The Coligny Calendar is a large, 2nd Century bronze tablet that records the only known pre-Christian Celtic calendar. Its smashed remains were fatefully unearthed in 1897, in an unsuspecting field just North of Coligny, Ain. It did not take long for scholars to piece the artifact’s fragments together and begin poring over its remarkable contents. In 1926, Seymour de Ricci drew a precise facsimile which stands as the essential reference work for the calendar today.
For additional reference, a high-resolution photo of the Coligny Calendar can be found here.
(Note: some of the smaller and more isolated fragments in that display were placed in the wrong location, cf. Seymour de Ricci’s facsimile. Also note that this photo must be presumed to fall under copyright.)
The tablet records a lunisolar calendar such as those of Ancient Greece and that which continues to be followed devotedly in Hinduism to this day, pointing to a probable common source. The calendar likely would have played a critical role in the timing of festivals and holy days, as well as instructing on the auspicious and inauspicious times for other activities. As such, reconstructing the Coligny Calendar should be a top priority for Celtic Pagan praxis.
But as can be seen, the Coligny is badly damaged and survives only in fragments. Ultimately, any practical lunisolar calendar must be made complete enough, convenient enough, and ideally beautiful and automated enough, that any Celtic pagan regardless of their means or skill can and will use it, allowing them to break free from the inadequate Roman solar calendar typically used today.
Clearly there is much work to do. To lay one cornerstone towards this collective reconstruction project, Tegos Skrībbātous is contributing a transcription of the Coligny Calendar with its blank portions filled in as completely and plausibly as could be mustered:
In brief, the Coligny Calendar is a parapegma (pegs were inserted into the holes to keep track of time) that appears to record a cycle made up of five years plus two intercalary months. Each year is divided into twelve true months (cf. PIE *mḗh₁n̥s ‘moon, month’), “true” meaning they are actually defined by lunar phases rather than merely being subdivisions of a solar year as in the modern Gregorian calendar.
As a lunar cycle averages 14.77 days, each month is thus divided into two periods that are either 15 or 14 days long. The second period, or fortnight, of each month is marked with the word ⟨ ATENOVX ⟩. This system bears comparison to those found in Germanic and Greek cultures as well as the Hindu lunisolar calendar that is still in use today.
By comparison with OIr cóicthiges and W pythefnos < Br pempedecan + nos ‘fortnight’, we could describe this as something like kʷenkʷedekam·kʷe noxtiom or (lit. ‘fifteen night’), although similar to English ‘fortnight’ (‘fourteen’ + ‘night’) the neo-Celtic forms appear heavily elided. While their meaning is clear, they are formally irregular and opaque, so more study is needed to arrive at a secure back-formation.
Lunisolar years are more complex and irregular than solar years, as they compromise between following both the Lunar and Solar Cycles. The Coligny Calendar’s keepers had devised and maintained a sophisticated system of intercalation in order to prevent it from drifting too far off of either one. (Intercalation simply means inserting extra time units into the calendar to correct a shortcoming. For a familiar example, cf. “Leap Day” in February). With the Coligny Calendar, intercalation took the form of adding an entire month every two-and-a-half years as well as possibly adding and removing days in the month of Equos.
The tablet also records many cryptic symbols and abbreviated words that mark various days as being significant in some way. It is presumed that these markings refer to feast days, lunar phases, and what could be some sort of interlinking wherein each month has days that refer to the month prior or the one yet to come. But the ongoing attempt to decipher these markings is beyond the scope of this post.
It is known that a 19-year Metonic Cycle arrives at a common multiple of the solar year and lunar month, so it is perhaps the simplest cycle to keep a lunisolar calendar in sync. The Coligny Calendar tablet is a quinquennium, or five-year cycle, which may have fit within a larger cycle similar to the Metonic one. Alternatively, Pliny speaks of a 30-year age that the Celts adhered to, so perhaps this five-year cycle is ⅙ of a 30-year cycle.
The calendar, no doubt, has extensive gaps in its record. But thanks to the fact that the tablet records a quinquennium, as well as owing to the inherently repetitive nature of timekeeping, it is possible to extrapolate patterns from the surviving fragments to fill in many of the gaps.
The method used here was a simple heuristic: First, each iteration of each month was collected (for example, the month of Samonios in years 1, 2, 3, 4, and 5). Then, patterns within the month were identified. If a pattern was discerned of any length (even if it only repeated one time) and it was not contradicted elsewhere in any of the other surviving pieces for that month, it was treated as reconstructible and thus projected onto its corresponding positions in the blank portions of the calendar. All surviving text is represented in black while reconstructed text is bright red so there can be no mistake about which is which.
This heuristic, like any heuristic, is an imperfect one. There are occasional instances, for example, where we may see a pattern in the surviving fragments repeat three or even four times, but then the pattern is broken after that. So it is always possible that what had repeated in the surviving fragments might have differed or not existed in the missing ones. For this reason, the bright red text should not be taken as a citable or authoritative reference. It does, however, allow us to imagine what might have existed in the missing fragments and it certainly helps facilitate the reading and comparison of the calendar.
It is worth noting that the creators of the tablet themselves appear to have made some errors. They also extensively used abbreviation and shorthand. So even if a pattern is not consistently written out in every instance, we cannot be certain that it wasn’t implicitly followed or meant to have been included.
Another shortcoming in the heuristic is its inability to anticipate the content of the intercalary months, since they are each unique. Luckily, the second intercalary month is largely intact. But unfortunately, the first one is mostly destroyed. The first fortnight of the following months was likewise too poorly preserved for the heuristic to be of much use there: Ogronios, Giamonios, Elembiu, and Edrinios.
Question marks appear in cases where it is evident that there was additional text that cannot be determined. The color of the question marks is insignificant, and I apologize that it was not made more consistent.
Hopefully deeper and more meaningful analyses of the content can be drawn from this reconstruction in the future.
For further reading about the Coligny Calendar, see Mac Neill’s On the Notation and Chronography of the Calendar of Coligny (1928). It can be read for free on JSTOR’s website with a registered account.
The reconstructed images were made with GIMP and are published by Tegos Skrībbātous under BY-NC-SA 4.0 license, a form of copyleft. They are typeset with Astigmatic’s Marcellus font family, which is published under the Open Font License.
Kowarisagetus ‘reconstruction’ — cf. OIr. córugud ‘arrangement, order,’ cóir ‘proper, fitting,’ W. cywair ‘proper condition, restoration, repair.’
Arewedonālos ‘preliminary’ — cf. OIr. fedan ‘conveying, leadership,’ W. arweiniol ‘leading, preliminary, introductory.’
Wlidyā ‘calendar’ — calque of OIr. félire which derives from Lat. vigilia (source of OIr. féil ‘feast-day’) + -āria (abstract noun-forming suffix)
Koliniākom — proto-form of ‘Coligny,’ attested as Cologniacum, Coloniacum in Medieval times. Root kolino- ‘holly’ + -(i)akom (neuter substantive of an adj. suffix), essentially meaning a place where holly grows.