[832] Appendix N, Nos. (i), (ii).
[833] The most recent authorities are Tille, Y. and C. 119; H. Usener, Religionsgeschichtliche Untersuchungen, i, Das Weihnachtsfest (1889); L. Duchesne, Origines du Culte chrétien (ed. 2, 1898), 247, and in Bulletin critique (1890), 41; F. C. Conybeare, The History of Christmas, in American Journal of Theology (1899), iii. 1, and Introduction to The Key of Truth (1898); F. Cumont, Textes et Monuments mithraïques, i (1899), 342, 355. I have not been able to see an article praised by Mr. Conybeare, in P. de Lagarde, Mittheilungen (1890), iv. 241.
[834] Conybeare, Am. J. Th. iii. 7, cites, without giving exact references, two ‘north Italian homilies’ of the fourth century, which seem to show this.
[835] Sermo ccii (P. L. xxxviii. 1033).
[836] The depositio martyrum, attached to the Fasti of Philocalus drawn up in 354, opens with the entry ‘viii kl. ianu. natus Christus in Bethleem Iudeae.’ December 25 was therefore kept as the birthday at least as early as 353. Usener, i. 267, argued that the change must have taken place in this very year, because Liberius, while veiling Marcellina, the sister of St. Ambrose, on the Epiphany, spoke of the day as ‘natalem Sponsi tui’ (de Virginibus, iii. 1, in P. L. xvi. 219). But it is not proved either that this event took place in 363, or that it was on Epiphany rather than Christmas day. Liberius refers to the Marriage at Cana and the Feeding of the Five Thousand. But the first allusion is directly led up to by the sponsalia of Marcellina, and both events, although at a later date commemorated at Epiphany, may have belonged to Christmas at Rome, before Epiphany made its appearance (Duchesne, Bulletin critique (1890), 41). Usener adds that Liberius built the Basilica Liberii, also known as Sta. Maria ad Praesepe or Sta. Maria Maggiore, which is still a great station for the Christmas ceremonies, in honour of the new feast. But Duchesne shows that the dedication to St. Mary only dates from a rebuilding in the fifth century, that the praesepe cannot be traced there before the seventh, and that the original Christmas statio was at St. Peter’s.
[837] Duchesne, Bulletin critique (1890), 44. This document also belongs to the collection of Philocalus.
[838] Conybeare, Key of Truth, clii-clvii, quoting an Armenian bishop Hippolytus in Bodl. Armen. Marsh 467, f. 338a, ‘as many as were disobedient have divided the two feasts.’ According to the Catechism of the Syrian Doctors in the same MS., Sahak asked Afrem why the churches feast Dec. 25: the teacher replied, ‘The Roman world does so from idolatry, because of the worship of the Sun. And on the 25th of Dec., which is the first of Qanûn; when the day made a beginning out of the darkness they feasted the Sun with great joy, and declared that day to be the nuptials [? ‘natals,’ but cf. p. 241, n. 1] of the Sun. However, when the Son of God was born of the Virgin, they celebrated the same feast, although they had turned from their idols to God. And when their bishops (or primates) saw this, they proceeded to take the Feast of the Birth of Christ, which was on the sixth of January, and placed it there (viz. on Dec. 25). And they abrogated the feast of the Sun, because it (the Sun) was nothing, as we said before.’ Mommsen, C. I. L. i2. 338, quotes to the same effect another Scriptor Syrus (in Assemanus, Bibl. Orient. ii. 164): cf. p. 235. The early apologists (Tertullian, Apol. 16; ad Nationes, i. 13; Origen, contra Celsum, viii. 67) defend Christianity against pagan charges of Sun-worship.
[839] Conybeare, J. Am. Th. iii. 8.
[840] Most of these dates were in the spring (Duchesne, 247). As late as †243 the Pseudo-Cyprianic de Pascha computus gives March 28. On the other hand, December 25 is given early in the third century by Hippolytus, Comm. super Danielem, iv. 23 (p. 243, ed. Bonwetsch, 1897), although the text has been suspected of interpolation (Hilgenfeld, in Berlin. phil. Wochenschrift, 1897, p. 1324, s.). Ananias of Shirak (†600-50), Hom. de Nat. (transl. in Expositor, Nov. 1890), says that the followers of Cerinthus first separated the birth and baptism: cf. Conybeare, Key of Truth, cliv. This is further explained by Paul of Taron (ob. 1123), adv. Theopistum, 222 (quoted Conybeare, clvi), who says that Artemon calculated the dates of the Annunciation as March 25 and the Birth as December 25, ‘the birth, not however of the Divine Being, but only of the mere man.’ Both Cerinthus (end of 1st cent.) and Artemon (†202-17) appear to have held Adoptionist tenets: cf. Schaff, iv. 465, 574. Paul adds that Artemon calculated the dates from those for the conception and nativity of John the Baptist. This implies that St. John Baptist’s day was already June 24 by †200. It was traditional on that day by St. Augustine’s time, ‘Hoc maiorum traditione suscepimus’ (Sermo ccxcii. 1, in Migne, P. L. xxxviii. 1320). The six months’ interval between the two nativities may be inferred from St. Luke i. 26. St. Augustine refers to the symbolism of their relation to each other, and quotes with regard to their position on the solstices the words ascribed to the Baptist in St. John iii. 30 ‘illum oportet crescere, me autem minui’ (Sermo cxciv. 2; cclxxxvii. 3; cclxxxviii. 5; Migne, P. L. xxxviii. 1016, 1302, 1306). Duchesne, 250, conjectures that the varying dates of West (Dec. 25) and East (Jan. 6) depended on a similar variation in the date assigned to the Passion, it being assumed in each case that the life of Christ must have been a complete circle, and that therefore he must have died on the anniversary of his conception in the womb. Thus St. Augustine (in Heptat. ii. 90) upbraids the Jews, ‘non coques agnum in lacte matris suae.’ March 25 was widely accepted for the Passion from Tertullian onwards, and certain Montanists held to the date of April 6. Astronomy makes it impossible that March 25 can be historically correct, and therefore the whole calculation, if Duchesne is right, probably started from an arbitrary identification of a Christian date with the spring equinox, just as, if Ananias of Shirak is right, it started from a similar identification of another such date with the summer solstice. But it seems just as likely that the birth was fixed first, and the Annunciation and St. John Baptist’s day calculated back from that. If the Passion had been the starting-point, would not the feast of Christmas, as distinct from the traditional date for the event, have become a movable one?
[841] The Armenian criticism just quoted only re-echoes that put by St. Augustine in the mouth of the Manichaeans in Contra Faustum, xx. 4 (Corp. Script. Eccl. xxv) ‘Faustus dixit ... solemnes gentium dies cum ipsis celebratis ut Kalendas et solstitia.’ Augustine answers other criticisms of the same order in the course of the book, but he does not take up this one.