It’s All Computus To Me!
- EPOCH
- 14 hours ago
- 9 min read
Rhiannon Warren | Trinity Hall, University of Cambridge
I knew before that this was the true celebration of Easter, but now I so fully learn the reason for observing this time, that I seem in all points to have known but little before concerning these matters. Therefore, I publicly declare and protest to you that are here present, that I will for ever observe this time of Easter, together with all my nation; and I do decree that this tonsure, which we have heard to be reasonable, shall be received by all clerks in my kingdom.” Without delay he accomplished by his royal authority what he had said. For straightway the Paschal cycles of nineteen years were sent by command of the state throughout all the provinces of the Picts to be transcribed, learned, and observed, the erroneous cycles of eighty-four years being everywhere blotted out.
– Bede, Historia Ecclesiastica Gentis Anglorum Book 5 Ch. XXI transl. A. M. Sellar
In the antepenultimate chapter of his Historia Ecclesiastica Gentis Anglorum, the Venerable Bede (c.627/3– 735) depicts the moment in 710 CE when Nechtan III, king of Pictland (what is now northern and eastern Scotland) accepted Abbot Ceolfrith’s lengthy arguments in favour of the nineteen year ‘Roman’ dating cycle over the eighty-four year ‘Irish’ cycle to calculate the date of Easter. This ‘science’ was computus: the calculation of Easter according to lunar cycles and, through this, the passage of time itself. Here, Nechtan’s proclamation results in significant, immediate action. Liturgical books were updated with the new computus tables, the new method of calculation taught, and clergy adopt the ‘Roman’ style tonsure: the monk’s ‘bald patch’ we are familiar with today over their previous, unknown style.
It may seem strange to the modern reader that Bede chose to close his history by recounting what is, on the face of things, a rather boring doctrinal debate, and resulting administrative changes. In fact, the proper keeping and dating of Easter were the subject of six complete chapters of Bede’s Historia, and the word ‘pascha’ (Easter) appears over two hundred times in the text. His treatises De Tempore (706) and De Temporum Ratione (c.716–725) were devoted to defending and codifying the date of the nineteen-year cycle. However, computus was not a preoccupation unique to Bede. The date of Easter was an important and contentious subject in the early medieval west. In Northumbria, the 664 Synod of Whitby, where King Oswiu accepted the 19-year cycle, was within living memory.
This event led Bede’s monastery Wearmouth-Jarrow to produce the Codex Amiatinus (Florence, Biblioteca Medicea Laurenziana, MS Amiatino 1), an ornate pandect (manuscript containing the full Bible) written and decorated in deliberate Roman style as a gift for Pope Gregory II to affirm their connection with Roman orthodoxy. In fact, the scriptoria did such a good job emulating continental uncial script (rather than eighth-century Britain’s typical Insular half uncial) and Roman decoration that the Codex Amiatinus was only recognised as the codex whose production is mentioned in the Vita Ceolfrithi, in 1888! Previously, the manuscript was deemed Italian.
Concern about the proper calculation of the date of Easter in the British Isles continued well beyond the Age of Bede. In the tenth century, Ælfric of Eynsham wrote De Temporibus Anni: an Old English handbook intended to practically help monks who were still confused about the complex calculations and explain the theological significance of the practice. Computus was a definite ‘hot-topic’ beyond the British Isles too. The Spanish theologian, Isidore of Seville (c.560–636), dedicated book five of his Etymologies, De legibus et temporibus (laws and times), to understanding the divine through calculations and relative temporality, noting in book three that calculating Easter helped explain scriptural passages. Many continental foundations encountered computus as a fundamental science as a result of their Irish founders. The ‘Bobbio Computus’ (Milan, Biblioteca Ambrosiana, H 150 inf) containing Irish and Frankish texts was produced at Bobio, a monastery founded by the Irish St Columbanus, in the ninth century. Similarly, in the eleventh century, Irish hermit monk, Marianus Scottus of Mainz used computus to argue Bede’s calculations of the age of the world were wrong by twenty-two years, using a deliberately complex paschal table to support his claim visually in the (partially autograph) manuscript Vatican, Bibliotheca Apostolica, MS Pal. Lat. 830.
But why was the date of Easter such a contentious problem? Why were the Northumbrians and the Picts following ‘the erroneous cycles of eighty-four years’ in the first place? Put simply, the date of Easter mattered because it was the most important feast day in the Christian year. The dispute over dating Easter properly stemmed from the difficulty of reconciling the solar Julian calendar (the widespread calendar in western Europe its 1582 Gregorian replacement) with the biblical lunar Hebrew calendar. The biblical Passion account is broadly internally consistent: the four gospels agree Jesus died during the Jewish Passover, celebrated on a fixed date on the 14th and 15th of Nisan, and that he died the day before the Sabbath and was resurrected on Sunday. Yet chronological disagreements still persisted: the Synoptic Gospels and John disagree on the exact lunar date of Christ’s death with the former claiming Jesus died on the Jewish Pascha/ Luna xv and the latter on Luna xiv. The early Christian author Tertullian (c.155–220) took his date of Luna xiv (the 25th of March) from John’s Gospel, forming the basis of the ‘Roman’ nineteen-year cycle. Hippolytus of Rome (c.165– 235) also favoured John’s Gospel, whereas Theophilus of Alexandrina’s (c.350–412) prologue to Emperor Theodosius made explicit an argument for Luna xv.
The first computistical work to date events concerning Creation, by Julius Africanus (c.160–240), does so by framing events in a chronology relative to Annus mundi (AM, ‘the year of the world’). It renders the birth of Christ as 5500 AM (and this as the start of the sixth age), and the Passion as 5531 AM – locating this relative to ‘real world’ events as the sixteenth regnal year of Emperor Tiberius. These dates were revised by Hippolytus of Rome, who placed Christ’s Passion in 5532 AM, and in Eusebius of Caesera’s (c.260/5–339) Chronicon, which placed Jesus’ birth in AM 5199 at 2015 years after Abraham’s birth.
Most influential to the early medieval understanding and construction of a chronology was Dionysius Exiguus’ (c. 470–544) introduction of Annus Domini (AD, ‘the year of the lord’) as a dating system in his Easter cycle of 525 CE, covering ninety-five years. This was based on a computistical table by pseudo-Cyril (an unknown author who’s work circulated under Cyril of Alexandria’s name) for 526–31 CE inclusive, inserting an additional nineteen years and changing the basis of the dating system from the years passed since the reign of Emperor Diocletian to years since the birth of Christ. In the medieval period, computistical works presenting the chronological passage of time tended to take their mode of measuring chronology and relative passage of time from Dionysius Exiguus’ model, though this (again) was not without challenges. In De Tempore Ratione, Bede extended Dionysius’ Easter table, but calculated Christ’s birth at 3952 AM and his death at Luna xiv 3985 AM, based on the Masoretic text of the Pentateuch – the authoritative Hebrew text of the first five books of the Bible. While this significant temporal shift required some defence in De Tempore Ratione, Bedan computus tables were widely accepted in the British Isles and the Carolingian Empire (which, at its height covered present-day France, Belgium, Netherlands, Switzerland, Germany, northern Italy, and parts of Spain). This table then spread to Scandinavia following Denmark’s conversion in c.965.
Given all of this debate about Easter itself and relative temporality, it is hardly surprising that over nine thousand surviving medieval manuscripts, dating from the sixth to sixteenth centuries, contain some form of computistical material. Many ecclesiastics developed memory aids to simplify the process: London, British Library MS Egerton 3314 fol 9r-13v consists of mnemonics and 73r-75v diagrams of hands, known as computus digitorum (‘finger reckoning’) to help with calculations. Many of these manuscripts are liturgical books, which contain computistical diagrams alongside calendars outlining the church year’s feasts and fasts. These diagrams served an important, practical function: informing the community when Easter should be celebrated each year, and therefore how to adapt the temporale and sanctorale: the churches’ cycles of movable feasts and saint’s feast days. In this context it is not surprising that the oldest surviving manuscript from medieval Iceland, AM 732 a VII 4to, is a computus table, written around one hundred years after Iceland’s conversion to Christianity in c.1000 CE - the amount of time it takes for the infostructure to develop for native book production. Computistical treatises continued to circulate widely throughout the medieval period accompanied by ever more elaborate diagrams. The early twelfth-century Thorney Computus (Oxford, St John’s College MS 17) contains a plethora of striking diagrams of the heavens with tables of calculations and dates. This was a book meant more to impress, than to be used practically.
The so-called Great Paschal table was the form of computus diagram most widely used to check the date Easter Sunday was supposed to be celebrated on. It consisted of a table of nineteen columns, corresponding with the nineteen years of the lunar cycle, and twenty-eight rows, corresponding to the twenty-eight years of the solar cycle. The resulting 523 ‘boxes’ represent the date of Easter for each year. To calculate the date of Easter the reader would first have to find which year of the nineteen-year cycle they were in by adding one to the number of the year they were in and then dividing by nineteen. The remainder would give you the number for your year known as the ‘golden number’. They would then have to calculate the ‘dominical letter’ to tell them which days were Sundays by adding nine to the year and dividing by twenty-eight. Again, the remainder informs you which letter on the diagram to look for: count down the rows according to this. To find the date of Easter they would then find the ‘golden number’ on the vertical axis and ‘dominical letter’ on the horizontal axis. The square where these two figures met provided the exact date for Easter Sunday, often with a symbol or letter that would have to be checked against tables on the following folios.
Using an Icelandic example, the fourteenth-century fragmentary calendar AM 249 d fol, we can see this work in practice, and just how far the Great Paschal Calendar spread by the late medieval period.
Say the year is 1361 CE. We are therefore in the thirteenth year of the cycle:
1361 + 1 = 1362
1362/ 19 = 71 remainder 13
Our dominical letter here is C because:
1361+ 9 = 1370
1370/ 28 = 48 remainder 26
Reading down the thirteenth vertical column and across the twenty-sixth horizontal column the date of Easter Sunday is given by the insignia .B: in this case corresponding to the 5th of April – the same as it was in the western tradition this year!
The validity of Easter’s calculation was a matter of intensely debated theological importance. Across the medieval west, computus became both a practical tool and a cultural marker, linking the peripheries of Christendom visually with the Catholic Church in Rome – whether that be Nechtan III’s command for the Picts to adopt a new style of tonsure or Bede’s own Wearmouth-Jarrow producing the Codex Amiatinus in the style of a Roman pandect . The calculation itself was a complex, often contested topic that required a lot of specialist teaching to understand. For many medieval people, these texts and diagrams were as much ‘all computus to me’ as they are for the vast majority of us today!
Further Reading:
Bender, John, and David E. Wellbery, eds, Chronotypes: The Construction of Time (Stanford: Stanford University Press, 1991)
Coleman, Janet, Ancient and Medieval Memory: Studies in the Reconstruction of the Past (Cambridge: Cambridge University Press, 2010)
MacCarron, Máirín, Bede and Time: Computus, Theology and History in the Early Medieval World (London: Routledge, 2020)
Verbist, Peter, Duelling with the Past: Medieval Authors and the Problem of the Christian Era c. 990–1135 (Turnhout: Brepols, 2010)
Warntjes, Immo, and Dáibhí Ó Cróinín, eds, Computus and its Cultural Contexts in the Latin West, AD 300–1200: Proceedings of the 1st International Conference on the Science of Computus in Ireland and Europe, Galway, 14–16 July 2006 (Turnhout: Brepols, 2010)
Warnties, Immo and Dáibhi Ó Crónín, eds. The Easter Controversy of Late Antiquity and the Early Middle Ages: Its Manuscripts, Texts and Tables: Proceedings of the 2nd International Conference on the Science of Computus, Galway, 18–20 July 2008 (Turnhout: Brepols, 2012)
Rhiannon Warren is a PhD Candidate in the Department of Anglo-Saxon, Norse, and Celtic at the University of Cambridge, funded by the AHRC OOC-DTP. Her research interests lie in book culture, palaeography, liturgy, and ecclesiastical history in the medieval west. Her doctoral research examines the use of liturgical books in Medieval Iceland, paying specific attention the physical use of manuscripts, office liturgy and survival contexts.