Why Is Christmas on December 25?
Part 4: What the Church Fathers Actually Say
Author: K. R. Harriman
I have already noted that there were differing traditions in the East and West about Jesus’s dates of death and birth (April 6 and January 6 on the one hand and March 25 and December 25 on the other), but the picture is even more complicated than that. The earliest record we have of anyone referring to the date of Jesus’s birth is in Clement of Alexandria’s Stromata 1.21, written around the turn of the second and third centuries (or at least Book 1 was, as indicated by his chronology ending with the death of Commodus). After presenting his long, complicated chronological calculations, which includes the calculation of the time from Adam to the death of Commodus to be 5,784 years, two months, and twelve days, he concludes that Jesus’s birth took place 194 years, one month, and thirteen days before the death of Commodus, which was on December 31, 192. We do not see here a theme that will later emerge of correlating creation and incarnation, but we see the peculiar interest in convoluted chronology among some teachers in the early church. And while this is not immediately obvious to us who do not have regular access to Alexandrian and Julian calendars from the time (hence why Clement refers to Egyptian names for months), Roland Bainton long ago calculated that Clement’s reckoning of the various dates in his chronology according to varying calendars would actually place Jesus’s birth at January 6, 2 BCE (“Basilidian Chronology and New Testament Interpretation,” 81–134, esp. 95–105). This was also the date (Tybri 11 in the Alexandrian calendar) on which some believed that John baptized Jesus, while others argued that it was four days later (Tybri 15 or January 10). According to Clement, these dates were correlated with Jesus’s birth because of Luke’s statement in 3:23 (with further information in 3:1 and 2) that Jesus began his ministry when he was about thirty years old (although this interpretation obviously requires ignoring the significance of ὡσεὶ, “about”). Others still placed his birth (or conception, as the term could refer to either) in the spring, such as in April 19 or 20 (Pharmuthi 24 or 25) or May 20 (Pachon 25). Only one of these dates that Clement lists — April 20 — was also considered by some to be the date of Jesus’s death. While only January 6 maintained acceptance long after Clement’s day, this text still indicates that there were Christian groups and offshoots like the Basilidians interested in chronological matters like identifying the dates of Jesus’s birth and death. It is impossible to say how widespread these interests were at this time, but at least in Alexandria — one of the great intellectual centers of the ancient world — around the end of the second century there were already multiple positions on these matters. We have no direct record of any particular celebration or festival as yet, except for the Basilidians celebrating the day of Jesus’s baptism, rather than his birth per se.
Only a couple of decades later in Rome we see further such chronological speculation attributed to Hippolytus. I say “attributed to” because there is controversy over the authorship of these texts, but their assignment to the first half of the third century is likely in any case. For the sake of simplicity, I will refer to the author as Hippolytus, although the argument does not depend on this identification. However, I will not be examining here the text in Book 4 of his commentary on Daniel for a few reasons: 1) we only have late copies of this text; 2) the copies we have present us with significant textual problems when it comes to his date for Jesus’s birth, particularly in terms of whether or not it was even present in the earliest manuscript; 3) even if it was not present in the earliest manuscript, we seem to have good reason for Hippolytus proposing December 25 as the date for Jesus’s birth anyway based on other precedents.
The first work of Hippolytus that is relevant here is his Canon, which is a lunar calendar written in 222 on a statue found in Rome (Thomas C. Schmidt, “Calculating December 25 as the Birth of Jesus in Hippolytus’ Canon and Chronicon,” 547). This was Hippolytus’s first recorded attempt to produce a chronology of Passovers for both the past and the future. But this proved difficult and, embarrassingly for Hippolytus, his error was literally written in stone. Indeed, the anonymous On Computing the Paschal Feast (De Pascha Computus; see below) written two decades later seems to have been aimed at correcting Hippolytus’s mistakes (Ibid., 548, 558). Still, this text is valuable for what it tells us of Hippolytus’s early chronology, such that he placed Jesus’s death on March 25 (the vernal equinox), 29 CE and his “genesis” on April 2, 2 BCE, which more likely means his conception rather than his birth (Ibid., 548–52: cf. Hippolytus, Refutation of All Heresies 4.3.5; Galen, On Semen 1.13.17; Clement of Alexandria, Eclogae Propheticae 50.2–3; Methodius of Olympus, Banquet of the Ten Virgins 2.1.31). It is possible that he could have conceived of a December 25 date at this time, but he does not say anything definite in this regard.
Hippolytus worked to avoid his errors in his Chronicon, written in approximately 235. This time, he restricts himself to talking about specific years as part of his larger chronology. His years for Jesus’s genesis and death still match what he claimed in the Canon when one remembers that the date he is calculating to is 235 and he places 206 years between Jesus’s death and the thirteenth year of Alexander Severus as well as thirty years between Jesus’s genesis and death (698–699). In two places, Hippolytus calculates the time from Adam to the current year to have been 5,738 years and the dates he assigns in both cases otherwise align (688, 700), but prior to the earlier reference he adds a note about nine additional months (686–687). In this case, it seems most likely that Hippolytus is adding nine months to arrive at Jesus’s birth rather than his conception (Ibid., 553–56). However, because he does not correlate precise dates in this work, it is hard to know on basis of the text itself if he regarded the dates of Jesus’s incarnation and death to be on the same day with a difference of nine months added for his birth. For this possibility, we must look to other evidence.
As noted above, an anonymous North African author (dubbed Pseudo-Cyprian) wrote On Computing the Paschal Feast in 243 as a correction of Hippolytus. In the process, the author notes that his predecessors regarded March 25, the date of the vernal equinox, as the day the world was created (4). Hippolytus was likely one of these predecessors, given the relationship of these documents. And by extension, he also contrasts himself with others about the first appearance of the moon (6), which he places on the fourth day of creation (March 28) while others placed it on the fifth day (March 29). Thomas Schmidt thus explains that one can calculate that Hippolytus concluded that the first Passover was on March 29 (Ibid., 559–60). This means that he placed the first day of creation on March 25. As such, when he adds the nine months in the aforementioned text, he would thus be placing Jesus’s birth on December 25. As for our North African author here, it seems he has placed Jesus’s birth and death on the same day: March 28 (Nothaft, “Early Christian Chronology and the Origins of the Christmas Date,” 252–54)
Another early teacher worth noting is Julius Africanus, a man known primarily for his Chronographiae, which exists now only in fragments. Like the aforementioned texts, this was an attempt to provide a chronological record to his present time, which was written, at the latest, in 222 (293 n. 5). He also places the first day of creation on March 25, the 29th of Phamenoth in the Egyptian calendar (F14b; 23, 25). He likewise linked this date to the day of Jesus’s advent/incarnation as being 5,500 years later, “announced” during the sovereignty of the Caesars (F15; 25). His later discrepancy of identifying Jesus’s death and resurrection as being 5,531 or 5,532 years after Adam (T93c–d; 289) is perhaps explicable if he placed the resurrection on March 25 as the mark of a new year in his chronology (Nothaft, “Early Christian Chronology,” 263–64). Unfortunately, Africanus nowhere says precisely what date he assigns to Jesus’s birth, but there is at least a decent probability that he could have dated Jesus’s birth to nine months later, on or around December 25.
Tertullian of Carthage, also writing in the early third century, does not directly present a date for Jesus’s birth. He is noted for being an early author to refer to Jesus’s death as being on March 25, eight days before the Kalends of April (Against the Jews 8.17–18). As for Jesus’s birth, he bases his dating for it on the prophecy of the 70 sevens in Daniel 9, which was also a key text in the chronologies of Julius Africanus and Clement of Alexandria as well as the later convoluted chronological work of Eusebius of Caesarea (Demonstration of the Gospel 8.2, which does not provide a month or day for Jesus’s birth). In this chronological schema, Christ’s birth was six months removed in the time of year from the destruction of Jerusalem on the 9th of Av (8.11–16). While precision is difficult here, not least because of the errors Tertullian makes in his calculations, it seems that Tertullian placed Jesus’s birth in winter, either in January or February (for more on Tertullian, see Ibid., 258).
Gregory Thaumaturgus, from Asia Minor, wrote a homily in the last half of the third century titled “Second Homily on the Annunciation to the Holy Virgin Mary.” In this homily, Gregory places the annunciation — and thus Jesus’s Incarnation/conception — in the month of the Passover, Nisan, since he identified it as the first month of the year, the first month of Mary’s pregnancy, and the sixth month of John the Baptist’s gestation. It is possible that he also linked the Incarnation with the Passover, unless he is citing the Passover to specify the month. In any case, this would at least roughly correspond with the time at which he died. This would also place Jesus’s birth in December or January.
A similar argument appears with Ephrem the Syrian, who wrote in the middle decades of the fourth century. In his Commentary on Exodus, he says that Jesus entered Mary’s womb on the 10th of Nisan. He is able to be more precise for two reasons. First, he plays on a well-established typology of Jesus and the Passover lamb, such that he thinks that the reference to the procurement of the lamb on the 10th of Nisan is a type for Jesus’s Incarnation (Exod 12:3). Second, he confirms this with the notion that John’s conception was on the tenth day of the seventh month of the Jewish calendar, when his father Zechariah was in the sanctuary (Luke 1:8–10). This is because of a view in the early church that Zechariah was high priest and that he was entering the sanctuary on this date because it was the Day of Atonement (cf. his Commentary on the Diatessaron 1.29; Protoevangelium of James 8.3; Ambrose, Exposition on the Gospel According to Luke 1.22; the aforementioned On the Solstices and Equinoxes; John Chrysostom, Homily on the Day of the Birth of Our Savior Jesus Christ 4–5; Nothaft, “Early Christian Chronology,” 258–61). In his Commentary on the Diatessaron, relying on these same arguments, he also specifies the date of Jesus’s birth in accordance with the Greek rendering, January 6 (1.29).
To return to a document cited earlier, On the Solstices and Equinoxes of the Conception and Birth of our Lord Jesus Christ and John the Baptist provides similar argumentation based on the same premise that Zechariah was serving in the temple as the high priest on the Day of Atonement. He also adds references to fasts in Zech 8:19 in the fourth, seventh, and tenth months (but not the fifth) that are occasions of great joy and he claims that John was conceived one day after the Day of Atonement, which he thought corresponded to the autumnal equinox in the Julian calendar (September 24). The months he uses from Zech 8 thus align with months of other significant dates. Starting with Nisan (which he equates to March), the month in which Jesus was announced according to this schema, the fourth month is June (summer solstice), the seventh month is September (autumnal equinox), and the tenth month is December (winter solstice). Since John was conceived on the autumnal equinox and he is six months older than Jesus, everything else in this schema falls into place. John was conceived on the autumnal equinox and was born on summer solstice, while Jesus was conceived on the vernal equinox and was born on the winter solstice, December 25. The author thus completes his correlation of these important dates with the four season-shifting astronomical events (for more on this text, see Ibid., 259–60; for a similar, but significantly later, argument, see Dionysius Exiguus, On Easter/The Paschal Cycle Argument 15). As noted above, our author also insists that because these dates of the seasons based on the position of the sun all serve as a testimony to Jesus, Jesus is likewise identified as the true sun, which would not have been unusual for early Christian texts (cf. Luke 1:78; Rev 22:16) and their use of Old Testament texts (Mal 4:2 is especially notable for its reference to the sun of righteousness).
Epiphanius of Salamis, in his compendium of heresies Panarion written in the late fourth century, presents us with a complicated picture. He notes that the Quartodecimans thought that Jesus died on the eighth day before the Kalends of April (March 25), but that others say it was the tenth day (March 23), and that he has concluded that it was the thirteenth day (March 20; 2.50.1.7–8; 51.26.1–4). Likewise, he condemns those who celebrate the birth of the Lord on the eighth day before the Kalends of January — which he polemically links with Saturnalia/Cronia/Cicellia — rather than the proper day of the eighth day before the Ides of January, thirteen days after the winter solstice (January 6; 2.51.22.3–6). This is clearly polemical to the extent that he is smearing his opponents and potentially misrepresenting at least one of his sources. While the non-extant text he quotes from Ephrem as support still comports with what we have already observed, the consul list he uses to support his claim for the proper birthdate of Christ actually dates it to the eighth day before the Kalends of January (December 25; for this source — preserved in the Monumenta Historiae Germanica, Auctores Antiquissimi IX — see page 218 here: https://www.dmgh.de/de/fs1/object/goToPage/bsb00000798.html?pageNo=218&sortIndex=010%3A010%3A0009%3A010%3A00%3A00&sort=score&order=desc&contextRows=10&contextStart=10&context=consularia+constantia&hl=false&fulltext=consularia+constantia). It is not out of the realm of possibility that an earlier version of this text said otherwise, especially if this was a point of controversy, but we have no evidence to suggest otherwise besides Epiphanius’s own polemics, which inclines me to think that that he is either misremembering (the more charitable and more likely reading) or intentionally misrepresenting his source. He clearly places himself among those of the Eastern tradition that celebrated the Nativity and the Epiphany on the same day of January 6 (2.51.22.12, 17, 29.7–30.3). He does not clearly correlate Jesus’s birth or annunciation with either his death or resurrection (2.51.27.4–6), but he still arrives at his conclusions on the bases of biblical interpretation, tradition/source work, and convoluted chronology.
Gregory of Nazianzus and John Chrysostom were contemporary Greek Fathers who accepted the December 25 designation for Jesus’s birthday. Gregory does not give us any sort of chronological argument to work with, but he seems to have been the first to preach in Constantinople in commemoration of the Christmas festival on December 25, 380 and acknowledged that day as Christmas (Oration 38.3; Beth Dunlop, Earliest Greek Patristic Orations on the Nativity, 13–14). I have already mentioned Chrysostom’s link of this date with the Western tradition (1) and his presentation of the argument based on the idea that Zechariah was the high priest (4–5). What is also notable is how he draws on sun imagery, particularly the “sun of righteousness/justice” image in Mal 4:2.
Augustine of Hippo, in his time and setting of the late fourth and early fifth centuries, can typically assume the December 25 date without offering a long chronological argument for it. He uses Jesus’s birth on December 25 and his annunciation and death on March 25 as part of a numerological argument rather than a directly chronological one (Trinity 4.5). In his sermons on the Nativity (Sermons 184–96 in Fathers of the Church vol. 38), he also assumes the establishment of this date and often draws on solar imagery, including the aforementioned sun of righteousness/justice.
Jerome, who worked in Judea around the same time Augustine was working, on the other hand, acknowledges that there are those in his day who place the Nativity and Epiphany on the same day — January 6, as we have seen in the early Eastern tradition (cf. the contemporary John Cassian, Conferences 10.2, who only mentions Egypt as celebrating in this fashion in contrast to the West). He is less hostile towards this group than Epiphanius was when the situation was reversed (he acknowledges that they both worship one Lord rather than calling the other group idolaters). But he still criticizes their view on three bases. First, he could at this point rely on the strength of the preceding tradition, claiming that it goes back to the earliest days. Second, he thinks that it is symbolically more appropriate for the Nativity to celebrate his generation and the Epiphany to celebrate his regeneration (since the day still retains its traditional association with his baptism). When these days are combined into one, the symbolism gets complicated as the day of Christ’s generation also becomes the day of his regeneration. Third, he points to the testimony of nature, as the celebration of this day on the winter solstice is followed by the increase in sunlight. This is fitting for the celebration of the Sun of Righteousness/Justice. He only briefly mentions a chronological point when he says that John the Baptist was born six months before Jesus (Homily 88). Since he does not expound on this point, presumably he is able to assume the aforementioned schema in which John the Baptist is born on or around the summer solstice. Otherwise, mentioning his birth as being six months before Jesus would not have been considered a decisive point in his favor.
Primary Sources
- Clement of Alexandria’s Stromata 1.21
- Hippolytus of Rome, Canon
- Hippolytus of Rome, Chronicon
- Hippolytus of Rome, Refutation of All Heresies 4.3.5
- Galen, On Semen 1.13.17
- Clement of Alexandria, Eclogae Propheticae 50.2–3
- Methodius of Olympus, Banquet of the Ten Virgins 2.1.31
- Pseudo-Cyprian, On Computing the Paschal Feast 4–6 (George Ogg, ed. and trans. The Pseudo-Cyprianic De Pascha Computus. London: SPCK, 1955.)
- Julius Africanus, Chronographiae (Chronographiae: The Extant Fragments. Edited by Martin
- Wallraff, Umberto Roberto, and Karl Pinggéra. Translated by William Adler. GCS 15. Berlin: de Gruyter, 2007.)
- Tertullian of Carthage, Against the Jews 8.11–18
- Eusebius of Caesarea, Demonstration of the Gospel 8.2
- Gregory Thaumaturgus, Second Homily on the Annunciation to the Holy Virgin Mary
- Ephrem the Syrian, Commentary on Exodus 12:3
- Ephrem the Syrian, Commentary on the Diatessaron 1.29
- Protoevangelium of James 8:3
- Ambrose, Exposition on the Gospel According to Luke 1.22
- On the Solstices and Equinoxes of the Conception and Birth of our Lord Jesus Christ and John the Baptist (available in Latin at: https://www.roger-pearse.com/weblog/wp-content/uploads/2020/01/Botte-Les_origines_de_la_noel_et_de_lepiphanie-1932-de_solstitia.pdf)
- Dionysius Exiguus, On Easter/The Paschal Cycle Argument 15
- Epiphanius, Panarion [2.]51
- Gregory of Nazianzus, Oration 38
- John Chrysostom, Homily on the Day of the Birth of Our Savior Jesus Christ
- Augustine of Hippo, Trinity 4.5
- Augustine of Hippo, Sermons
- John Cassian, Conferences 10.2
- Jerome, Homily 88
Secondary Sources
- Bainton, Roland H. “Basilidian Chronology and New Testament Interpretation.” JBL 42 (1923): 81–134.
- Dunlop, Beth. Earliest Greek Patristic Orations on the Nativity: A Study Including Translations. PhD Diss., Boston College, 2004). (available online: http://www.dec25th.info/pdf%20books/Dissertation--3122121.pdf)
- Nothaft, Carl Philipp Emanuel. “Early Christian Chronology and the Origins of the Christmas Date: In Defense of the Calculation Theory.” QL 94 (2013): 247–65. (available online on academia.edu)
- Schmidt, Thomas C. “Calculating December 25 as the Birth of Jesus in Hippolytus’ Canon and Chronicon.” VC 69 (2015): 542–63.
Read all 5 installments in K. R. Harrimans excellent series:
- Why Is Christmas on December 25?. Part 1: Scope, Stakes, and Saturnalia | by SAROLTA TATAR | Christianity — History and Culture | Dec, 2020 | Medium
- Why Is Christmas on December 25?. Part 2: The Sol Invictus Factoids | by SAROLTA TATAR | Christianity — History and Culture | Dec, 2020 | Medium
- Why Is Christmas on December 25?. Part 3: The Apologetic False Trail | by SAROLTA TATAR | Christianity — History and Culture | Dec, 2020 | Medium
- Why Is Christmas on December 25?. Part 5: Summary and Conclusions | by SAROLTA TATAR | Christianity — History and Culture | Dec, 2020 | Medium
