Article by William Bidelman - 1991
Share |

The Bimillenary of Christ's Birth:
The Astronomical Evidence

Dr. William P. Bidelman
Warner and Swasey Observatory
Case Western Reserve University
Cleveland, Ohio 44106

[reprinted from the Planetarian,September 1991]

Everyone knows that we are approaching the two-thousandthanniversary of the birth of Christ, but exactly when should it be celebrated?If we accept the conclusion of the sixth century monk Dionysius Exiguusthat he was born on December 25 of the year 1 B.C., then the bimillenarywould occur on Christmas Day of the year 2000 [if we ignore, for simplicity,the difference between the Julian and the Gregorian calendar]. However,it is widely believed that Dionysius' date is some four years too late.Is this correct? It is the writer's belief that the astronomical evidence,properly considered, indicates that Dionysius was considerably less in errorthan his critics have thought.

There have been several attempts to date the eventsin Christ's life by reference to astronomical phenomena. Since he was certainlyborn before the death of Herod the Great, much has been made of Josephus'statement that Herod died shortly after an eclipse of the moon visible fromJericho. Most chronologists, following the lead of the great Johann Kepler,have assumed that the relevant eclipse was that of March 13, of the year4 B.C. Yet it has been known for centuries that there were several othereclipses, one as late as January 10 of 1 B.C., that could also qualify (Mosley1987 and Sinnott 1986). Regrettably, it must be concluded that the astronomicalevidence on this point is inconclusive and that we must rely on other indicationsfor the date of Herod's death. And unfortunately, there is a substantialdifference of opinion on this point (Martin 1978).

A somewhat similar situation exists concerningthe date of the crucifixion. While the correlation between date of the monthand day of the week offers only two prime candidates (April 7 of A D. 30and April 3 of A.D. 33), the latter has been considered preferable becauseof a possible connection with a lunar eclipse supposedly seen from Jerusalemon that date. But Ruggles (1990) has recently shown that there is actuallyno astronomical reason to choose between two dates. And, in any case, thereis some uncertainty as to Christ's age at his death.

Another astronomical approach to the problem hasbeen to attempt to determine the nature and hopefully the time of appearanceof the "Christmas Star," the sight of which inspired the wisemen's visit to Israel (Hughes 1979). In this connection it is well to referto Matthew's exact words concerning this phenomenon. I quote from the 1961New English Bible: "After his birth astrologers from the east arrivedin Jerusalem, asking, `Where is the child who is born to be king of theJews? We observed the rising of his star, and we have come to pay him homage.'"And again: "They set out at the king's bidding; and the star whichthey had seen at its rising went ahead of them until it stopped above theplace where the child lay. At the sight of the star they were overjoyed."

There are at least two important points raisedby this account. First, the "star" apparently had meaning onlyfor the astrologers; no one else is reported to have seen it. And second,it is implied that the astrologers had lost sight of the "star"during their trip from their homeland (Babylon? Persia?) to Jerusalem, butthat they then saw it again on their way to Bethlehem.

The fact that the "star" was not generallyseen gives considerable credence to the idea that what the astrologers observedwas not a conventional star like a nova or a supernova but rather what maybe called a "sign in the heavens" such as a conjunction, or closeapproach, of two or more planets. And if indeed, a planetary conjunctionwas meant, it appears that the same planets were involved in two conjunctions,occurring several months apart.

Now the idea that planetary conjunctions are involvedin the "Christmas Star Problem" has ancient roots: Kepler, whilebelieving that the "star" was a newly-appeared celestial object,nevertheless thought that its coming was associated with several planetaryconjunctions involving Jupiter, Saturn and Mars (Burke-Gaffney 1937). Sometwo hundred years later the idea that planetary conjunctions alone wereinvolved became popular through the work of Ideler. One of the more noteworthysuggestions was that of Stockwell (1892) (actually of the writer's own institution!)that the "star" was a moderately close approach of Venus and Jupiteron May 8 of the year 6 B.C. Stockwell believed the death of Herod to haveoccurred in 4 B.C. and also considered only morning conjunctions. His suggestionwas roundly criticized on several grounds by Swift (1893) who, interestinglyenough, noted that Stockwell's conjunction could not have accounted forthe astrologers' second sighting. The solution to the problem came, in thewriter's view, from the discovery by Sinnott (1968) of two notable conjunctionsof Venus and Jupiter revealed in the extensive planetary tables of Tuckerman(1962). Both occurred in Leo. The first of these was a close morning conjunctionon August 12 of 3 B.C. which Mosley (1987) has shown to have been only about4.3 arcminutes in separation, while the second was an extremely close (0.5arcminutes) evening conjunction on June 17 of 2 B.C. Sinnott and othershave assumed that it was the second conjunction that led to the astrologers'trip to Jerusalem, but the writer prefers the scenario envisaged by Federer(1968), the longtime editor of Sky and Telescope magazine, who equated theten-months interval between the sightings with the travel time from theastrologers' homeland to Jerusalem.

If one accepts Federer's interpretation, then thedate of the astrologers' visit to Bethlehem can be precisely stated to havebeen June 17 of the year 2 B.C. This can be true, of course, only if a laterdate for the death of Herod can be shown to be valid. But even if this isso, are we now much closer to the date of Christ's birth? Unfortunatelywe are given no definite information on the age of the "child"when the astrologers arrived. It may be inferred, however, from Luke's accountof the parents' visit to Jerusalem to present the child to the Lord, whichcould not have occurred until at least 40 days after his birth, that thisvisit must have been made before the arrival of the astrologers, since theHoly Family apparently left for Egypt very soon after that time.

Consequently a reasonably accurate estimate of the time of the birth of Christ would be in March or April of the year 2 B.C., perhaps pure speculation around the time of the Lyrid meteor shower. It is difficult to put the birth much earlier in the year in view of Luke's testimony concerning the shepherds in the fields, which seems to imply that Christ was not born in the winter. It may be suggested, then, that the bimillenary birth date might be, say, April 1 of the year 1999.

Two closing comments: (l) the writer has recentlyshown (Observatory 111, 121) that pairs of morning and evening conjunctionssuch as those described by Sinnott are actually very common. In fact whenevera Venus-Jupiter morning conjunction of more than 19ยท elongation occurs,it is followed about ten months later by a similar evening one. Unknownboth to Stockwell and to Swift, an evening conjunction occurred on March9 of 5 B.C. While it would be attractive to those preferring an earlierdeath date of Herod to assume that the 6 and 5 B.C. conjunctions were involvedin the astrologers' visit rather than the later ones, this would put thebirth of Christ rather early in the year; also it is unlikely that Herodwould have been at his palace in Jerusalem in early March rather than inhis usual winter quarters in Jericho.

And (2) it appears probable to the writer thatthe reason that Luke's narrative of the Christmas story differs so significantlyfrom Matthew's (in not mentioning the astrologers, the "star,"or the flight to Egypt) may simply be that Luke was anxious to disassociateastrology completely from his account of the origin of Christianity, whereasMatthew was not so sensitive on that point. This question can best be leftto the theologians!


I am much indebted to Professor Eldon J. Epp of the Department of Religionof Case Western Reserve University for introducing me to some of the recentliterature on this subject, and to the University Library for making manyreferences available through interlibrary loan.


Burke-Gaffney, W. 1937, J. Roy. Astr. Soc. Canada, 31, 417.
Federer, C. A. 1968, Sky & Telescope, 36, 396.
Hughes, D. 1979, The Star of Bethlehem, (New York: Pocket Books).
Martin, E. L. 1978, The Birth of Christ Recalculated!, (Pasadena:Foundation for Biblical Research).
Mosley, J. 1987,
The ChristmasStar, (Los Angeles: Griffith Observatory).
Ruggles, C. 1990, Nature, 345, 669.
Sinnott, R. W. 1968, Sky & Telescope, 36, 384.
Sinnott, R. W. 1986, Sky & Telescope, 72, 632.
Stockwell, J. M. 1892, Astr. J., 12, 124.
Swift, J. 1893, Astr. & Astr., 12, 105.
Tuckerman, B. 1962, Planetary, Lunar, and Solar Positions 601 B.C. toA.D. 1, (Philadelphia: American Philosophical Society).

Dr. William Bidelman is now an Emeritus Professor following some 41 yearsof astronomical teaching and research at some of America's finest universities.He has worked mainly in stellar spectroscopy, and is perhaps best knownas the discoverer of the "phosphorous" and "mercury"stars. This is his first paper dealing with the solar system.

Reprinted from the Planetarian, Vol 20, #3, September 1991. Copyright1991 International Planetarium Society. For permission to reproduce pleasecontact Executive Editor, Sharon Shanks.

about ips | find a planetarium | officers | privacy policy

International Planetarium Society, Inc | c/o Ann Bragg, Treasurer | Marietta College, 215 Fifth Street | Marietta, Ohio 45750 USA