction of Jupiter and Saturn in Pisces to mean: The king who shall reign over the Golden Age to come is born in Palestine. And so they set out for Jerusalem to find the new king and to pay him homage.


We have seen that, given the common interpretations of Jupiter, Saturn, and Pisces, the above interpretation of the conjunction would be a natural one for the wise men to make. But is there any evidence that anyone ever made it? Jewish commentaries from about the time of Christ speak of the Messiah as heralded by a star, without referring specifically to Jupiter, Saturn and Pisces. But some later references are more definite. The Jewish writers Abraham bar Hiyya (about 1100 AD) and Isaac Abravanel (1437-1508) both believed that the triple conjunction of Jupiter and Saturn in Pisces in 1464 heralded the coming of the Messiah. (They believed that just such a conjunction had occurred in 1395 BC (by my reckoning that should be 1398 BC), three years before the birth of Moses.) Moses Maimonides (1135-1204) rejects this theory, but says that it is widespread among Jews. The writings of Abravanel (or Abarbanel) attracted the attention of the Christian astronomer Kepler, who calculated that just such a conjunction had taken place not long before the birth of Jesus in the last years of Herod's reign. Now we know that many Jews in the first century were interested in astrology (although others denounced it as a heathen superstition), but we do not know how far back the belief in this conjunction as a sign of the Messiah goes. However, it seems unlikely that the belief would arise in Jewish non-Christian circles as a result of the conjunction of 7 BC and its close connection with the birth of Jesus, so we must suppose either that it arose afterwards, independently, and is just a curious coincidence, or that it already existed at the time, and this latter seems at least probable.


We take it, then, that a group of astrologers living somewhere east of Jerusalem saw the conjunction and decided to act on it. They were probably from Babylon, where astronomical observation was expert and astrological belief was nearly universal, and which was the home of many Jews and a center of Jewish learning. Perhaps they were themselves Jews. If not, they were familiar with Jewish beliefs and the expectation of a Messiah. Perhaps they were encouraged by the fact that the second of the three conjunctions (May 29, October 3, December 4) took place on Yom Kippur, the Day of Atonement, the holiest day of the Jewish year. Note that a constellation rises about four minutes earlier each day than it did the day before, thus gaining two hours on the sun each month, and twenty-four hours each year. The sun would have been in the constellation Pisces throughout most of March, but by May 29 Pisces and with it the two planets Jupiter and Saturn would have been rising long enough before the sun so that they would be visible for an hour or two before dawn. This may be what the Magi meant when they said, "We have seen his star in the rising." (Not, as the King James translation has it, "in the east," which would be "in risings" -- plural and no definite article) On the other hand, they may have used "in the rising" as a technical term meaning the first of the three phases of a triple conjunction. Either way, they seem to be referring to the May 29 conjunction. It has been disputed whether they would have called a conjunction a "star," but the usage is not without precedent in Greek, and considering that Greek was probably not the native language of the wise men, this objection must not be given too much weight. They headed for Jerusalem and made inquiries. Someone reported them to Herod and they were brought before him. He told them, "Go to Bethlehem and bring me word so that I may come and pay him homage also." This remark has been ridiculed on the grounds that Herod was too clever and too suspicious to have depended on the Magi. Surely he would have had them followed. But how do we know that he did not? Perhaps the Magi, because they were clever, or because they were lucky, or because God was with them, managed to give Herod's spies the slip and come to Bethlehem untailed. They may not have told anyone the details, and indeed may not have known anything about it themselves. There were many Jews who hated Herod and who were hoping for the Messiah. There was an active guerrilla movement, called the Zealots, and it surely had spies at Herod's court. Perhaps when the Magi left Jerusalem, followed at a discreet distance by a few of Herod's spies, a small band of horsemen rode out of the hills, slit the spies' throats, and rode away without the Magi's ever knowing about it. We are told that as the Magi travelled from Jerusalem to Bethlehem, "The star went before them until it stopped over the place where the child was." This has caused some people to say that the star could not have been anything like a planet or comet or real star, and that what we are being asked to believe in is something much more like ball lightning or will-o'-the-wisp: some kind of luminous object flitting merrily along in front of the Magi not very far off the ground. This is not the only possible interpretation. Bethlehem is about five miles south of Jerusalem. A reasonable travel time, by camel or on foot, say at three miles an hour, would be about an hour and forty minutes. Assuming that we are dealing with the second conjunction, the October one, the planets, which at the first conjunction in late May were rising a few hours before sunrise, would now be rising a few hours before sunset. There would have been about an hour and forty minutes between the time when the planets became visible after sunset and the time when they reached their meridian (i.e. when they were due south of the observers and had reached their maximum height above the horizon). If the Magi left Jerusalem when they saw the star in the southern sky, the fact that the road to Bethlehem led them straight toward the star, give or take a few degrees, and that when they reached Bethlehem the star was at its greatest height (which is probably what is meant by saying that it halted) would have struck them as very definitely a good omen, worthy of comment. I take it that "halted over the place where the child was" refers to Bethlehem. There is no need to suppose that the star picked out the very house for them. Bethlehem was not a large town, and a few discreet inquiries (they had learned something in Jerusalem about the need for discretion) probably got them the addresses of the two or three women who had just given birth for the first time (a younger brother is not the heir apparent to a throne). The Magi, being warned in a dream, did not report back to Herod, but returned home by a different way (or "by different ways"). They may have been influenced by the fact that Mars was due to join Jupiter and Saturn in February, and Mars is the planet of war and bloodshed. This, plus what they had learned of Herod, by meeting him and in other ways, may have made them suspect trouble ahead. Joseph, being warned of God in a dream, took the child and his mother and fled to Egypt. Painters have shown the infant Jesus playing on his mother's lap with the Pyramids in the background. It is not necessary to suppose that they went as far southwest as that. What we now call the Gaza strip was then a part of Egypt. Antony had given it to Cleopatra as a wedding present, and it stayed Egyptian after their deaths. Joseph would have had to travel only about 35 miles from Bethlehem to be out of Herod's jurisdiction.


I should like to turn now to Luke's account and come back to Matthew later. Luke says (chap. 2, RSV):

+ In those days a decree went out from Caesar Augustus that all + the world should be enrolled. This was the first enrollment, + when Quirinius was governor of Syria. And all went to be + enrolled, each to his own city. And Joseph also went up from + Galilee, from the city of Nazareth, to Judea, to the city of + David, which is called Bethlehem, because he was of the house + and lineage of David, to be enrolled with Mary, his betrothed, + who was with child.

This passage has been called unhistorical on several grounds: (1) Augustus did not order an empire-wide census. (2) Herod's kingdom was not part of the empire, and would not have been included in a census. (3) Joseph would have been enrolled where he lived, not sent back to his birthplace. (4) If Joseph had been sent to Bethlehem, he would not have been required to bring Mary. (5) The first Roman census in Palestine was in 6 AD (under Quirinius), and we know that it was the first, because it caused an armed uprising. (6) Quirinius was not governor of Syria during Herod's lifetime.

Let us consider these objections in order.

OBJECTION 1: Augustus did not order an empire-wide census.

REPLY: This is a curious argument, since the taking of several empire-wide censuses is one of the accomplishments which Augustus commanded to be engraved on his monument. It probably arises because the critic is thinking of a modern census, conducted everywhere simultaneously. The American census, for example is taken in April of every year divisible by ten. Before 1960 (when the government gave up and started mailing forms out) census takers were supposed to call on every home in the country during April, and on one night they would hit all the hotels, motels, and rescue missions, and try to check out all the bridges, park benches, and alleys, so as to record those with no fixed addresses. The Roman procedure was different. The census takers worked continuously, moving from place to place. The first stage in conducting a census in an area not previously subject to the Roman census would have been mainly a matter of listing all the real estate in the country (APOGRAPHE), with surveyors' descriptions and the names of the owners. After this had been done, the next stage was the assessment of taxes (APOTIMESIS) on the said real estate.

OBJECTION 2: Herod's kingdom was not part of the empire, and would not have been included in a census.

REPLY: It is argued that a Roman census would not have included Herod's territory, since he was nominally not a subject of Rome but an independent king, a Friend of Caesar and Ally of the Roman People. But we know that the city-state of Apamea in Syria, although officially described as autonomous, was subject to a census (carried out, as the inscriptions show, under Quirinius). The Nabataean kingdom of Petra, which minted its own silver coins (unlike Herod, who was allowed by the Romans to mint copper coins only), was likewise subject to Roman supervision of its taxation system. It is unlikely that Herod would be exempt. Moreover, in 8 BC, Augustus quarrelled with Herod, and told him that he would thereafter treat him not as a friend but as a subject. (Josephus says that he soon restored him to favor, but Josephus has an axe to grind: he is concerned to magnify the influence of the man who represented Herod in this matter. Even if Augustus ceased to be angry, it is contrary to the nature of a department of taxation to give up a jurisdiction after having once established it.) So that it seems probable that Augustus, having curbed Herod's independence in 8 BC, would order a census of his territory beginning in 8 or 7 BC. We know that in 7 BC Herod had several thousand Pharisees put to death for refusing to swear allegiance to Caesar. Presumably such an oath would have been administered in connection with the imperial census. We therefore have good grounds, even apart from Luke's statement, for supposing that there was a Roman census in Palestine in about 7 BC.

OBJECTION 3: Joseph would have been enrolled where he lived, not sent back to his birthplace.

REPLY: This is by no means certain, or even probable. In the first place, we have papyrus records of the Roman census in Egypt, and there people were explicitly required to return to the villages of their birth to be registered -- not just the heads of families, but everyone. Now it is possible that the Egyptian rules were special. It has been suggested, for example, that Egyptian peasants had a tendency to leave the farms and migrate to the cities, and that the Romans, who depended on Egyptian grain, were determined to get them back to the farms. We do not know whether the Egyptian rules were applied to the census in Syria and Palestine. But even if they were not, there is another reason why Joseph may have thought himself bound to make the journey. If Joseph's family had come from Bethlehem, he may very well have been owner or part owner of some small bit of land thereabouts, not necessarily of great commercial value, but of great value to him as representing his ancestral inheritance. When the census takers were working at Bethlehem, that land would be surveyed and registered, and any owner or part owner not presenting his claims in person stood to lose them. Now the Jews believed that their land was sacred, that God had given it to his people, and that it had been divided among the 12 tribes and, within each tribe, among families, when the Israelites first entered the land around 1200 BC. The land was not to be sold, but was to pass down from father to son forever. We may doubt that Joseph's little plot of ground had belonged to his family for 12 centuries, or even for five (since the return from the Babylonian captivity). But if it had been in his family for several generations, and if he believed that probably it had once belonged to his remote ancestors, perhaps to David himself, family honor might seem to demand that the land, given to Salmon in the days of the Conquest, passed on to Boaz, Obed, Jesse, and David, later lost to strangers, and finally recovered by Joseph's own great-great-grandfather, should not be lost again simply because the journey to Nazareth would be a hardship.

OBJECTION 4: If Joseph had been sent to Bethlehem, he would not have been required to bring Mary.

REPLY: If the census was being conducted under the Egyptian rules, then obviously Joseph would have been required to bring Mary. If the Egyptian rules did not apply, then he could legally have left Mary behind. However, if he were going to Bethlehem to save the family estate from confiscation, and if Mary had inherited some Bethlehem property in her own right (a possibility we shall come back to later), then she would come along for the same reason. The Roman methods of census taking meant that Joseph would probably have to stay in Bethlehem for some time; thus, it would make sense for Mary to come with him rather than waiting alone in Nazareth for the baby to be born. Another consideration is that Matthew makes it explicit that Joseph and Mary were not formally married until after Mary was already pregnant. In order to spare Mary the gossip that would result from her giving birth less than nine months after the wedding, Joseph may have decided to bring her to Bethlehem fairly early in her pregnancy, and to remain there with her for at least a few years. Christmas pageants give the impression that they arrived in Bethlehem a few hours before the birth, as well as the impression that the Magi arrived a few minutes behind the shepherds. (As soon as the shepherds have gathered around the manger and have finished the last chorus of "While Shepherds Watched Their Flocks By Night," the wise men start down the aisle, singing "We Three Kings of Orient Are.") But the gospels give no grounds for this. It is purely a dramatic convenience. Mary and Joseph may have been in town for weeks before the birth. Nor do we have grounds for dramatics about hard-hearted innkeepers who would not give a pregnant woman any place to sleep except a stable. The inn (or guest-chamber -- same word as in Luke 22:11) would have been a large room with people lying out on mats wall to wall. A woman in childbirth would have found more privacy, and probably more comfort, on a heap of clean straw in a stable. And, despite the familiar pictures, it is unlikely that they had to share the stable with animals. Remember that the shepherds were "abiding in the fields, keeping watch over their flocks by night." If at that season of the year the animals were being kept in the fields day and night, the stable could be cleaned out and scrubbed down, and Joseph the carpenter could put up a few partitions and make the whole place much more pleasant than the guest-chamber. An empty manger would have seemed an ideal crib. Nor do swaddling clothes mean that they had nothing better. Wrapping a newborn baby like a mummy was standard practice. (Thus we see that, whatever special graces Mary may have received, an acquaintance with the child-rearing theories of Dr. Benjamin Spock was not one of them.)

OBJECTION 5: The first Roman census in Palestine was in 6 AD (under Quirinius), and we know that it was the first, because it caused an armed uprising.

REPLY: Josephus mentions no census in the last years of Herod, but he does mention a census in 6 AD when the Romans deposed Herod's son Archelaus and began to govern the land directly through Roman prefects rather than through puppet kings. Moreover, he tells us that it caused a major revolt, put down with much bloodshed. (Luke refers to the same matter in Acts 5:37.) This objection overlooks the delay between the registering of property and the collection of the first taxes on it. These were separate events. Even in Egypt, which was always well organized where land taxes were concerned, there was an interval of several years, and in Gaul it took forty years from the beginning of the APOGRAPHE to the end of the first APOTIMESIS. People who submit quietly when their land is surveyed and title deeds are filed may very well revolt a few years later, when the tax collector comes around demanding cash, just as experience has shown that young men who submit peacefully to draft registration may rebel at actually being drafted. Besides, a dissatisfied people is more likely to rise up at the beginning of a reign than in the middle, when the king is firmly in the saddle and has shown himself prompt, capable, and ruthless in suppressing previous revolts. Also, Herod, who understood Jewish customs and ideas, and made an effort (remodeling and enlarging the Temple, for example) to be popular with his subjects whenever that didn't get in the way of more important things, may have known, better than a Roman governor, how to conduct a census with a minimum of offense, or at any rate a minimum of resistance.

OBJECTION 6: Quirinius was not governor of Syria during Herod's lifetime.

REPLY: See following sections.


Matthew clearly dates the birth of Jesus before the death of Herod, and Luke either before or very soon after. But Luke associates the birth with a census conducted while Quirinius was governor of Syria, and in fact Quirinius was not made governor of Syria until 6 AD, nine years after Herod's death in 4 BC. Whom shall we blame for this seeming error?


We might simply say that Luke has made a mistake. Even the most conscientious scholar occasionally slips, and a man writing probably about 60 years after the event might confuse two censuses a dozen years apart without forfeiting all claim to be considered a careful writer. If we were asking whether Luke is divinely protected against making mistakes, this would be a serious matter, but my policy throughout this paper is to approach Matthew and Luke as I would any other ancient historians, asking only whether their accounts are reliable by ordinary human standards. I am not taking int account any claim (because it lies outside the scope of my inquiry) that they were infallible, or inspired, or otherwise special as historians go, and I could live with an error about Quirinius without any trouble.


We might say that the text of Luke has been miscopied. Since different ancient manuscripts of parts of the Bible contain variations (usually trivial), no one argues that copying errors are excluded. But clearly it is a cheap dodge to assume a copying error whenever a sentence contains difficulties. Are there any special reasons for assuming an error here? One reason is simply that the reference to Quirinius is a kind of parenthetical explanatory remark. And this is just the sort of thing that a copyist is likely to add, in a spirit of helpfulness to the reader. Unfortunately, he does not always check his facts. A book appeared a few years ago about the internment of Japanese-Americans during World War II. The author mentioned that the internment had had the support of the then Attorney General of the State of California, Earl Warren. His editor, without consulting him, added a footnote: "Not, of course, the same Earl Warren who later became Chief Justice of the Supreme Court." A reasonable, helpful, clarifying footnote, except that it really was the same Earl Warren. The editor had simply assumed that it couldn't be, and didn't want the reader confusing the bad guys with the good guys. Similarly, a copyist might very well insert a note: "You know. The census under Quirinius. The one that caused all the trouble." So the Quirinius remark could be a copyist's insertion. Again, it could be an over-zealous correction. Saturninus was governor of Syria from 9 to 6 BC. Perhaps Luke wrote "Saturninus" instead of "Quirinius" and a copyist who knew that there had been a census under Quirinius made the change. If there were some early manuscripts of Luke that read "Saturninus" instead of "Quirinius", this would certainly be a respectable hypothesis. Now we do not have any manuscripts of Luke that mention Saturninus, but we have something similar. Tertullian, writing around 200, says that Jesus was born at Bethlehem during the census taken by Saturninus. One possibility is that Tertullian has calculated the date of the birth on his own, looked up the dates of the Roman governors, and come up with Saturninus as the correct name. Unlikely. Tertullian was not that kind of scholar. Besides, when he does give a date for the birth, it is too late (around 2 BC, if I remember correctly). Can he have found the Saturninus reference in some other work, now lost? This too is unlikely. He knows the four gospels, and refers to them in language that makes it clear that he regards them as the final authority on the life of Christ. It is improbable that he would accept any evidence that Saturninus was the correct name so long as his copy of Luke said that it was Quirinius. I conclude, therefore, that Tertullian's copy said, "Saturninus." If it did, then we have the equivalent of two alternative manuscript readings at this point, and are entitled to prefer the factually correct one.


Instead of supposing that Luke, or at least our present version of Luke, is wrong, we may question whether Luke has been correctly understood. The Greek reads: "This enrollment first occurred of Quirinius governing Syria." Two words here require careful examination: (1) EGENETO, meaning "occurred" or "came to pass," and (2) PROTOS, meaning "first" or "earliest."

Luke uses EGENETO again in Acts 11:28, where a prophet predicts a famine, which Luke explains "occurred" (later) under the emperor Claudius. Thus, it is suggested, Luke may have meant that the census was commanded by Augustus and begun in about 7 BC, but completed under Quirinius, and hence known to historians as the census of Quirinius, or the census of 6 AD, or the like. The objection to this is that if Luke had meant "completed," he should have said so. A strong objection, but not, I think, a fatal one. Luke may have been thinking in terms of a command by Augustus: "I want a list given me of all the property in the empire and who owns it." A dozen years later, Quirinius says: "Here is the list, all checked out." If one is thinking of the result (the whole empire organized on a rational, businesslike, efficient, bureaucratic basis), and not of the process of organization, the expression "occurred" is perfectly natural. Suppose that a speaker says, "Sports fans have long been waiting to see whether Hank Aaron would hit as many home runs as Babe Ruth, and this finally occurred on September 12, 1977." No one takes this to mean that Aaron hit more than seven hundred home runs on the day mentioned, or that in place of "this finally occurred" the speaker ought to have said, "this feat was finally completed," or something of the sort. I maintain that a perfectly possible interpretation of Luke's statement is: "In those days, Caesar Augustus expressed a wish that the domains of King Herod might be completely surveyed and added to the tax rolls. In the days when Quirinius was governor of Syria, his wish was fulfilled." Alternatively, we may suppose that Luke is using the word APOGRAPHE to refer, not to the preliminary surveying, registration, and assessment of real estate, but to the collecting of the taxes. In this case, we translate he statement somewhat as follows: "In those days, Augustus commanded that real estate everywhere should be taxed, and that the preliminary surveying and registration should begin at once. The tax was actually collected (in Palestine) in AD 6 when Quirinius was governor of Syria." This interpretation has Luke using the same term for the registration and for the taxation, but this is not surprising in anyone not a tax lawyer. Given this interpretation, Luke's use of "occurred" rather than "completed" for the events of AD 6 is above even the most nit-picking criticism.

Turning to the word PROTOS, translated "first," we note that it it means "earliest," but is also sometimes used to mean "earlier, prior, previous." John the Baptist says (John 1:15,30), "he was first of me", meaning, "he was earlier than I." Greek regularly uses "of" (the genitive case) rather than "than" to express comparison. So that Luke's words can be translated: "This enrollment occurred earlier than Quirinius governing Syria," meaning, "This was the census just before the big one (the one that everyone knows about, because it started a rebellion) under Quirinius." There are in fact three plausible ways of parsing the phrase: (1) We can read it as "the first census, Quirinius being governor of Syria." This treats "Quirinius governing Syria" as a genitive absolute, similar to the Latin ablative absolute or the English nominative absolute as in "Jones took notes, the regular secretary being absent," or the more frequent, "The picnic will be on Tuesday, weather permitting". This is the construction assumed by most English translators. (2) We can read it as "the census earlier than Quirinius governing Syria." This treats "Quirinius governing Syria" as a genitive of comparison. (3) We can read it as "the census earlier (than the census) of Quirinius governing Syria." This assumes that Luke omits the second occurrence of the word "census", as if I were to say, "My dog is smarter than his," expecting people to understand that I meant, "My dog is smarter than his dog." Such omissions of repeated words are standard in many languages. In John 5:36, we have, "The testimony I have is greater than (the testimony) of John." In 1 Corinthians 1:25, we have "The foolishness of God is wiser than (the wisdom) of men, and the weakness of God is stronger than (the power) of men." If this is the construction Luke intended, then "Quirinius" is an ordinary genitive of possession, modifying "census" understood. On either the second or the third interpretation, all difficulties vanish. George Ogg, who thinks that Luke is wrong about Quirinius, objects that there are no undisputed passages in which PROTOS followed by a participial phrase in the genitive case clearly means "before." I think this is unreasonable. He does not deny that PROTOS often means "before," or that participial phrases can occur in the genitive of comparison. I think his objection is little like questioning the authenticity of the Gettysburg Address on the grounds that Lincoln nowhere else uses "fourscore" to mean "eighty." Even granted that he is right, this would eliminate the second construction listed above, but leave the third (which he does not consider) as a perfectly good possibility.


Instead of trying to defend Luke against a charge of saying that Quirinius was governor of Syria during the lifetime of Herod, we may admit that he did say it and maintain that he was right. I know of two arguments used for this purpose. The first argument relies on an incomplete inscription known as the Tiburtine inscription. This lists the accomplishments of an illustrious Roman. Unfortunately, the top of the stone is broken off, so we cannot be sure which illustrious Roman. However, some of the details match known details in the career of Quirinius, and the mystery man was apparently governor of Syria twice, with other appointments between. If the inscription does refer to Quirinius, then there is no reason to dispute Luke's assertion (if he does assert it) that Quirinius was governor around 7 BC. The other argument does not try to fit Quirinius into a list of governors of Syria, but into a list of vice-emperors of the East. Agrippa held this post from 23 BC until his death on March 12, 12 BC. At that time, Augustus released Quirinius, one of the two consuls for that year, from his consular duties. The inference is that he was released so that he could take Agrippa's place. For the next 27 years he is found in the Orient. Agrippa's son Gaius was appointed vice-emperor of the East (ORIENTI PRAEPOSITUS) in 2 BC, but died in 4 AD at the age of 23. Given his youth, he probably governed under Quirinius's tutorship. After his death, Quirinius took full charge again, until AD 16, when Germanicus was appointed. If I read my secondary sources correctly, there is no explicit record of Quirinius's receiving the title ORIENTI PRAEPOSITUS. However, Pompey, Mark Antony, and Agrippa held it before him, Germanicus held it after him, Gaius held it for five years in the middle, and during the rest of the time from 12 BC to 16 AD no one seems to be acting more like an ORIENTI PRAEPOSITUS than Quirinius. His name comes up repeatedly in connection with military campaigns and censuses. Given that the census began in 8 BC, with Saturninus holding the title "Governor of Syria" at that time, it seems likely that Quirinius outranked Saturninus in Syria and elsewhere, that he took charge of the census, and that, unlike Saturninus, he was around from start to finish, so that it is most appropriately described as taking place under his governorship.

SUMMARY: I conclude that Luke is clearly right about all aspects of the census except the assertion that Quirinius was governor of Syria at the time. Here, perhaps Luke has wrongly confused two censuses, perhaps he was miscopied, perhaps the translation that makes him assert this is wrong, perhaps Luke described Quirinius's role non-technically, and perhaps Quirinius really was governor of Syria at the time. It is unsatisfactory not to know which, but when we are dealing with ancient history, sometimes the evidence is ambiguous. We have seen that Luke is right on all points but one, and on that one cannot be shown to be wrong. And I think we must leave it at that. </PRE>