Numerous suggestions have been made over the centuries for the identity of the star that led the Magi to Jesus. Some of these proposals may be classified as purely natural, in that no miraculous intervention is postulated - e.g., Halley's comet, some other comet, one or more of various planetary conjunctions, or an exploding star, such as a supernova. Other proposals can be labeled purely supernatural, since they suggest that a miraculous object - perhaps an angel, the Shekinah glory, or some other localized, movable bright light in the sky - was responsible for the phenomenon.
 

The advantage of this first class of proposals is that it may be possible to find other historical references to the object, or (for Halley's comet or planetary conjunctions) to re-enact the scenario using modern computers. The problem with such proposals is that the bright object, being millions of miles out in space, does not naturally fit Matthew's comment that the star led the Magi to the place where the child was. Supernatural models, on the other hand, can be easily imagined in which a relatively local bright object will do the guiding. But it would be very unlikely that such an object would leave any extra-biblical traces in the historical record.
 

Here we propose a hybrid model - part natural and part supernatural - that both leaves a distinctive mark for us to investigate two thousand years later, and which will also fit a straight forward reading of Matthew's account. Our proposal involves a modification of Ernest Martin's interpretation of a series of planetary and stellar conjunctions, to which is added a localized bright object that would have been recognized by the Magi as the "same" star and which, on that crucial night, led them to the house where the baby Jesus was.
 
 

Retrospective calculations of planetary positions for ancient times have occasionally been made since the time of Johann Kepler (d 1630). But such efforts (being done by hand) were enormously time-consuming, exacting and tedious until very recently. An early fruit of the development of electronic computers was the publication in 1962 of Bryant Tuckerman's tables of Planetary, Lunar and Solar Positions 601 BC to AD 1 by the American Philosophical Society. In the following years, Roger Sinnott worked carefully through these tables and located all the close planetary conjunctions that occurred around the probable time of the birth of Jesus. In particular, Sinnott found one strikingly close conjunction of Venus and Jupiter in the constellation of Leo (the Lion) on the evening of 17 June 2 BC, which would have been quite rare. He proposed that this was the Christmas star, publishing his results in the December 1968 issue of Sky and Telescope, the premier magazine for amateur astronomers. His suggestion has since been adopted and presented each Christmas in many planetarium programs. Ernest Martin went on to develop the suggestion with much additional detail and historical research in his book The Star That Astonished the World.
 

Martin noticed a cluster of additional conjunctions involving one or the other of these two planets within a year of this conjunction. He suggested that these would have had such symbolic significance as to send the Magi to Judea to look for the newborn king of the Jews.
 

Martin suggests that these conjunctions above, plus a configuration of sun and moon with the constellation Virgo on 11 Sept 3 BC, would have been read as follows:
 

12 August 3 BC. Jupiter, the king planet, having left the Sun, the supreme Father, is now conjoined with Venus, the mother. The Sun, Moon and Mercury (the messenger) are also located with Jupiter and Venus in Leo, the lion, representing the tribe of Judah (Gen 49:9; see fig. 1). The message of this configuration: some important royal event is to happen among the Jews.

31 August 3 BC. Mercury, the heavenly messenger, having left the Sun, now arrives at Venus, the mother. The Sun is now in the constellation Virgo, the virgin. Mercury and Venus are in Leo and Jupiter is just entering the same. Message: God's messenger is sent to the mother.
 

11 September 3 BC. One of the visions in Revelation, that of the woman about to give birth in chapter 12, verses 1-5, can easily be understood astrologically to mark an exact date. If the woman is taken to be the constellation Virgo, the virgin, then the Sun clothes her body for about one month per year. The moon will pass through that region in the course of its monthly cycle so that it is (just) under her feet on one particular day. In 3 BC, that day is September 11. Martin takes this to be the date of Jesus' birth. See fig. 2.

14 Sept 3 BC, 17 Feb 2 BC, and 8 May 2 BC. Three conjunctions of Jupiter with Regulus, the brightest star in the constellation Leo, the lion, and located between the lion's feet. Martin notes the prediction in Gen 49:10: "The scepter will not depart from Judah nor the ruler (or ruler's staff) from between his feet, until he comes to whom it belongs and the obedience of the nations is his." Jupiter, the king planet, circles above Regulus, the king star, placing a crown on the king star. See fig. 1.
 

17 June, 2 BC. A very rare close conjunction of Jupiter and Venus, which are variously estimated to have a separation of 3 minutes of arc (Sinnott), 0.5 minute (Martin), or 0.1 minute (Carroll). The frequency of such a close conjunction (depending on which estimate is correct) would range from once in about 1150 years to once in about 34 thousand years. See my calculations in the appendix. Martin sees this conjunction as having the same significance as the one of 12 Aug 3 BC, and suggests it was this 17 June conjunction that sends the Magi on their way, arriving in Jerusalem about December 25.
 

26 August 2 BC. A close conjunction of Jupiter and Mars (war) in Leo, with Venus and Mercury also in Leo, and the Sun in Virgo. Martin suggests this sign may mean these events will lead to war.
 

Recently I have noticed a number of things about these conjunctions that point to a modification of Martin's view. When these changes are made, the symbolism of this sequence of events suddenly snaps into focus in a remarkable way.
 

1. Martin's 11 September event comes only two weeks after the 31 August conjunction, in which God's messenger comes to the mother. This 31 August event sounds like Gabriel's annunciation to Mary, which would hardly be just two weeks before the birth.
 

2. Taking 11 September as the conception, then the really close conjunction of 17 June is just over nine months later. This makes best sense as the birth event, which (after all) is what Matthew 1-2 is all about. So Jesus is born 17 June 2 BC.
 

3. This arrangement gives a different flavor to the three conjunctions of Jupiter with Regulus, as the baby Jesus is in Mary's womb at this period. Thus Jupiter can be seen as God hovering over him. As an unexpected benefit, the name Regulus is the Latin diminutive of Rex, "king." So the king planet is hovering over the little king!
 

4. The 26 August 2 BC massing of planets - all the anciently-known planets but Saturn clustered within a few degrees of one another in the space between Leo and Virgo (Sun, Moon, Mercury, Venus, Mars, Jupiter) - may indeed imply war, as Martin suggests, given the presence of warlike Mars for the first time in this sequence of conjunctions. If we assume with Martin that the Magi set out from Babylon (or thereabouts) at the really close conjunction of 17 June, they would have just 70 days to get to Judea by this date. This is not unreasonable if they didn't dawdle. See the appendix. My scheme thus looks like this, with a date or two adjusted to fit the results of my own planetarium program:
 
 

12 August 3 BC. Jupiter, the king planet, joins with Venus, the mother, in Leo. An important royal event - presumably a birth - is about to occur in Judah. See fig. 3.

26 August 3 BC. Mercury, the heavenly messenger, comes to Venus the mother. The message is delivered to the mother-to-be. See fig. 4.

11 September 3 BC. The Sun clothes the body of the Virgin while the Moon is under her feet. The conception of the child. See figs. 2 and 5.

14 September, 17 February and May 8, now 2 BC. Jupiter the king planet hovers over Regulus, the little king, who is in his mother's womb.

17 June 2 BC. The rare close conjunction of Jupiter with Venus in Leo, almost exactly the same distance East of Regulus as the earlier 12 August 3 BC conjunction was West of Regulus. See fig. 1. The birth. The Magi set out for Judea and Jerusalem, its capital. See fig. 6.

28 August 2 BC. The clustering of Sun, Moon, Jupiter, Venus, Mercury and Mars. This means war. The Magi arrive, and Herod makes his move to kill the baby Jesus. See fig. 7.

How rare is the sequence of events we are considering here? As noted above, this will depend on the separation of Jupiter and Venus at the close conjunction of 17 June 2 BC. With Sinnott's suggested separation of 3 minutes, this sort of conjunction will occur in Leo about every 1154 years. With Martin's separation of 0.5 minute, it will occur about once every 6923 years. With Carroll's separation of 0.1 minute, once every 34,615 years.

The event is seen to be even rarer when we add in the triple conjunction of Jupiter with Regulus, which only occurs in this particular year because the eleven degree wide loop made by Jupiter at opposition happens to lie across the position of Regulus. This has a probability of 11 degrees divided by 360 degrees, or about .03. Thus the frequency of this close Jupiter-Venus conjunction plus the triple conjunction with Regulus can be expected to happen once every 38 thousand years (Sinnott), every 228 thousand years (Martin), or every 1 million 142 thousand years (Carroll). A rare event indeed!
 

The symbolism of these events (and their chronological order) is also impressive:

12 August 3 BC. Magi pay attention! A royal birth in Judah.

26 August 3 BC. The heavenly messenger informs the mother.

11 September 3 BC. The conception occurs.

The triple conjunction. The king of the gods hovers over the little king.

17 June 2 BC. The birth occurs.

28 August 2 BC. This means war.

The apologetic value of this sequence of events - which can easily be reconstructed two thousand years later by anyone on a home computer using a planetarium program - is enormous.
 
 

The purely natural models tend to run aground in dealing with the text of Matthew 2:9-10:
 

(9) After they had heard the king, they went on their way, and the star they had seen in the east went ahead of them until it stopped over the place where the child was. (10) When they saw the star, they were overjoyed.

The problem here is that objects at astronomical distances cannot guide someone to explicit places on the surface of the earth by mere movement alone. Instead this is characteristic of objects that are only a few tens of feet above the ground, as for example, the pillar of cloud and fire in the wilderness.
 

Martin's model takes the verb "stopped" to be used in a technical astronomical sense, for Jupiter ceasing its westward motion relative to the stars, and beginning to move eastward once more, which he sees as happening about 25 Dec 2 BC, when he believes the Magi were at Jerusalem. This is nothing an observer would notice by watching Jupiter that night. Its stopping would only become apparent over the course of many nights as its relative motion among the stars would gradually come to a halt and then change direction. And "stopped over the place" Martin takes to mean that when the Magi came out of Jerusalem that night to go to Bethlehem, they saw Jupiter "stopped" in the southern sky over the village of Bethlehem, not over the particular house in which the baby Jesus was located. Since the Magi would need to know that Jupiter had reached its stopping point (before moving east again) by means of calculations or charts they had prepared in advance, they would hardly be surprised or overjoyed as Matthew tells us they were. This Greek expression is quite emphatic.
 

Another suggestion for using an astronomical object to mark a location on the surface of the earth appears in E. W. Maunder's article on the "Star of the Magi" in the International Standard Bible Encyclopedia (1939) 5:2849, under the section "The Legend of the Well." Maunder reports a legend at Bethlehem that one of the Magi, looking down into the well at the inn, saw the star reflected in it (so directly overhead) and knew they had come to the right place. This is a clever suggestion, but it hardly does justice to the details of the text of Matthew.
 

Supernatural models, by contrast, handle this problem easily. An object which is only a few tens of feet above the earth can easily guide people by moving at a speed they can keep up with and by stopping over the location to which the people are being led. In fact, the terminology used in the ancient Greek translation of the Old Testament to describe the guidance provided by the pillar of fire and cloud is rather close to that used by Matthew here. Matthew uses the verb proago to indicate that the star "went before" them or "led" them, the verb erchomai to indicate its "moving" or "coming" and histemi to indicate its "stopping" or "standing." In Exodus 13:21 God "leads" them (ago) by a pillar of cloud to show Israel the way. In Ex 14:19 the angel of God "goes before" them (proerchomai), and to protect them from the Egyptians, he moves the pillar of cloud to "stand" (histemi) behind them. In Ex 33:9, whenever Moses would enter the tent of meeting, the pillar of cloud would descend and "stand" (histemi) at the door of the tent, and when the people saw the pillar "standing" (histemi), they would each stand at the door of their own tents and worship. In Numbers 14:14, Moses tells God that the Egyptians have already heard that God's cloud "stands over" Israel (ephistemi) and that by or in the pillar of cloud He "goes before" them (poreuomai proteros). The verb poreuomai is pretty much a synonym of erchomai. In Deuteronomy 1:33, Moses rebukes Israel for not trusting God who "went before" (proporeuomai proteros) them in fire by night and cloud by day to search out places to camp and show them the way they should go.
 

So. With a purely natural model, no ordinary reading of the text of Matthew or the phenomena of conjunctions will guide the Magi to Jesus. In a purely supernatural model, this is handled easily, but the events described above in our section "The Natural Part" are just an astonishing (but meaningless) fluke. But how about a hybrid model? The natural part is open to retrospective calculation, while the supernatural part fits the Matthew narrative naturally and gets the Magi to the right house in Bethlehem. Such a natural-supernatural hybrid might also make sense of the extreme joy which the Magi experience when they see the star again.
 

Matthew 2:9-10 tells us that they again saw the star which they had seen in the east, so there is a strong indication that we are to identify the star they now see with that which sent them on their way. In the model we propose, if they arrive in Jerusalem on or about 28 August of 2 BC, then the cluster of Sun, Moon, Mercury, Venus, Mars and Jupiter in Leo-Virgo will set with the Sun; in fact the Sun, being the easternmost member of this group, will set last. So if we imagine the Magi coming out of Jerusalem as or shortly after the Sun sets, we can construct a scenario that would work very well.
 

Suppose (we speculate) they see a supernatural light about the size and brightness of the Jupiter-Venus conjunction of 17 June. It arises in the west from where the Sun has just set and comes toward them until it is low overhead. The "star" then turns southward and leads them (like the pillar in the wilderness) until it stops over the house where Joseph, Mary and the child Jesus are now living. After all their travels, wouldn't that be something to make the Magi "rejoice exceedingly with great joy" (a literal approximation to the Greek)?
 

That is my proposal for a reconstruction of the phenomenon of the Christmas star.
 
 

Carroll, Susan S. "The Star of Bethlehem: an Astronomical and Historical Perspective," 1998 <http://sciastro.net/portia/articles/thestar.htm> accessed 26 Sept 2000.

Chester, Craig. "The Star of Bethlehem," Imprimis (December, 1993): 1-4.

Martin, Ernest L. The Star That Astonished the World. Portland, OR: ASK Publications, 1991.

Maunder, E. W. "Star of the Magi," International Standard Bible Encyclopedia (1939).

Sinnott, Roger. "Thoughts on the Star of Bethlehem," Sky and Telescope (December 1968): 384-386.

Tuckerman, Bryant. Planetary, Lunar, and Solar Positions 601 B.C. to A.D. 1. Philadelphia: American Philosophical Society, 1962.

My simulations of the sky were made using the program SkyGlobe 3.6 for DOS, copyright 1993, Klass M Software.

My calculations for the frequency of conjunctions of Jupiter and Venus were made using my own program FCONJ, written in QBasic 4.5.
 
 

Calculating the Frequency of Close Conjunction:
(Astronomical data from World Almanac '97: 455)
 

The Jupiter Venus separation of our conjunction is variously estimated as 3' (Sinnott), 0.5' (Martin), or 6" = 0.1' (Carroll). This will give us a range of frequencies.
 

The inclinations of orbits of these planets to the ecliptic:
 

I(Jupiter) = 1deg 18' 17" = 78' I(Venus) = 3deg 23' 41" = 204'
 

so the separation of Jupiter and Venus at a conjunction can range from 0' to 282'. Assuming an equal probability for each separation, the fraction of conjunctions as close or closer than ours is:
 

F(Sinnott) = 3'/282' = 1/100

F(Martin) = .5'/282' = 1/600

F(Carroll) =.1'/282' = 1/3000
 

How often do Jupiter and Venus have a conjunction? I attempted in several ways to make a diagram from which a simple calculation could be made, but did not get consistent results. So I tried two alternatives. (1) I used my computer planetarium program to look at the number of conjunctions in the past ten years, and found 12 or 13 (one case being too close to call), for an average of 1.25 per year. But this is not long enough to give an accurate average. (2) I set up a computer simulation in which three planets travel in circular orbits around the Sun at the average distances and speeds of Venus, Earth and Jupiter, respectively, and counted the number of times Jupiter and Venus would be in the same direction from Earth. This was run for 100,000 days (about 275 years), and gave an average of about 1.04 per year. I believe this second method is closer to the correct value, as the number of conjunctions seems to fluctuate rather substantially over shorter periods of time. Let us use 1.04 per year.
 

But only 1/100 of these conjunction are close enough to match or better the separation calculated by Sinnott, or 1/600 for Martin's, or 1/3000 for Carroll's. In addition, to have the conjunction occur in the proper constellation, there is only one chance in 12, so frequency f of conjunctions is:
 

f(Sinnott) = (1.04)(1/100)(1/12) = 1/1154, one conjunction in every 1154 years.

f(Martin) = (1.04)(1/600)(1/12) = 1/6923, one in every 6923 years.

f(Carroll) = (1.04)(1/3000)(1/12) =1/34615, one in every 34,615 years.
 

Add to this the requirement for a triple conjunction of Jupiter with Regulus. My computer planetarium program (SkyGlobe 3.6) shows that Jupiter has a "loop" at opposition of about 11 degrees, so to have this loop intersect Regulus has a probability of 11 deg/360 deg, or .03.

This decreases the frequency of the event we are describing to:
 

f(Sinnott) = (1/1154)(.03) = 1/38082, one such in every 38,082 years

f(Martin) = (1/6923)(.03) = 1/228492, one such in every 228,492 years

f(Carroll) = (1/34615)(.03) = 1/1142460, one such in every 1,142,460 years.
 
 

Travel-Time for the Magi:
 

In Ezra 7:9 we have information on the travel of Ezra from Babylon to Jerusalem in the fifth century BC. According to the conversion of dates recorded in the New Living Translation, Ezra left Babylon on April 8 and arrived in Jerusalem August 4. This is just under four months or about 120 days.
 

Applying this travel-time to the Magi, if the close conjunction of June 17, 2 BC was the event that moved them to journey to Jerusalem, then they would have set out sometime after this date. Traveling at about the same speed as Ezra's entourage, they would have arrived no earlier than mid-October.
 

To get Martin's arrival date of December 25, we must assume that they spent over two months longer than this, either in preparation for departure or in travel. This is possible, but seems rather long.
 

But Ezra's group was quite large, and so moved rather slowly. The Magi, by contrast, may have felt the need to hurry. If they averaged 10 miles per day, they could have traveled 700 miles by the 26 Aug conjunction; if they averaged 15 miles per day, 1050 miles; if 20 miles per day, 1400 miles. From the Macmillan Bible Atlas map #243, it looks like the total distance from Babylon to Jerusalem (up the Euphrates and across to Palmyra, then via Damascus) is about 3 inches on the map, or about 760 miles using the scale given there. This is doubtless somewhat of an underestimate, as the roads are not as straight as a ruler, but it looks like the Magi could easily have made the trip in 70 days if they averaged between 10 and 15 miles per day, not at all difficult if they were riding horses or camels.
 
 

Figure 1: The constellation Leo (the Lion), showing the location of its chief star Regulus and the two close Jupiter-Venus conjunctions of 12 August 3 BC and 17 June 2 BC. Adapted from a figure in Roger Sinnott's article "Thoughts on the Star of Bethlehem," Sky and Telescope (December 1968): 384-386.

Figure 2: The Zodiac in the vicinity of Leo (the Lion) and Virgo (the Virgin), showing the location of the Sun and Moon in early September 3 BC. Martin identifies 11 September 3 BC as the date of the birth of Jesus. I see it as the date of Mary's conception, and thus the incarnation of Jesus. Figure from Ernest L. Martin, The Star That Astonished the World (Portland, OR: ASK Publications, 1991).

Figures 3-7: Star charts of the various conjunctions as viewed from Jerusalem, reconstructed by the planetarium program SkyGlobe 3.6. Planet names are given in three-letter abbreviations (SUN, MOO, MER, VEN, MAR, JUP), as are the constellation names (LEO, VIR).
 
 

  Return to the IBRI Home Page

Last updated: January 14, 2002