Hostname: page-component-797576ffbb-tx785 Total loading time: 0 Render date: 2023-12-05T13:25:53.679Z Has data issue: false Feature Flags: { "corePageComponentGetUserInfoFromSharedSession": true, "coreDisableEcommerce": false, "useRatesEcommerce": true } hasContentIssue false

Travelling Festivals in Late Antiquity: How Christmas Came to the Greek East

Published online by Cambridge University Press:  08 June 2023

University of Göttingen
Rights & Permissions [Opens in a new window]


This article argues that the liturgical tradition of celebrating Christmas on 25 December travelled from the Latin West to the Greek East at the behest of Theodosius I upon his arrival in Constantinople in AD 380. From there it made its way to Cappadocia, Pontus and Syrian Antioch by means of travelling clerics who belonged to a pro-Nicene network. The essay also makes the larger methodological point that in late antiquity liturgical traditions did not travel of their own accord; rather, they were often carried by networks of travelling bishops and ‘radiated out’ from major sees to minor ones.

Eusebius Prize Essay
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright © The Author(s), 2023. Published by Cambridge University Press

Unlike during the Liturgical Movement of the nineteenth and twentieth centuries, liturgies in late antiquity were not serendipitously discovered by scholars in long-lost books.Footnote 1 The movement of liturgical traditions, many of which were newly invented or in the midst of significant development during late antiquity, required human actors to carry them from one location to another. In this article, it is argued that bishops operated as the primary means of the transmission of liturgical traditions in the fourth and fifth centuries, as these clerics travelled to maintain bonds of friendship with other bishops within their networks. Just as participation in the liturgy is always an embodied act, so in late antiquity the movement of liturgical traditions required embodied actors to carry them from place to place.Footnote 2

However, if one examines the scholarship on the introduction of Christmas to the eastern half of the Roman Empire, it can almost seem as if it floated across the Mediterranean from the city of Rome. It is not at all clear what were the means – whether personal, social, institutional or material – of this transmission. Despite the ever-growing mass of scholarly literature devoted to the origins of Christmas,Footnote 3 many questions remain with respect to its transmission: who brought it from the Latin West to the Greek East? How was it established so quickly in certain places in the East, but not in others? Why, despite its contradiction of the local tradition of Epiphany, did some clerics so readily embrace it? Certain vague explanations have been provided: for example, it was introduced by Nicene clerics as an anti-Arian holiday. Nevertheless, this argument has rightly come under fire, as Arians and Nicenes alike held to the human birth of Jesus of Nazareth.Footnote 4

The establishment of Christmas in the East is particularly puzzling because, despite being established in the 380s and 390s in Constantinople, Antioch, Cappadocia and Pontus, it did not come to be celebrated in Palestine and Egypt until some decades later, probably between the Councils of Ephesus (431) and Chalcedon (451). Martin Connell is one of a few who attempt to provide an explanation for why Christmas arrived in some locations but not in others:

The places that were in contact with the Latin churches most frequently and that sought communion with Rome, like those in West Syria, were those that accepted the Roman date of Christmas earlier, as early as any non-Roman churches in the West. Other places, such as Jerusalem and Egypt, resisted for some time.Footnote 5

While ‘communion’ between Churches is potentially a helpful criterion for the reception of the festival, there is little discussion of how such communion functioned in late antique Christianity, or by what means it was established or maintained. Connell's account is further problematic because Alexandria and Rome did actually have very close relations during the fourth century and beyond. Martin Wallraff has also offered very brief comments on the adoption of Christmas in the East, to the effect that the festival was first introduced to ‘the Eastern half of the empire’ under the reign of Constantius ii (337–61), and gradually came to be accepted ‘beginning in Syria and Asia Minor’.Footnote 6 While Hans Förster has shown this account to be unlikely on other grounds,Footnote 7 it also provides no explanation of the uneven success of Christmas in the East in the late fourth century. Förster himself suggests that the holiday came over from the West in fits and spurts, first from Rome to one of Antioch's Nicene factions (that of Paulinus) and then to the rival Nicene faction (that of Meletius). Förster suggests that the widespread adoption of the holiday in the East ‘compelled’ (nötigte) the rival faction to begin celebrating it (though, again, Christmas was not ubiquitous in the late fourth century).Footnote 8 Förster thinks that after arriving in Antioch, Christmas came to Constantinople during Gregory of Nazianzus’ short-lived episcopacy, through the influence of the Western emperor Theodosius, who himself came from a family of devoted Nicene Christians.Footnote 9 While I will follow Förster's second suggestion, the first is less plausible. First, it imagines that a new holiday would be readily accepted by clerics when it came from a group with whom they had been at odds for decades. Second, it provides no explanation why Paulinus adopted the celebration of Christmas but other allies of Rome – namely the Alexandrians – did not. None of these explanations addresses why the holiday was adopted in some locations and not in others, and none supplies plausible accounts of the material and social means of its transmission.

In what follows, then, I provide an account of the transmission and non-transmission of this holiday, which falls into two parts. First, in agreement with Förster, I suggest that the celebration of Christmas was introduced from the West to Constantinople by the emperor Theodosius or those of his court when the emperor took up residence in the eastern capital in 380. Theodosius’ authority in the capital, coupled with his support for Nicene Christianity, made it easy for Gregory of Nazianzus and other Nicene bishops in the East to accept his liturgical authority, and thus to transmit the celebration of the feast to their allies. Second, I argue that the pre-established pro-Nicene episcopal network, which included Gregory and was headed by Meletius of Antioch and Basil of Caesarea, provided the means for the holiday to spread to other parts of the Greek East. Due to the strong ties of this particular Nicene ecclesiastical network, the celebration of the Nativity on 25 December spread quickly to Cappadocia under Gregory of Nyssa (Basil's brother) and to Antioch under the bishop Flavian and the priest John Chrysostom (both disciples of Meletius). Furthermore, while these network ties allowed this feast to be adopted so quickly in these locations, the boundaries of the network also served to limit the spread of the holiday. The much-delayed adoption of the festival in Palestine and Alexandria stems from the limes between one Nicene network and another, one surrounding Antioch and the other surrounding Alexandria. Following this account of the transmission of the tradition of Christmas in particular, I offer some thoughts on the transmission of liturgical traditions in late antiquity more broadly: namely that, like Christmas, they often spread by means of the travels of bishops, radiating out from major sees (for example, Constantinople) to minor ones.

A brief note about methods. This study relies in part on sociological theories concerning social networks, arguing especially that ‘strong ties’ among individuals within a network allowed the tradition of Christmas to travel. The term ‘strong ties’, coined by Mark Granovetter, has been used by many scholars of early Christianity.Footnote 10 Although Granovetter mostly worked to show that the widespread transmission of knowledge often occurs through ‘weak ties’, scholars of early Christianity – especially Elizabeth Clark – have shown that theological traditions in late antiquity often travelled by way of ‘strong ties’.Footnote 11 I propose that the celebration of Christmas likewise travelled through those who were related by strong bonds of friendship. However, behind this quantitative terminology also stands a qualitative assessment of the relationships within the network in question.

The introduction of Christmas to Constantinople

Although it has been the subject of some debate, Christmas appears to have been introduced into the Christian East around 379 or 380. The best evidence for its arrival comes from John Chrysostom's sermon On the day of Christ's birth, which he delivered in Antioch: ‘It has not even been ten years since this day became clear and well-known to us.’Footnote 12 While Chrysostom can ‘fudge’ numbers,Footnote 13 we should not assume that this claim is wildly inaccurate, not least because his audience would have known well the novelty of the festival. Because this sermon can be dated with some certainty to 387, the first celebration of Christmas in Antioch would have been at the very earliest nine years previous, in 378.Footnote 14 However, it seems that Chrysostom was rounding up (‘not even ten years’), and the feast was probably introduced a little later than this, in the early 380s. This date coincides closely with a reference made to the celebration of Christmas in Gregory of Nyssa's oration On his Brother Basil. This oration was delivered either on 1 January 381– the first anniversary of Basil's death – or, perhaps, on the same date in one of the following years.Footnote 15 This sermon thus gives some of the earliest evidence for the celebration of Christmas in the Greek East, either in 380 or a year or two later. Finally, Gregory of Nazianzus’ Oration 38, often titled On the theophany, was most likely delivered at Christmas during one of Gregory's winters in Constantinople, and therefore in either 379 or 380, with the latter being more likely.Footnote 16 Assuming that John Chrysostom's statement is accurate, Gregory of Nyssa's mention of Christmas and Gregory of Nazianzus’ Christmas sermon would appear to testify to celebrations of Christmas in Constantinople and Cappadocia around the introduction of Christmas to Antioch – namely, around 380.

Coming to a firm date for Gregory of Nazianzus’ oration is particularly important because scholars have often thought that this sermon was delivered upon the first celebration of Christmas in Constantinople. The evidence here is far from certain. Hermann Usener considered that an outburst against the restless audience in the middle of Gregory's sermon is an indication of the novelty of the feast;Footnote 17 and, following Usener and Bernard Botte, much has been made of Gregory's reference to himself as the feast's ἔξαρχος – that is, either its ‘originator’ or ‘leader’.Footnote 18 Despite this tenuous evidence, Thomas Talley has shown that there is good reason for thinking that the sermon was preached on either the first or the second celebration of Christmas in Constantinople. Because Christmas was so obviously a tradition of the western half of the Roman empire – and therefore Nicene in heritage – Talley thinks it ‘highly unlikely that the festival would have been adopted from Rome under an Arian emperor [i.e. Valens]’.Footnote 19 That is, Constantinople, which was decidedly not Nicene even during Gregory's tenure in the city, probably would not have celebrated the Western/Nicene holiday prior to the death of Valens. The introduction of Christmas to Constantinople at this time is further confirmed by the account of Epiphanius of Salamis, dating from the 370s, that the birth of Christ was celebrated in Constantinople on 6 January (and, by implication, not yet on 25 December).Footnote 20 It is therefore exceedingly likely that this holiday was first introduced to Constantinople after the accession of Theodosius in the East in 379.

While Christmas appears to have been introduced to Constantinople at this time, why or how it was introduced has not been adequately discussed. Theological reasons are often adduced: Gregory of Nazianzus himself introduced Christmas because the feast's theology is inherently Nicene. However, as others have shown, this is not an adequate reason for introducing the feast, as celebrating the birth of Christ is no more Nicene than it is Arian. And even if the holiday had associations with the Nicene West, this link was merely a traditional one, and some historical mechanism for the introduction of the holiday is still required. Although such mechanisms are often impossible to clarify, the first celebration of Christmas in the East (as witnessed to by Gregory of Nazianzus, Gregory of Nyssa and John Chrysostom) coincides with a monumental event that took place in 380, just a month before Christmas: the baptised, Western, Nicene emperor Theodosius arrived in Constantinople.Footnote 21

The chronology of Theodosius’ ascent to the purple and his arrival in Constantinople is well established. Appointed by Gratian as co-emperor on 19 January 379, Theodosius was already an accomplished general. Much of that year was consumed by a fraught war against the Goths in Illyricum, while his court was based in Thessalonica, with victory being proclaimed (if not in fact achieved) on 17 November 379. Three months later, on 28 February 380, Theodosius, along with Gratian, promulgated the famous Cunctos populos (the ‘Edict of Thessalonica’), adopting the Nicene Christianity of the city of Rome – the ‘Catholic’ Christianity of the Apostle Peter and Pope Damasus – as the religion of the empire. All other forms of teaching were ‘heresy’, and adherents of other forms of Christianity were expelled from the churches. It is pertinent that although the edict was probably by implication directed to the whole empire, it was explicitly addressed to the people of Constantinople. Theodosius’ Nicene commitments are confirmed yet again later in 380: having been grievously ill, and, thinking he was on his deathbed, he was baptised by the Nicene bishop of Thessalonica, Ascholius. When he unexpectedly recovered, he quickly made his way to Constantinople. Upon arriving there on 24 November 380, the emperor requested that the Arian bishop Demophilus embrace the homoousion and become a Nicene. When Demophilus declined, Theodosius drove him and his Arian flock out of the city on 26 November. The following day, Theodosius processed together with Gregory of Nazianzus to the imperial Church of the Holy Apostles, under armed guard, ostensibly (though non-canonically) installing Gregory as his chosen bishop of Constantinople. Thus, the Nicene Theodosius' arrival in Gregory's Constantinople coincides almost exactly with the first celebration of Christmas in the Greek East.Footnote 22

While there is no direct evidence that Theodosius or members of his court told Gregory to celebrate Christ's birth on 25 December, there are a couple of hints that this might have been the case.Footnote 23 First, in a sermon delivered before Theodosius’ entry into the city, Gregory appears to be aware of the emperor's imminent arrival and his plans to bring Nicene Christianity to the capital.Footnote 24 By whom was he informed? It seems likely that there was some sort of correspondence between the elite pro-Nicene circles of the emperor Theodosius and those of the bishop Gregory. Second, like Roman emperors before him, Theodosius appreciated the importance of public festivals.Footnote 25 It is under Theodosius that there is the first evidence of imperial legislation of Christian holidays.Footnote 26 And since Theodosius intentionally adopted a form of Nicene Christianity centred on the ancient capital, it seems likely that he would have been interested in importing Roman liturgical traditions – and one as important as the Nativity – to New Rome.Footnote 27 Christmas was a tradition belonging to the city of Rome, and, as John Chrysostom's sermon shows, everyone knew it.Footnote 28 Perhaps an expression of Theodosius’ piety, the introduction of Christmas was also a political symbol: to celebrate Christmas is to be a good Roman (Nicene) Christian.

The reasons that Gregory would choose to celebrate a holiday that was suggested, or ordered, by the emperor are, in part, obvious. Despite the emperor's forceful way with Gregory himself, he was turning out to be a fierce supporter of the Nicene cause, and – if he understood Greek at all – he certainly appears to have appreciated Gregory's theological and oratorical abilities. Gregory was not so sensitive to the dangers of imperial patronage as to be unaware of its benefits. Additionally, there are at least two other reasons for Gregory's sympathy with this new, Western emperor. First, Theodosius showed favour not only to Gregory himself, but also to his network: when Theodosius invited the bishops to a council (Constantinople 381), he chose Gregory's ally Meletius to preside. Second, Gregory had friends among the aristocracy who had been responsible for his success in the capital from the beginning. The most obvious of these was Gregory's first cousin Theodosia, who so warmly welcomed him to Constantinople and offered him her home (which Jean Bernardi rightly refers to as ‘un palais’) to use as his church: the ‘Anastasia’.Footnote 29 Although uncertain, it is quite likely that Theodosia was well connected in Constantinople, such that she and her family could recommend Gregory to the imperial court.Footnote 30

Where Christmas spread: a pro-Nicene network

Throughout the 380s and into the 390s there is evidence that the celebration of Christmas spread throughout Asia Minor and Syria – specifically to Antioch, Cappadocia and Pontus. In addition to Gregory of Nyssa's mention of the holiday in his oration On his brother Basil, there is also a sermon that he preached at Christmas. This sermon, On the birth of Christ, was probably delivered in 386, while the bishop was resident in Nyssa.Footnote 31 Asterius, bishop of Amaseia in Pontus from the 380s to 400s, also mentions Christmas in two of his sermons.Footnote 32 And there is also John Chrysostom's sermon On the day of Christ's birth, delivered in Antioch in 387, in which he refers to its recent introduction to the city.Footnote 33

These three clerics had significant connections with one another. The most obvious tie was between John Chrysostom and Gregory of Nyssa. As far as can be ascertained, the two never met; however, they both belonged to a network of ecclesiastics – particularly bishops – who were responsible for shaping an alliance of supporters of the homoousion throughout Syria and Asia Minor. At the head of this alliance stood two bishops: Basil of Caesarea and Meletius of Antioch. Other prominent figures included the bishops Diodore of Tarsus, Gregory of Nazianzus and Amphilochius of Iconium. Although some of these figures disagreed significantly on other theological issues – for example, Diodore and Gregory had different approaches to ChristologyFootnote 34 – their alliance was formed out of a common ‘moderate’ Nicene theology. That is, while the Trinitarian theology of the Cappadocians was in some ways a novel departure from Nicaea (having developed in debate with, among others, Eunomius), the alliance to which they belonged did not depend on this particular theology, but to a broader commitment to the symbol of Nicaea, which was often combined with sympathy for those who had previously preferred the language of homoiousion or homoion to denote the Son's relationship with the Father. While this is undoubtedly a simplification of the theological matters under debate, the attempt to straddle a position between the homoousion and what some (often Alexandrians) deemed to be an ‘Arian’ position is confirmed by Basil's early friendship with Eustathius of Sebaste,Footnote 35 Meletius’ early homoian leaningsFootnote 36 and Gregory of Nazianzus’ attempts not to upset the ‘Arians’ in Constantinople by instead attacking those with whom no ‘mainstream’ clerics agreed, namely the Eunomians.Footnote 37

John Chrysostom belonged firmly to this network. He was connected by way of two of his mentors, Meletius of Antioch and Diodore of Tarsus, both of whom were well acquainted with the Cappadocians, and who were politically active in attempting to solidify a Nicene position in the East.Footnote 38 Meletius and Basil were responsible for the placement of Gregory of Nazianzus in Constantinople, and Diodore was responsible for electing Gregory's more politically savvy successor Nectarius.Footnote 39 John Chrysostom himself was also the successor of Nectarius as bishop of Constantinople in 397 – and thus in a capital which was formed deeply by this pro-Nicene alliance. Likewise, the bishop Flavian, under whom Chrysostom served when he preached at Christmas, was Meletius’ successor and had already served as the administrator of Meletius’ Nicene community in Antioch during the bishop's exiles.Footnote 40 Naturally, Gregory of Nyssa was also well ensconced in this alliance, being introduced to it by his brother Basil, and being a friend and correspondent of Gregory of Nazianzus. The Nyssen was also already well acquainted with Meletius from the Council of Antioch in 379 and from Meletius’ exiles in Armenia, before they both attended the Council of Constantinople (381), over which Meletius briefly presided. Gregory of Nyssa was even chosen as the eulogist for Meletius, after the latter's untimely death at the beginning of the council.Footnote 41

Furthermore, although very little is known of Asterius, it is not hard to imagine that he was a part of this larger network of bishops: his surviving sermons show not only that he made use of Cappadocian texts, but also that he clearly supported the Cappadocian brand of anti-Eunomian theology.Footnote 42 And while geography is not decisive, Pontus was not far from Cappadocia. Thus, the celebration of Christmas on 25 December in the late fourth century spread within this close-knit network, with its demonstrably strong ties.

Strikingly, the celebration of Christmas in the East appears not to have travelled beyond this particular Nicene network. For, although Antioch is not too far from Palestine and Egypt, and there was indeed a fair amount of travel among these locations in the fourth and fifth centuries, this date for Christmas took many decades to be established in Palestine and Egypt. In a sermon delivered in 411 in Bethlehem, the presbyter Jerome notes that the locals did not celebrate this feast, and that Epiphany (6 January) continued to be the festival celebrated in honour of Christ's birth.Footnote 43 John Cassian, who later wrote of his travels in the last decades of the fourth century, also clearly states that Egyptian Christians celebrated Epiphany, and not the Western holiday of Christmas, in commemoration of the birth and baptism of Christ.Footnote 44 It seems to have taken several more decades until Christmas was celebrated in these regions.Footnote 45

Although it is not immediately evident why this date for the Feast of the Nativity did not take hold in Palestine or Egypt, the discontinuity of the Antiochene/Cappadocian Nicene network with the Alexandrian Nicene network provides a good explanation for Alexandria being very slow to adopt the date. Although there were connections between the Nicene network already described and other Nicenes throughout the Roman Empire, these were decidedly weak ties. Furthermore, any strong ties that Alexandria had with Antioch were not with the alliance of Meletius and Basil.Footnote 46 Rather, the Alexandrian hierarchy – along with that of Rome – had historically sided against Meletius and with the other Nicene bishops of Antioch (the ‘Eustathians’), including Meletius’ rival, Paulinus. Athanasius had taken the side of Paulinus as early as the Tome to the Antiochenes (362)Footnote 47 – albeit in a more conciliatory tone than usual – and Alexandrian bishops continued to do so through the Council of Constantinople, notably in opposition to Gregory of Nazianzus and the Antiochenes with whom he was connected.Footnote 48 There continued to be significant animosity between Antioch and Alexandria into the fifth century: Theophilus of Alexandria opposed John Chrysostom, beginning with his consecration as bishop of Constantinople and ending with his exile and death.Footnote 49 And the afterlife of this rivalry continued throughout the Christological controversies. Thus, while the network of Basil and Meletius accounts for the spread of the holiday, opposition between networks accounts for its limits.

The importance that networks held as the causes for the introduction of Christmas in some locations and not others has been hinted at before. Martin Connell understands ‘communion’ among Churches to be the deciding factor of the festival's spread, even if he fails to recognise the significance of this claim.Footnote 50 In contrast, Förster, who is very aware of the ecclesiastical dynamics in Antioch, has argued that the holiday travelled across communions or networks: introduced from Rome to Paulinus or his predecessors, Christmas was eventually adopted by their rivals, the party of Meletius.Footnote 51 This situation is unlikely not only because, intuitively, rival sects are unlikely to take over one another's novel holidays, but also because, as Nathalie Rambault has argued on the evidence of the Apostolic constitutions, the Meletians more often took over homoian than Eustathian traditions.Footnote 52 Thus, while Connell is right to note that holidays travel within communions, such communion exists not between places, but between people – that is, within personal networks. Geography is not decisive, since we also know that at times ‘good fences make good neighbours’. Rather, to understand the meaning of communion, we must look to the historical – personal, material, practical – means of establishing and maintaining networks of relationships.

Although it is frequently assumed that information travels almost of its own accord, through the ‘weak ties’ of mass media – and now social media – this was not how information travelled in antiquity. Rather, it relied on individuals establishing and maintaining the ‘strong ties’ of social and ecclesiastical networks. It is really unlikely that a novel (or disruptive) doctrine or ritual practice would travel through ‘weak ties’ and thus beyond the bounds of a network. This has been shown not least in the pioneering work of Elizabeth Clark, in which she demonstrated that it was more often one's network, rather than doctrine, which was the deciding factor for which side of a controversy one ended up on.Footnote 53 This explains why, despite the shared commitment to the symbol of Nicaea, the Alexandrians did not accept this new holiday: they were not a part of the network so strongly established among the allies of Basil and Meletius. In other words, although the holiday travelled within a Nicene network, it was not the theology of the holiday that drove the adoption of Christmas. Rather, the force of network ties broke through any resistance that might have been felt at the novelty of the holiday and its potential to supplant the pre-established holiday of Epiphany.Footnote 54

How Christmas spread: travelling bishops

The importance of networks for the transmission of this liturgical knowledge is reinforced by the simple fact that neither imperial nor ecclesiastical communication was at all centralised in late antiquity. As Claire Sotinel has shown, if one wanted to propagate one's idea of orthodoxy or orthopraxy, a substantial amount of personal effort was required. Writing on the topic of how information spread in Augustine of Hippo's circles, Sotinel states ‘Ideas did not circulate by themselves, even though ancient literature abounds in vague formulae defining (or rather failing to define) the channels through which they were transmitted. It is important to underline the extent to which circulation depended on the actions of individuals.’Footnote 55 Here Augustine's circle was not the exception, but the norm.Footnote 56 Things as important as conciliar decisions – and perhaps even something like Gratian's and Theodosius' Cunctos populous – were not ‘officially’ promulgated, but only travelled within networks, which were often very limited in geographical scope.Footnote 57

Because it is still possible to imagine that the celebration of Christmas flowed organically through the tentacles of this network from Gregory of Nazianzus to Gregory of Nyssa, Asterius of Amaseia and Flavian of Antioch, it is important to consider concretely how knowledge about Christmas travelled within the network. Like doctrine, liturgical traditions did not travel of their own accord. Rather, the celebration of Christmas, like other ideas and practices, was carried by clerics – especially bishops – throughout the nodes of the network. Indeed, from the beginning of the fourth century, bishops appear to have spent much of their time travelling.Footnote 58 Spurred on by Constantine's decision to permit them use of the cursus publicus, bishops travelled so often, and for such questionable reasons, that throughout the fourth century a series of councils attempted to limit their movement.Footnote 59 Nevertheless, bishops travelled not only from self-interest, but also to attend synods and councils (some more formal than others), consecrations of bishops and disciplinary hearings, among other things.Footnote 60 In these travels, bishops moved within preestablished networks, working to maintain relationships. As Carmen Cvetković writes, ‘Thanks to networks of hospitality … gifts were exchanged, relics were circulated, books and letters were delivered, news was disseminated.’Footnote 61 Like these other goods, Christmas did not travel of its own accord. Rather, those bishops already mentioned – and undoubtedly many whose names are not known – carried it with them throughout their networks as they travelled.

While the simple fact of the frequency of episcopal travel can inform us about the probable means of the movement of Christmas throughout this network, specific patterns of episcopal travel also confirm the likelihood of Christmas first coming to Constantinople and then radiating out from there. Most episcopal travel was undertaken by bishops of less significant sees to visit the bishops of more important cities. Since ecclesiastical centres were often also seats of political power, provincial bishops could both appeal to officials and strengthen connections with their more powerful episcopal brothers. Such a phenomenon is seen throughout the Roman empire in the fourth and fifth centuries. For example, the minutes of the Council of Serdica tell of the bishop of Thessalonica complaining that he receives too many episcopal visitorsFootnote 62 and Augustine was often away from Hippo Regius, visiting Carthage. More relevant to this issue is that Gregory of Nyssa was often away from his see, visiting Constantinople, while Palladius of Hierapolis makes mention of large gatherings of bishops in Constantinople during John Chrysostom's episcopacy (397–404), possibly precursors to the ‘endemic synods’ of which we hear several decades later.Footnote 63 Similarly, Severian, the bishop of the small Syrian city of Gabala, appears to have stayed in Constantinople for years at the pleasure of the imperial family but to the chagrin of Constantinople's bishop, John Chrysostom. Thus, like other political and ecclesiastical centres of power, Constantinople was a gathering place for bishops. And, naturally, when bishops returned to their sees, they would bring with them their newfound knowledge – of doctrine and of liturgy. Sometimes they would bring physical documents full of prayers and readings, and sometimes they carried knowledge in their memories. This is very likely how bishops in Syria, Cappadocia and Pontus brought Christmas back with them after visits to the capital.

That bishops in particular were responsible for the movement of liturgical traditions is exceedingly likely. Episcopal visits – which were often prolonged – included participation in local liturgical life. Although other clerics and monks travelled extensively (for example, Jerome and John Cassian), they were neither responsible for the maintenance of networks, nor authoritative in liturgical matters. Moreover, other forms of travel such as pilgrimage and visits to famous ascetics had fundamentally different goals than episcopal travel.Footnote 64 It was travelling bishops who had both the knowledge of liturgical traditions and the liturgical authority to introduce into their dioceses the holidays that they experienced while visiting cities such as Constantinople. This is most probably how the celebration of Christmas travelled throughout networks in the East in the last two decades of the fourth century.

Based on what is known of the circulation of information within Christian networks of the Mediterranean in the fourth and fifth centuries, the most plausible explanation for the arrival and spread of Christmas in the Greek East is that it was introduced to Constantinople and made its way through the Nicene network of Constantinople's bishop, Gregory of Nazianzus. Simple geographic proximity was not sufficient for Christmas to spread. Liturgical traditions are not contagious: they cannot spread of their own accord. Rather, individual agents, usually bishops, learned of Christmas through their visits to those within their networks and introduced it to their own flocks. Those who were outside these networks had neither the knowledge nor the impetus to celebrate the holiday. For not only were some ‘out of the loop’, but others also belonged to rival networks and were not inclined to share.

In the process of arguing for these specific means whereby Christmas spread, more has been demonstrated than the simple idea that Christmas did not travel of its own accord; it is now possible to generalise about the movement of liturgical traditions in late antiquity. First, they often spread by means of travelling bishops. Bishops had both the opportunity to learn about liturgical innovations and the authority to introduce them into their own jurisdictions. Second, holidays appear to have ‘radiated out’ within networks of bishops from major sees to minor ones. Since bishops from all over the empire – and sometimes beyond – came for prolonged visits to important cities such as Thessalonica, Carthage and Constantinople (to mention just a few cities), such cities became the sources of more widespread liturgical traditions. Bishops from Pontus who were visiting Constantinople could learn about its holidays and then begin celebrating them in their own dioceses. Analogous to Förster's larger argument that liturgical traditions radiated out from Jerusalem by means of returning pilgrimsFootnote 65 – which phenomenon is of course unique to that city – there is also a broader phenomenon throughout the empire of bishops travelling to metropoleis and bringing back home with them ‘souvenirs’: that is, holidays.


The article has benefited greatly from conversations and correspondence with Jeremiah Coogan, Carmen Cvetković and Dorothee Schenk. All mistakes remain my own. The research for this article was generously supported by the Alexander von Humboldt Foundation.


1 Joseph Ratzinger referred to this aspect of modern liturgical renewal as ‘archaeological enthusiasm’ in his preface to Reid, Alcuin, The organic development of the liturgy: the principles of liturgical reform and their relation to the twentieth-century liturgical movement prior to the Second Vatican Council, 2nd edn, San Francisco, Ca 2005Google Scholar.

2 On the centrality of embodiment in liturgical development, against an ‘evolutionary’ approach, see Belcher, Kimberly Hope, ‘Ritual systems, ritualized bodies, and the laws of liturgical development’, Studia Liturgica xlix (2019), 89110CrossRefGoogle Scholar.

3 See especially Susan K. Roll, Toward the origins of Christmas, Kampen 1995, and Förster, Hans, Die Anfänge von Weihnachten und Epiphanias: eine Anfrage an die Entstehungshypothesen, Tübingen 2007CrossRefGoogle Scholar. See also Nothaft, Philipp, ‘Early Christian chronology and the origins of the Christmas date: in defense of the “calculation theory”’, Questions Liturgiques/Studies in Liturgy xciv (2013), 247–65Google Scholar.

4 In support of the theological rationale see Roll, Toward the origins, esp. p. 165 and McArthur, A. Allan, The evolution of the Christian year, London 1953, 47–8Google Scholar.

5 Connell, Martin F., Eternity today: on the liturgical year, i, New York 2006, 163Google Scholar.

6 Wallraff, Martin, Christus verus sol: Sonnenverehrung und Christentum in der Spätantike, Münster 2001, 180Google Scholar.

7 Förster, Anfänge von Weihnachten, 184–7.

8 Ibid. 19, 174–9. Although Martin Wallraff appears to find this section convincing, he does not say why: review of Förster, Die Anfänge von Weihnachten, Gnomon lxxxii (2010), 342.

9 Förster, Anfänge von Weihnachten, 17, 190, 197–8 and passim. McArthur also links Theodosius to Christmas in Constantinople, but for theological reasons, and Roll follows him: McArthur, Evolution of the Christian year; Roll, Toward the origins, 190. See also Talley, Thomas J., The origins of the liturgical year, Collegeville, Mn 1991, 137–8Google Scholar.

10 Granovetter, Mark, ‘The strength of weak ties’, American Journal of Sociology lxxviii (1973), 1360–80CrossRefGoogle Scholar. See most recently Collar, Anna, ‘Strong ties, social networks, and the diffusion of ideas’, in Collar, Anna (ed.), Networks and the spread of ideas in the past: strong ties, innovation and knowledge exchange, London 2022, 127CrossRefGoogle Scholar, esp. pp. 8–9.

11 Clark, Elizabeth A., The Origenist controversy: the cultural construction of an early Christian debate, Princeton 1992, 18–9CrossRefGoogle Scholar. Peter Gemeinhardt, in contrast, sees much of Athanasius’ influence operating by means of ‘weak ties’: ‘Polemics and networking in fourth-century Trinitarian debates: Athanasius’ writings from his third exile revisited’, in Peter D'Hoine, Geert Roskam, Stefan Schorn and Joseph Verheyden (eds), Polemics and networking in Graeco-Roman antiquity, Turnhout 2021, 461–95, esp. p. 463.

12 οὔπω δέκατόν ἐστιν ἔτος, ἐξ οὗ δήλη καὶ γνώριμος ἡμῖν αὕτη ἡ ἡμέρα γεγένηται: In diem natalem 1, PG xlix. 351.27–8.

13 As he does when he narrates his own life, making his biography difficult to write. See, for example, Kelly, J. N. D., Golden mouth: the story of John Chrysostom: ascetic, preacher, bishop, Ithaca, NY 1995, 296–8Google Scholar.

14 His reference to a series of sermons that he preached against Jewish festivals – a clear reference to one of the series Adversus Judaeos – allows us to be very precise: In diem natalem 1, PG xlix. 351.27–8. With the main series being delivered in the autumn of 387, this Christmas sermon was most likely delivered at Christmas 387. It is also possible, but less likely, that John is referring to a shorter series of sermons against the Jews in autumn 386, in which case In diem natalem would date to 386 instead of 387: Pradels, Wendy, Brändle, Rudolf and Heimgartner, Martin, ‘The sequence and dating of the series of John Chrysostom's eight discourses Adversus Judaeos’, Zeitschrift für Antikes Christentum vi (2002), 90116CrossRefGoogle Scholar. Relatedly, some scholars, following Bernard Botte, have argued, based on John Chrysostom's mention in De Pentecoste of only three holidays – Epiphany, Easter and Pentecost – that De Pentecoste was preached before Christmas was introduced to Antioch: see especially Talley, Origins of the liturgical year, 135–6; Förster, Anfänge von Weihnachten. However, this argument fails to consider the rhetorical and exegetical contexts of Chrysostom's comment. In accordance with the Old Testament parallel of the Jews being commanded to meet ‘three times a year’, Chrysostom mentions only the three most important Christian holidays, and excludes the more novel festivals of Ascension and Christmas. Furthermore, Christmas seems to have been celebrated every year that Chrysostom was a priest (and thus preacher) in Antioch: he mentions Christmas in De beato Philogonio, which he delivered in 386, having been ordained earlier that year. Even in this sermon, Christmas does not seem to be entirely new to Antioch.

15 Daniélou, Jean, ‘Chronologie des sermons de Grégoire de Nyssa’, Revue des sciences religieuses xxix (1955), 351–3Google Scholar. Although Connell assumes that this encomium was delivered in the immediate aftermath of Basil's death, and therefore soon after 1 Jan. 379 (Eternity today, 161–2), this is untenable because (1) Gregory would have needed more time to craft an oration of this calibre; (2) there is a certain emotional distance to be observed in the oration; and (3) it was almost certainly delivered as a logos epitaphios, and therefore sometime after Basil's death, specifically on an anniversary of his death. Pierre Maraval has shown that the oration ‘is certainly pronounced [on] a [1 Jan.], but it does not seem possible to specify a year’: ‘Chronology of works’, in Lucas Francisco Mateo-Seco and Giulio Maspero (eds), The Brill dictionary of Gregory of Nyssa, Leiden 2010, 164–5.

16 While not certain, there is consensus that this sermon was delivered at Christmas (25 Dec.) rather than on Epiphany (6 Jan.). It is the first of three closely related sermons, all of which relate to Christ's birth or appearance, and the other two sermons (Oratio xxxix; Oratio xl) were clearly preached at the Epiphany vigil and on Epiphany respectively. Oratio xxxviii was therefore delivered sometime prior to 5 Jan. Although John McGuckin speculates that Oratio xxxviii was delivered as part of an Epiphany ‘triduum’ this is something otherwise unheard of in antiquity: Saint Gregory of Nazianzus: an intellectual biography, Crestwood, NY 2001, 336–7. The most reasonable conclusion is that it was preached two weeks earlier, in one of the first instances of the celebration of Christmas in the East. Furthermore, whereas Oratio xxxix and and Oratio xl are closely related – with the latter being the continuation of the former – Oratio xxxix simply recalls Oratio xxxviii: Grégoire de Nazianze: discours 38–41, ed. Claudio Moreschini and Paul Gallay, SC ccclviii, Paris 1990, 16. All scholars have admitted that it is impossible to be certain whether this sermon was preached in 379 or 380, at least since Justin Mossay, Les Fêtes de Noël et d’Épiphanie d'après les sources littéraires cappadociennes du IVe siècle, Louvain 1965, 9. Nevertheless, having weighed up the history of scholarship, Moreschini agrees with Jean Bernardi in preferring Christmas 380: Grégoire de Nazianze: discours, 22; Bernardi, Jean, La Prédication des pères Cappadociens: le prédicateur et son auditoire, Paris 1968, 199202Google Scholar. For the relevant bibliographies see Grégoire de Nazianze: discours, 16–22.

17 Usener, Hermann, Das Weihnachtsfest: Kapitel I bis III, Bonn 1889, 260Google Scholar.

18 Botte, Bernard, Les Origines de la Noël et de l'Epiphanie, Louvain 1932, 2730Google Scholar.

19 Talley, Origins of the liturgical year, 137–8.

20 Epiphanius, Adversus Haereses li.22.4. See Förster, Anfänge von Weihnachten, 181.

21 While Gregory himself might have introduced Christmas to Constantinople from his native Cappadocia, this is implausible, as in the 380s and 390s Christmas appears to have been rather new and precarious there as well: Mossay, Fêtes de Noël, 29–30. Strangely, it is usually Gregory who is mentioned as the originator of the feast, and not Theodosius. For example, McGuckin downplays Theodosius' role: ‘The arrival of Theodosius, a Spanish Christian, might also have accelerated the Eastern adoption of the date of December 25’: Saint Gregory of Nazianzus, 336–7.

22 In this paragraph I rely on the exemplary account of these years in Leppin, Hartmut, Theodosius der Grosse, Darmstadt 2003, esp. pp. 6674Google Scholar.

23 Other than Förster, Anfänge von Weihnachten, I can locate no suggestions that Theodosius was the one who introduced Christmas to Constantinople, but only that Gregory, under Theodosius, did so.

24 McGuckin takes Oratio xxxiii as evidence that Gregory knew of Theodosius’ plans to offer the Arian bishop, Demophilus, the episcopacy, if he opted to become Nicene: Saint Gregory of Nazianzus, 325.

25 Graf, Fritz, Roman festivals in the Greek East: from the early empire to the middle Byzantine era, Cambridge 2016, 105–27Google Scholar.

26 Codex Theodosianus ii.8.19.

27 Ibid. xvi.1.2.

28 See John Chrysostom, In diem natalem.

29 See Bernardi, Jean, ‘Nouvelles Perspectives sur la famille de Grégoire de Nazianze’, Vigiliae Christianae xxxviii (1984), 352–9CrossRefGoogle Scholar, esp. p. 354.

30 See McGuckin, Saint Gregory of Nazianzus, 238, following Bernardi, ‘Nouvelles Perspectives’. However, see an alternative assessment of Gregory's less significant standing in Constantinople in Neil McLynn, ‘The other Olympias: Gregory Nazianzen and the family of Vitalianus’, Zeitschrift für Antikes Christentum ii (1998), 227–46.

31 Daniélou, ‘Chronologie des sermons’, 367–8. Bernardi follows this dating: Prédication des pères Cappadociens, 290–1. However, Daniélou later dated the Christmas sermon to the year 382, in ‘La Chronologie des oeuvres de Grégoire de Nysse’, Studia Patristica vii (1966), 159–69. Maraval himself finds 386 the most acceptable, because of its thematic similarities to Gregory's two sermons on St Stephen: ‘Chronology of works’, 164–6. The second sermon In sanctum Stephanum can be securely dated to 27 December 386 (because of a coincidence of a feast dedicated to the Apostles), and based on thematic similarities among the three sermons it is inferred that In diem natalem and the first sermon In sanctum Stephanum were preached on the preceding days (25 and 26 December, respectively).

32 Asterius, Homiliae iv.3.3; xii.1.1. These sermons are impossible to date more precisely.

33 Additional evidence for the celebration of the Nativity on 25 December comes from the Apostolic constitutions, a compilation of liturgical material that derives from the party of Meletius, probably from around 380, but not much later than that: Les Constitutions apostoliques, i, ed. M. Metzger, SC cccxx, Paris 1985, 58–60.

34 Beeley, Christopher A., ‘The early Christological controversy: Apollinarius, Diodore, and Gregory Nazianzen’, Vigiliae Christianae lxv (2011), 376407CrossRefGoogle Scholar.

35 On Basil's relationship and subsequent falling-out with Eustathius of Sebaste see Philip Rousseau, Basil of Caesarea, Berkeley, Ca 1998, 73–82, 100–3, 139–244.

36 For the relevant sources and a discussion see Karmann, Thomas R., Meletius von Antiochien: Studien zur Geschichte des trinitätstheologischen Streits in den Jahren 360–364 n. Chr., Frankfurt 2009Google Scholar.

37 McGuckin, Saint Gregory of Nazianzus, 241.

38 On John's theological debt to the Cappadocian tradition, and perhaps to their writings, see Karmann, Thomas R., ‘Johannes Chrysostomus und der Neunizänismus: eine Spurensuche in ausgewählten Predigten des antiochenischen Presbyters’, Sacris Erudiri li (2012), 79107CrossRefGoogle Scholar; Lai, Pak-Wah, ‘The Eusebian and Meletian roots of John Chrysostom's Trinitarian theology’, Scrinium xiv (2018), 3762CrossRefGoogle Scholar, and ‘John Chrysostom's reception of Basil of Caesarea's Trinitarian theology’, Scrinium xv (2019), 62–78; and Edwards, Robert G. T., ‘Divine incomprehensibility and human faith in John Chrysostom’, Vigiliae Christianae lxxvi (2022), 434–62Google Scholar.

39 Sozomen, Historia ecclesiastica vii.8.

40 See, for example, Theodoret, Historia ecclesiastica iv.22; Theodoret, Philotheos historia viii.6–7.

41 See Gregory of Nyssa, Oratio funebris in Meletium episcopum.

42 On his use of Gregory of Nyssa's sermons see Cornelis Datema, Asterius of Amasea: homilies I-XIV: text, introduction and notes, Leiden 1970, pp. xxviii–xxxii. On his use of Cappadocian anti-Eunomian theology see especially Asterius, Homilia viii.14.2: Datema, Asterius of Amasea, 95–6.

43 Jerome, Homilia de nativitate Domini, CCSL lxxviii. 527. This is also indicated by the Armenian Lectionary, deriving from Jerusalem, probably from the episcopacy of Anastasius i (458–78): Méndez, Hugo, ‘Revising the date of the Armenian Lectionary of Jerusalem’, Journal of Early Christian Studies xxix (2021), 6192CrossRefGoogle Scholar. Although there is a major lacuna in Egeria's diary around Epiphany, it is likely that the eight-day celebration that she describes is the same one mentioned in the Armenian Lectionary.

44 John Cassian, Collationes x.2. This is corroborated by the Canons of Athanasius (latter half of the fourth century?), which knows only three feasts: Pentecost, Epiphany and Pascha: Canons of Athanasius, xvi: The canons of Athanasius of Alexandria, ed. Wilhelm Riedel and W. E. Crum, London 1904, 27.

45 On the likely mid-fifth century date of the Alexandrian adoption of 25 December see Talley, Origins of the liturgical year, 139.

46 For the early part of this split see Johannes Zachhuber, ‘The Antiochene synod of ad 363 and the beginnings of neo-Nicenism’, Zeitschrift für Antikes Christentum iv (2000), 83–101.

47 See Silke-Petra Bergjan, ‘Konkurrenz unter den Nizänern: die Christen Antiochiens im 4. Jahrhundert’, in Silke-Petra Bergjan and Susanna Elm (eds), Antioch II: the many faces of Antioch: intellectual exchange and religious diversity, CE 350–450, Tübingen 2018, 383–420.

48 Gregory of Nazianzus had his own ingenious plan, sadly doomed to failure: McGuckin, Saint Gregory of Nazianzus, 351–5.

49 Usener earlier recognised the animosity between Antioch and Alexandria as the basis for Christmas not spreading to Egypt until later: Weihnachtsfest, 329–31.

50 Connell, Eternity today, 163.

51 Förster, Anfänge von Weihnachten, 174–9.

52 Nathalie Rambault, ‘Pâques et l'Ascension au temps de Jean Chrysostome’, in Pascal Grégoire Delage (ed.), Jean Chrysostome: un évêque hors-contrôle: actes de la septième Petite Journée de Patristique 21 mars 2015 – Saintes, Royan 2015.

53 Clark, The Origenist controversy. See also Schor, Adam M., Theodoret's people: social networks and religious conflict in late Roman Syria, Berkeley, Ca 2010Google Scholar.

54 In keeping with Collar's sensitive analysis (‘Strong ties’), it is clear that it is not just the strength (quantity) of the ties that is important for the transmission of knowledge, but also the quality of the connections.

55 Claire Sotinel, ‘Augustine's information circuits’, in Mark Vessey (ed.), A companion to Augustine, Maldon 2012, 129.

56 See, for example, Carmen Angela Cvetković, ‘Niceta of Remesiana's visits to Nola: between sacred travel and political mission’, in Carmen Angela Cvetković and Peter Gemeinhardt (eds), Episcopal networks in late antiquity: connection and communication across boundaries, Berlin 2019, 179–200.

57 Sotinel, ‘Augustine's information circuits’, 127–8. On Cunctos populos see Neil McLynn, ‘“Genere Hispanus”: Theodosius, Spain and Nicene orthodoxy’, in Kim Bowes and Michael Kulikowski (eds), Hispania in late antiquity: current perspectives, Leiden 2005, 77–120.

58 Carmen Angela Cvetković, ‘Ecclesiastical travel and communication networks in late antiquity’, in Madalina Dana and Matthias Haake (eds), People of knowledge on the move: networks, connectivity and mobility in the ancient Mediterranean world from the early archaic period to late antiquity, Stuttgart (forthcoming). See also Leyerle, Blake, ‘Mobility and the traces of empire’, in Rousseau, Philip (ed.), A companion to late antiquity, Malden 2009, 110–23CrossRefGoogle Scholar.

59 See Cvetković, ‘Ecclesiastical travel’, and Claudia Rapp, Holy bishops in late antiquity: the nature of Christian leadership in an age of transition, Berkeley, Ca 2005, 265–76.

60 Cvetković, ‘Ecclesiastical travel’; Leyerle, ‘Mobility and traces of empire’, 112.

61 Cvetković, ‘Ecclesiastical travel’.

62 Hess, Hamilton, The early development of canon law and the Council of Serdica, Oxford 2002, 201–9CrossRefGoogle Scholar.

63 Throughout his Dialogus de vita Joannis Chrysostomi, Palladius of Hierapolis speaks of anything between 25 and 42 bishops who belong to the ‘synod of John’, who appear to have gathered with him at the episcopal residence on occasion: Dialogus viii. These were bishops of sees from Thrace (as far as Thessalonica) and throughout western Asia Minor. On ‘endemic synods’ see Joseph Hajjar, Le Synode permanent (ΣΥΝΟΔΟΣ ΕΝΔΗΜΟΥΣΑ) dans l’Église byzantine des origines au XIe siècle, Rome 1962, 52–80.

64 See, among others, Frank, Georgia, Memory of the eyes: pilgrims to living saints in Christian late antiquity, Berkeley, Ca 2000CrossRefGoogle Scholar; Maraval, Pierre, ‘The earliest phase of Christian pilgrimage in the Near East (before the 7th century)’, Dumbarton Oaks Papers lvi (2002), 6374CrossRefGoogle Scholar; and Bitton-Ashkelony, Brouria, Encountering the sacred: the debate on Christian pilgrimage in late antiquity, Berkeley, Ca 2005CrossRefGoogle Scholar.

65 Förster, Anfänge von Weihnachten, esp. pp. 306–8.