The star of Bethlehem, or, as I shall normally call it here, the Christmas star, has fascinated people for a very long time. Many have asked what the star might have been, and so theories abound. Late each year most planetaria feature programs exploring some of these possibilities. The account of the Christmas star is found only in the second chapter of Matthew’s gospel. Matthew wrote primarily to a Hebrew audience showing with reference to Old Testament prophecies that Jesus was the Messiah. The Christmas star is an obvious allusion to a Balaam’s prophecy of Numbers 24:17 which reads,

I shall see him, but not now: I shall behold him, but not nigh: there shall come a Star out of Jacob, and a Sceptre shall rise out of Israel, and shall smite the corners of Moab, and destroy all the children of Sheth.

In 2007, the attorney Frederick A. Larson produced a DVD entitled The Star of Bethlehem. Since its introduction, this DVD has been popular with many Christians, particularly near Christmas. In this DVD, Larson presents his journey to determine what the star of Bethlehem was. He assumed that the star was a real event, not some fabrication by the gospel writer Matthew to authenticate Jesus as the Messiah, as many liberal theologians maintain. He also assumed that the star was an astronomical event that could be reproduced with precise calculations using our modern knowledge of how the planets move. To do this, Larson used “Starry Night,” software that runs on virtually any desktop computer to see what the night sky might have looked like to observers in the Middle East about the time of the birth of Jesus.

An Earlier Proposal

One must know at least the approximate date of the birth of Jesus in order to compute a possible natural astronomical event as the star of Bethlehem. Since Matthew chapter two requires that Jesus was born before the death of Herod, the time of Herod’s death is very important in this calculation (Herod ordered the baby boys in Bethlehem up to the age of two years old killed). Josephus reported that Herod died near the time of Passover and that there was a lunar eclipse visible from Israel shortly before Herod’s death. Astronomers and historians have long known that there was a partial eclipse of the moon visible from Jerusalem on March 13, 4 BC, which was a month before Passover, so most historians date Herod’s death to that year. Thus, most historians conclude that Jesus was born about 5 BC or possibly a year or two earlier.

Using this clue, the famous German astronomer Johannes Kepler suggested that a triple conjunction of Jupiter and Saturn that occurred in BC 7-6 was related to the Christmas star. A conjunction is when two astronomical bodies appear to pass very close to one another in the sky. Due to their orbital motion around the sun, Jupiter, Saturn, and the other planets slowly move eastward with respect to the stars as seen from the earth. It takes Jupiter 12 years to pass once through the stars, and it takes Saturn 30 years to do so. Roughly every twenty years, Jupiter overtakes and laps Saturn, causing a conjunction. However, being superior planets, both Jupiter and Saturn undergo retrograde, or a small backward, westward, motion from time to time. On rare occasions retrograde motion combines with the conjunction of Jupiter and Saturn so that the two planets pass each other three times instead of the usual once. This is a triple conjunction. Since the triple conjunction of 7-6 BC, Jupiter and Saturn have come into conjunction approximately 100 times, but only 11 times since then has it been a triple conjunction. The most recent was in 1980-1981. Kepler noted that this triple conjunction was in the constellation Pisces, a constellation (and astrological sign) identified with the ancient Hebrews. This is a relatively rare event, and Kepler reasoned that this unusual event in a sign associated with the Hebrews was the star that the magi saw about the time of the birth of Jesus. Nearly everyone who has attended a planetarium show near Christmas has seen a presentation of this idea, and it remains one of the more popular explanations of the Christmas star. Right away one might question the connection of Pisces to the Hebrews. Such references are rather late, from the medieval period, and there are several different signs that different sources claim as being specially related to the Hebrews. All those signs cannot be related to the Hebrews. The ancient Hebrews were forbidden to embrace astrology, so obedient Hebrews would not make this connection. In the late Roman era, Jews began to Hellenize, and astrology crept in, giving rise to these medieval traditions.

Mosley (1981) has pointed out that there is more to the story about Kepler and this explanation of the Christmas star than is normally relayed. Kepler noted that the triple conjunction of Jupiter and Saturn was followed the next year, in 6 BC, by a massing of the planets Mars, Jupiter, and Saturn in the constellation Aries. A massing is a close grouping of three or more objects, though generally less close than a conjunction. Kepler observed a similar massing of these three planets in the year 1604 in the constellation Sagittarius, and that massing was immediately followed by a bright “new star” that was visible for several months. Today we recognize that this “new star” was a supernova. Astrologers recognize three “fiery signs,” Aries, Leo, and Sagittarius. Kepler thought that since the 1604 massing of three planets in one of the fiery signs produced a “new star,” the 6 BC massing of the same planets in another fiery sign produced a “new star”; it was this historically unreported “new star,” not the triple conjunction, that was the Christmas star. Kepler reasoned that the triple conjunction began the astronomical events that got the attention of the magi, but it was this nova that was the true Christmas star. However, as many people have already noted, a nova does not fit the behavior of the star that Matthew described.

A Later Date for Herod’s Death and a New Proposal

In recent years some historians have come to the position that Herod died in 1 BC, about three years later than normally thought. There was another lunar eclipse visible in Jerusalem on January 9, 1 BC, three months before Passover. Proponents of this theory suggest that in 4 BC the mere month between the lunar eclipse and Passover made it difficult for other events recorded by Josephus to have happened between the eclipse and the death of Herod, making the later date of Herod’s death more likely. Proponents further point out that the 1 BC eclipse was total and that the 4 BC eclipse was partial, thus making their preferred eclipse for dating Herod’s death more impressive. However, Josephus made no special claims about the eclipse, but rather remarked that one was visible shortly before Herod’s death, which could be either partial or total. We cannot be totally certain about either date, but most historians still favor the traditional 4 BC date. This is important, for all of what Larson claims relies upon the later date of Herod’s death. If Herod died in 4 BC, then nothing else that Larson claims can be relevant.

If Herod died in 1 BC, then the 7-6 BC triple conjunction of Jupiter and Saturn can be ruled out as the star of Bethlehem. So what was the Christmas star? Using his desktop software, Larson found that on August 12, 3 BC there was a conjunction of Jupiter and Venus. This was followed by a triple conjunction of Jupiter and the bright star Regulus in September 10, 3 BC; February 17, 2 BC; and May 8, 2 BC. This was then followed by a very close conjunction of Venus and Jupiter on the evening of June 17, 2 BC. Larson argues that this conjunction was so close that for a while Venus and Jupiter merged into a single object as viewed with the naked eye. Venus is the third brightest object in the sky, and Jupiter usually is the fourth brightest object (after the sun and moon), so such a merger into a single object would be most impressive. Jupiter was viewed as the king of the gods, and Regulus was viewed as a kingly star (we get the word “regal” from the same root as Regulus). Furthermore, Regulus is in the constellation of the lion, and so this marked the birth of an important king, the lion of Judah.

Did the magi actually see the two planets appear to merge into a single star, and did it have the significance that Larson claims? There is some doubt, and we ought to examine four factors to decide this. First, we need to determine how closely the two planets actually appeared to come to one another. There is some uncertainty in any computation of positions of planets into the distant past and future, so the reported separation of the two planets varies from source to source. However, all sources agree that the minimum separation was less than ten arc seconds, and since most people with good eyesight can resolve separations of about one arc minute, the two planets certainly did merge for at least some observers. A second question to ask is what the maximum separation would have had to be for the two planets to appear as one. Though I just mentioned that one arc minute is the minimum resolution standard, bright objects such as these tend to bleed over and overlap, thus raising the limit somewhat above one arc minute. Additionally, the close separation of Jupiter and Venus would have happened low in the sky, and atmospheric blurring would have made visible separation of even two arc minutes difficult. Third, the close separation of Jupiter and Venus would have changed rapidly and would have appeared as one for only a restricted range of longitude. Did the magi live along that range of longitude? The minimum separation was not observable from the Middle East, for the minimum separation occurred some time after the two planets set that evening. Sinnott (1968) showed that at sunset in Babylon on June 17, 2 BC Jupiter and Venus were separated by eight arc minutes, and were thus easily resolved as two objects then. The two approached one another before they set three hours later. Sinnott concluded that the pair likely appeared to merge for most people when they were four arc minutes apart. At that point they were only ten degrees above the horizon, and they set about 45 minutes later. At sunset, Sinnott says that the two objects were separated by three arc minutes. Given the uncertainty of how distinct that the two planets would have been, we cannot say for certain if an observer with good eyes in Babylon would have seen them as one star. Fourthly, the conclusion that the magi saw this as the Christmas star required that the weather was very good in their location. Since they likely lived in a desert location, this had a high probability, but it is not certain.

Larson doesn’t state how long that Venus and Jupiter could have appeared as a single star, though a layman viewing the DVD might easily conclude that it was visible for some time. As we have seen, if the magi saw this at all, it would have been for only a few minutes one evening. The omission of this fact in Larson’s DVD is a bit misleading, but it is likely unintentional. Any significance that this possible sighting might have had for the magi is conjecture, for we don’t know anything about them. If they were astrologers (which is likely) and if they were familiar with some of the OT (which is less likely), there is no guarantee that they would have interpreted this rare event as Larson and others conjecture. Piecing together a remarkably close conjunction of the (supposed) kingly planet with another bright planet near a kingly star might mean the birth of a king to some people, but not necessarily to all. And the identification with Judah because of this event occurring in Leo is interesting, but is it necessarily what astrologers from 2,000 years ago would have thought? This is conjecture. Recall that the triple conjunction of Jupiter and Saturn in the constellation Pisces (a sign some say was identified with the Hebrews) in 7 BC was supposed to herald the birth of the Messiah. Clearly, this sort of musing amounts to a sort of Rorschach test.

Many speculate that the magi would have been familiar with the writings of Daniel, from which the vision of 70 weeks allowed to at least estimate the time of the coming of the Messiah. They further speculate that their knowledge of Numbers 24:17 allowed the magi to know the significance of the star. However, the magi appear to have been ignorant of the Micah 5:2 prophecy of the Messiah being born in Bethlehem. Micah was written a century-and-a-half before Daniel was written, so Micah certainly could have been as available to the magi as the Pentateuch was. Of course, one could speculate that the magi did not have access the entire OT but just happened to have studied Daniel and Numbers, but this further compounds speculation.

Of particular interest to Larson is September 11, 3 BC. He reasons that on that date Jupiter and Regulus were close to one another. The sun was in Virgo (the virgin), and the crescent moon was below the virgin, though not visible in the morning when Jupiter was visible, but visible in the evening to mark Rosh Hashanah, the first day of the first month of the Jewish civil new year. Larson believes that this arrangement is what Revelation 12: 1–5 is referring to, though few, if any, commentaries endorse this view. True, most commentators think that Revelation 12:1–5 does speak of the birth of Jesus, but the passage speaks of far more. For instance, the child born in Revelation 12:5 certainly refers to Jesus, but the woman mentioned in these verses is almost certainly not Mary alone, as would be required by Larson’s teaching (though this is the teaching of the Roman Catholic Church). There is much symbolism here, and making direct correspondences here can lead one astray. Making what appear to be direct astrological parallels appears dangerous and even unbiblical. Furthermore, the circumstances described here by Larson are not that unusual. The thin waxing crescent moon is in this part of the sky every year at Rosh Hashanah. Every twelve years Jupiter is near Regulus and is so for much of a year, so the circumstances Larson describes for this happen every twelfth year, which is not as remarkable as Larson implies to his audience.

There are other problems. Matthew 2:9 states that after the magi left Herod

. . . the star, which they saw in the east, went before them, till it came and stood over where the young child was.

Larson believes that this standing of the star refers to a stationary point in Jupiter’s retrograde motion. This retrograde motion would have appeared low in the southern sky, and so Larson thinks that the star’s standing over the place where Jesus was was merely an illusion as the magi traveled southward from Jerusalem to Bethlehem. Larson further dates this event to December 25! Scholars long ago debunked whether December 25 had any real connection to the birth of Jesus. Instead, it is virtually certain that the early Roman Church adapted a pagan observance of the winter solstice, which was on December 25 of the Julian calendar two millennia ago. Thus, this retrograde motion stop on December 25 is nothing more than coincidence, for if the early Church had adopted some other date, it would mean nothing to us today.

There are problems with Jupiter’s stationary point being the star that guided the magi. First, Matthew 2:10 suggests that the magi had not seen the star for a while, for they were glad to see it on their way to Bethlehem from Jerusalem. The magi would have seen Jupiter much of the time that they traveled, so why the delight now? Second, the appearance of Jupiter over Bethlehem was of no use in finding Bethlehem, for they already had been told to go there and likely were given directions to the town just a few miles from Jerusalem. Third, the magi seemed to have had no difficulty identifying which house in Bethlehem to go to, suggesting that the star appeared over a particular house, rather than just appearing over the town in general. Thus, while Larson puts forth an argument for the star fitting Matthew’s description, the argument really doesn’t survive close scrutiny.

Larson further speculates on the date of the crucifixion. If Jesus was born in 2 BC, then he would have begun his ministry about AD 30. We know that he began his ministry about the age of 30 years, and there was no year zero: 1 BC was immediately followed by 1 AD. This means that the three-year ministry could have ended in the spring of 33 AD. Larson notes that there was a lunar eclipse on April 3 that year, and that evening marked the Passover. At Pentecost 50 days later, Acts 2:20 records Peter quoted Joel 2:31 with reference to the sun being darkened and the moon turned to blood. Mark 15:33 records that there was darkness between noon and 3:00 PM on the day of the crucifixion. Larson argues that since an eclipsed moon often appears deep red, the moon that rose that night was fulfillment of the blood moon and the darkness at the time of the crucifixion of Jesus was the fulfillment of Joel’s prophecy that Peter noted in his sermon at Pentecost.

How correct are Larson’s claims about this eclipse? It is true that a totally eclipsed moon does often appear reddish. However, a totally eclipsed moon does not always look red, but may be completely dark or it may have a great range in color from black to yellow, so we don’t know what the moon looked like during that particular eclipse (For instance, the July 6, 1982 very deep total lunar eclipse appeared partially black and partially peach-colored—no one could recall such an odd eclipse). If there were a total lunar eclipse on the night of April 3, AD 33, then it might have appeared red, but this particular eclipse was not total. The standard reference on eclipses is Oppolzer’s Canon of Eclipses (1887). In what follows, I shall use information from the 1962 Dover translation of this work.

This particular eclipse is number 1914 in the series. The half duration of the eclipse was 81 minutes, and the universal time (UT) of mid-eclipse was 15:06. Thus, the partial phase began at 13:45 UT and ended at 16:27 UT. According to Oppolzer’s Table VI, there is no appreciable correction for the equation of time for this date. Jerusalem is at longitude 35° 13´E, which corresponds to a 2:21 correction from UT to local time (LT). Therefore, for Jerusalem the partial phase of the eclipse began at 16:06 LT, mid-eclipse was at 17:27 LT, and the partial phase ended at 18:48 LT. In the first century the vernal equinox was on March 25, so this eclipse was just nine days after the vernal equinox. At the vernal equinox, moon rise would be at 6:00 PM, so moon rise on April 3 would be delayed at most 2-3 minutes; we will ignore this factor, and assume that moon rise was at 18:00 LT. The eclipse had already passed its deepest point more than a half hour before the time that moon rose that night. There was at best 48 minutes of partial phase remaining when the moon rose. Normally, any deep red color noticeable during a total lunar eclipse is not at all obvious during nearly all of the partial phases. This is because the much brighter portion of the un-eclipsed moon overwhelms the subtle color of the dark portion. I have seen many lunar eclipses. I have yet to see a partially eclipsed moon that I would describe as resembling blood. Since this eclipse was far from total, and it was not observable from Jerusalem until very late stages, the moon must have been very bright, and there is absolutely no basis for Larson’s claim that this was a blood moon. Larson’s pronouncements on this are very misleading, for the vast majority of people watching his DVD would have no idea that his discussion of the alleged “blood moon” are not true. One could speculate that unusual atmospheric conditions near Jerusalem at the time could have deeply reddened the full but partially eclipsed moon that night, but then there is no reason to appeal to an eclipse to account for this.

Larson claims that the eclipse began at 3:00 PM (15:00 LT) in Jerusalem. This is an hour before the partial phase began, so this time must refer to the penumbral phase, which Oppolzer did not compute. This is supported by an image in the DVD that appears to depict the sun in the penumbral shadow. Oppolzer probably did not compute penumbral shading, because penumbral shading is extremely difficult to detect, so there is not much of significance in it. Why such an unimpressive type of eclipse occurring at the time of Jesus’ death that no one particularly noticed has any significance is not clear.

In a special feature on the DVD, Larson shows how that this lunar eclipse, if viewed from the lunar surface, would have been a total eclipse of the sun in the constellation Aries (the ram, a sacrificial animal) at the very time of the death of Jesus. It is not clear how a solar eclipse that no one could witness (because it was visible only on the moon) could have any real significance. However, any total phase visible from the moon would not have happened until an hour after Jesus died, not during his sacrifice as Larson at least implies.


In conclusion, Larson’s thesis is fraught with problems. It completely relies upon the late death of Herod, something that few historians have embraced. It has some obvious astrological connections that we have not discussed in any detail here. These astrological connections are related to the so-called gospel in the stars theory, which is without foundation, and I have discussed elsewhere. Though Larson does not date the birth of Jesus to December 25, there is a suspicious clinging to that date that has no real connection to the birth of Jesus. This star doesn’t fit the description of Matthew 2. Finally, Mr. Larson means well, and many people who have watched the DVD have gleaned much encouragement from it, but it does take us into a dangerous direction. This DVD attempts to give some scientific answer for why the Bible must be true. The Bible never makes such claims for itself, for it stands as propositional truth, and we reject it at our peril. The DVD attempts to bolster peoples’ faith by showing that there is a natural explanation for the star of Bethlehem, as if we doubted that it actually happened. But by giving a natural explanation for the Christmas star, this cuts both ways and offers an out to the unbeliever. That is, the skeptic can now claim that an unusual, but perfectly natural, set of astronomical events was folded into the set of myths that we call the Bible. One may as well search for natural explanations for other miraculous events in the Bible, such as the crossing of the Red Sea or the plagues of Egypt. While these attempts to somehow prove the Bible might encourage some believers, it frequently has the opposite effect on non-believers.

I have another concern. The DVD strongly implies that the identification of the Christmas star with the previously discussed phenomena involving Jupiter, Regulus, and Venus during 3 – 2 BC were entirely the work of Larson. However, others had previously published on the subject. For instance, Sinnott (1968) had called attention to the close conjunction of Jupiter and Venus nearly four decades earlier. Sinnott accepted the conventional date of 4 BC for the death of Herod, but almost no one questioned that date then. Ernest L. Martin’s The Star that Astonished the World (1991) is briefly mentioned in a review of the DVD on Larson’s website, but neither there nor on the DVD does Larson credit that work. Martin earlier had argued for a later birth of Jesus in his book, The Birth of Christ Recalculated (1978), and much of what Larson teaches is found there. Craig Chester (1993, reprinted in 1996) also published much of Larson’s thesis long before Larson’s DVD. For the most part, nothing on Larson’s DVD was original to him, so why didn’t Larson credit those earlier works? Either Larson didn’t know about those earlier works when he did his research and produced his DVD or he did know about them. If he did not know about those works, then he didn’t do proper background research before starting his work (though he does seem to state that he did his homework). On the other hand, if Larson knew about those works and failed to credit them while suggesting to his viewers that he came up with his findings on his own, then that is plagiarism.

If the star of Bethlehem was not a planetary alignment and conjunction with a star, then what was it? The only record we have of the star is Matthew 2. Presumably, the magi were very educated men, well versed in astronomical knowledge of the time. As previously mentioned, many have speculated that they were from Persia, and hence knew of Daniel’s prophecy of the 70 weeks, and thus knew that the time of the Messiah was nigh, as did many Jews at the time. Then they saw something in the sky that grabbed their attention and caused them to travel hundreds of miles to seek out this new king. If the magi were well schooled in astronomy, they would have known that any planetary conjunctions, though unusual, would have been repeated in the past and future, and so would not be the unique event suggested in Matthew 2. The Matthew 2 text suggests that the star wasn’t always visible to the magi, nor was it necessarily visible to others. This suggests that the star may have been a localized object specifically fashioned as a message to the magi. This supernatural object would not be bound by the motions of objects normally found in the sky, and thus its odd behavior, such as appearing over where Jesus was, is easy to explain. In short, the star of Bethlehem likely was a unique and miraculous local apparition to fulfill God’s purpose and one of the ordained purposes for stars to be for signs as in Genesis 1:14. This is not unusual, for the Lord used the Shekinah glory to guide Israel in the wilderness. Of course, others before me have noted the difficulty of relating Matthew’s account to any known astronomical object and have suggested such a thing (Custer (1977), DeYoung (1989), Gitt (1996), Lisle (2008), MacArthur (2006)).


Chester, Craig. 1993. The Star of Bethlehem. Imprimus. December (reprinted in the December 1996 issue of Imprimus).

Custer, Stewart. 1977. The Stars Speak: Astronomy and the Bible, pp. 75–82. Greenville, South Carolina: Bob Jones University Press.

DeYoung, Don B. 1989. Questions and Answers on Astronomy and the Bible, pp. 65-66. Grand Rapids, Michigan: Baker.

Gitt, Werner. 1996. Stars and their Purpose: Signposts in Space, pp. 103-111. Bielefeld, Germany: Christliche Literatur-Verbreitung e. V.

Lisle, Jason. 2008. “What was the Christmas Star?” in The New Answers Book 2, ed. Ken Ham, pp. 157–162. Green Forest, Arizona :Master Books.

MacArthur. 2006. The John MacArthur Study Bible. Thomas Nelson. P. 1362 (note on Matthew 2:2).

Martin, Ernest L. 1978. The Birth of Christ Recalculated.

Martin, Ernest L. 1996. The Star of Bethlehem: the Star that Astonished the World. Portland, Oregon: ASK Publications.

Mosley, John. 1981. “Common Errors in ‘Star of Bethlehem Planetarium Shows’” The Planetarium. Third quarter, 1981.

Von Opplozer, Theodor Ritter. 1887 Canon of Eclipses. Translated by Owen Gingerich. 1962 Dover Edition.

Sinnott, Roger W. 1968. “Thoughts of the Star of Bethlehem.” Sky and Telescope. December:384–386.

Help keep these daily articles coming. Support AiG.