Coligny Calendar: Gallo-Roman Astronomy

The underlying scaffold of the Coligny Calendar, a metonic cycle of probably 30 years, was likely first compiled by Druidic leaders of the Gauls in the first century BCE (Dunbavin 2018, McKay 2024, Olmsted 2001). It was not the oldest successful luni-solar calendar — the earliest known to date was created by the Babylonians in the 5th century BCE, during the reign of the Achaemenid dynasty Persian king Xerxes I (ca. 485-465 BCE), which fact deserves a quite interesting digression if you choose to follow it. Attempts to align the pathways of the moon and sun are much older than that, of course, and are in evidence throughout the world (Ossendrijver 2018, Hannah 2005). The Coligny calendar was certainly based at least in part on the Babylonian version.

The Coligny Calendar is intriguing because — it's messy. The calendar contains evidence of the tinkering it took to keep it up to date over a period of probably 250 to 300 years since its compilation. The results of the tinkering evinced by the calendar provide insights into the condition of the science of time-keeping in the Gallo-Roman culture after its conquest by Rome (McKay 2025, McCluskey 1990, Olmsted 2001).

Approximately 150 fragments of the plaque were found in 1897 at Charmoux, just northwest of the French commune of Coligny, in a deposit mixed with pieces of a bronze statue thought to represent the Roman god Mars. The calendar fragments amount to about 40% of the original, a thin (5 millimeters or about 0.2 inches) rectangle of bronze measuring approximately 150x90 centimeters (59x35.5 in). According to Visiter Lyon, the arranged pieces are currently on permanent display at the Musée Lugdunum in Lyon, France.

Each of the five years engraved onto the plaque includes 12 lunar months of alternating 29 and 30 days. There are five repetitions of 12 lunar months on the plaque; and two intercalery months are interleaved, one at the beginning and one 30 months later, in the middle of the third year. In metonic calendars, intercalery months are added to calibrate and accommodate the differences between the lunar year (about 354.4 days in length) and solar year (about 365.24 days). The months are contained in 16 columns, and start with the first intercalary month in the upper left-hand corner in Column 1 and end in Column 16 with the fifth occurrence of the month Cantlos.

The Gaulish year was divided into two parts: summer and winter. There are twelve named months in the calendar, six summer and six winter months. Independent researcher Helen T. McKay conducted an extensive investigation (2024) into determining the seasonality of the calendar and concluded that the first half of the Gaulish year on the Coligny calendar started one month before the summer solstice; the second half began one month before the winter solstice.

On the calendar, each of the 12 named and two intercalery months are broken into two fortnights — the first half is 15 days long, the second half has 14 or 15 days. Each month begins at sunset on the sixth day after the New Moon, when the First Quarter moon hangs in the sky. The first half of each month is thus dedicated to the gibbous moon, the second half to the crescent moon. (McKay 2016, Olmsted 2001)

This image is one of the larger intact pieces of the calendar. It spans Columns 3 and 4 and shows the second fortnight ("ATENOVX" or "returning night"), that is, the second half of the months of Eqvos (month 9 in Year 1) and Samonios (month 1 in Year 2), what our modern calendars would list as days 16-29 or 30. Each day's listing begins with a punctured hole that acted as a socket for a peg or pin that the user would move as each day began. Next to that are Roman numerals representing the day number. The days of the first half of each month are listed by Roman numerals from 1-15 (I-XV); the second half are listed from 1-15 or 1-14, depending if the month had 30 or 29 days. Next to the Roman numerals is a space for the three vertical lines called "triplets" or "counters," and after that are varied notes about each day and its importance.

Descriptions of each day have provided linguists with about 2,000 Gaulish terms, include a multitude of abbreviations representing approximately 60 different concepts. D by itself is day or days ("divos" in Gaulish); N by itself is night or nights ("nocts"). MAT means "complete, auspicious," and AMB (possibly "ambios") generally means ""incomplete, not good, inauspicious"; so together D AMB is a warning of an inharmonious day. IVOS means "feast"; N INIS R is a reference to a night (N) in which an "unwholesome moon" can be found. "Prinni laget" is a reference to the falling solar pathway and the count from midsummer to midwinter. "Prinni loudin" refers to the rising solar pathway and tracks counts from midwinter to midsummer. Together these two Prinni terms divide the year into four equal parts and may signify the solstices and equinoxes. (Olmsted 1988, Aldhouse-Green 1998), Garcia Quinela and Gonzalez-Garcia 2017, Parisot 1992)

In 2001, U.S. anthropologist and linguist Garrett Olmsted published his digital reconstruction of the entire plaque. He used Adobe Photoshop (TM), copying information from the surviving fragments of the same months, making some adjustments based on his comprehension of the plaque itself. Olmsted's reconstruction largely supported previous understandings of the calendar, but it also brought new details about the engravers' process and the chronological units used to create the calendar. (Olmsted 2001)

Many scholars have wrestled with the most intriguing part of the Coligny Calendar: transmission errors, made while the plaque was being constructed. By 200 CE, either the tracking methods changed or people who made the calendar were unfamiliar with its original workings. U.S. historian Stephen C. McCluskey (1990) suggests that the changes in evidence occurred as astronomer priests tracked the lunar cycles, and informed the plaque makers that certain festivals had fallen out of synch with the solar calendar.

To adjust for such changes, the plaque makers simply swapped out the complete listings for the previous feast day for those into new feast day location, and moved information from the new landing place into the previous one. Unfortunately (or actually, fortunately for the modern scholars), the makers copied all of the information in the listing, rather than selecting just the feast day; and some of that information for those days pertaining to the solar movements was now incorrect.

The Samonios Year 2 image above illustrates a swap of a day from the third day of the month of Samonios to the 17th day, or 14 days later. In 1988, Olmsted suggested that the memo TRINVX in the new location is a corrupted abbreviation of the archaic Celtic ordinal numeral three, and thus TRINVX SAMO should be translated as "third day of Samonios." This is only one of a multitude of such swaps in evidence.

One significant source of information about swaps shifting are triplets or counters, what scholars call three vertical lines, one either longer than the other two, and/or capped with a tiny crossbar — you can see them in their original form on the image above at Days X, XI, and XII (Samonios 25, 26, and 27) in the second fortnight of Samonios Year 2. They are listed on the plate right after the Roman numerals, and occur on subsequent days in the same order, TII, ITI, and IIT. The pattern was clearly disrupted by the move. Olmsted (1988) supposed that the TII marks were part of a critical counting scheme to determine the solar and lunar positions. But most researchers, supported by McKay's extensive 2018 analysis, believe that the marks represent some form of triple division of the daylight hours. Whatever they were used for, disrupted patterns have been a flag for researchers to recognize that a transfer has occurred. McKay found that of the 167 triplets manifested on the calendar, only 57 of them remain in their original location. Apparently in the process of the many swaps over the decades, the triplets lost their meaning and/or usefulness, but the marks were left behind. Copies of copies of copies. Now that is a fortuitous mess.

A good question remains about why the Gaulish months begin on the sixth day, and it is something of a debate in scholarly circles. Pliny the Elder, writing in his Natural History in the first century CE, tells us that's the case.

"Mistletoe is, however, rather seldom found on a hard-oak and when it is discovered it is gathered with great ceremony, and particularly on the sixth day of the moon (which for these tribes constitutes the beginning of the months and the years) and after every thirty years of a new generation, because it is the moon rises in strength and not one half of its full size." (Pliny 1938)

But most lunar calendars, including the oldest known Babylonian one, begin within 2-3 days after the new moon, when the first sliver of the crescent is visible in the night sky. Olmsted (2001) argued that the original version of the Gaulish calendar probably started with the second or third day after the new moon, because it was clearly based on the Babylonian version or one of the widespread copies. If the calendar had originated 1,000 years before Pliny (as Pliny suggested it did), that would account for the three or four day lag in the start. Alternatively, McKay (2024) argued that the month began on the sixth day because that's the first quarter moon, when the moon has a straight edge and is readily identifiable by eye.

A few fragments of another version of the Coligny calendar have been found in the French commune of Villards d'Heria, the first from a river bed in 1807 and then 5-8 others from a nearby Romano-Gaulish shrine excavated in 1967 and dated between the 1st and 3rd centuries CE. Those pieces were similar to the Coligny, but contain differences in the details (Swift 2002). There are no other versions known at present.

When the Gauls stopped using the calendar is not clear. The Coligny calendar was deliberately broken and the pieces buried about 200 CE, but whether it was the last version is unknown. It's possible that the Coligny Calendar was discarded because it was just too difficult to maintain, too full of updates to continue. It's also possible the Gauls were finally forced to adopt the Julian calendar, part of the Roman desire to crush out pagan religions in their empire. The adoption happened eventually — the four major holidays of the Celtic calendar are today driven by the Julian calendar: Samhain (November 1st), Imbolc (February 1st), Beltaine (May 1st), and Lughnasas (August 1st). When that convention was established is unknown. It could also be that archaeologists simply haven't found other copies yet, or the metal might have been melted down and reused. (Aldhouse-Green 2021, Green 1998, Garcia Quinela and Gonzalez-Garcia 2017, McCluskey 1990.)

There is much to be said about metonic calendars that deserves its own Yurt; and there are some earlier versions that I'll need to dive into. But in the meantime, the Coligny calendar is an intriguing and detailed example of how people merge sky-based astronomy and earth-based festivals, struggling to make sense of enormously important celestial bodies with eccentric orbits.

- Photograph of the Coligny calendar at the Lugdunum Museum. LugdunumMTR, Public domain. https://commons.wikimedia.org/wiki/File:Calendrier_de_Coligny_-_BR.001_-_Mus%C3%A9e_Lugdunum.jpg

- Coligny calendar, as transliterated by Seymour de Ricci (1881-1942), in his 1926 article, Le calendrier celtique de Coligny (de Ricci 1926). Public domain. https://commons.wikimedia.org/wiki/File:Seymour_de_Ricci,_Le_calendrier_celtique_de_Coligny_(1926).jpg

- Structure of a Gaulish month as it appears in the Colighy Calendar. Coicise is the Irish word for "fortnight". Macfeegles 4.0 https://commons.wikimedia.org/wiki/File:The_structure_of_a_coligny_month.png

- Base image is a detail of the Calendar of Coligny, including the second fortnight of the Eqvos and Samanios. Base image credit: Gozitano, annotations by K. Kris Hirst. CC 4.0. https://commons.wikimedia.org/wiki/File:Calendrier_de_Coligny.jpg

- Base map is Month 14 (Samonios year 2) from the Coligny calendar, as transliterated by Seymour de Ricci (1881-1942, de Ricci 1926). Annotations by K. Kris Hirst, CC 4.0. https://commons.wikimedia.org/wiki/File:De_Ricci_Coligny_1926_detail_month14.jpg

- Garrett Olmsted's Reconstructed Coligny Calendar. Garrett Olmsted CC 4.0 https://commons.wikimedia.org/wiki/File:Whole_Reconstructed_Calendar_300_jpeg.jpg

- Phases of the Moon. Orion 8, CC 2.0 unported https://commons.wikimedia.org/wiki/File:Moon_phases_en.jpg

- Reconstruction of a Gallo-Roman temple at Archeon. Image credit: RomeinenNU cc 4.0 https://commons.wikimedia.org/wiki/File:Gallo-Roman_temple_Archeon.jpg

Metonic cycle Mythical Ireland

Xerxes I World History.org

Shanati Project - reconstruction of the Ancient Babylonian calendar by David Danzig, includes an excellent video explaining the issues

The Coligny (Gaulish) Calendar Timey's Calendarium, includes a dynamic Gaulish calendar for snicks.

Archeon Events Archeon is both a museum and a living history location in the Netherlands.

- Aldhouse-Green, Miranda. Rethinking the Ancient Druids: An Archaeological Perspective. University of Wales Press, 2021. New Approaches to Celtic Religion and Mythology, Jonathan Wooding.

- de Ricci, Seymour. "Le Calendrier Celtique De Coligny." Journal des Savants vol. 10, 1926, pp. 448-449, https://www.persee.fr/doc/jds_0021-8103_1926_num_10_1_5822

- Dunbavin, Paul. "On the Coligny Calendar and the Neolithic Calendar in Plato’s Critias." Chronology & Catastrophism Review, vol. 2018, no. 2, 2018, pp. 50–53. https://www.sis-group.org.uk/issue/chronology-catastrophism-review-2018-2/

- García Quintela, Marco V. and A. César Gonzalez-Garcia. "Archaeological Footprints of the “Celtic Calendar”?" Journal of Skyscape Archaeology, vol. 3, no. 1, 2017, pp. 49–78, doi: https://dx.doi.org/10.1558/jsa.31039

- Green, Miranda J. "The Time Lords: Ritual Calendars, Druids, and the Sacred Year." Prehistoric Ritual and Religion : Essays in Honour of Aubrey Burl edited by Aubrey Burl et al., Sutton, 1998, pp. 190–202.

- Hannah, Robert. Greek & Roman Calendars: Constructions of Time in the Classical World. Duckworth, 2005.

- McCluskey, Stephen C. " The Solar Year in the Calendar of Coligny." Études celtiques, vol. 27, 1990, pp. 163–74, https://www.persee.fr/doc/ecelt_0373-1928_1990_num_27_1_1925

- McKay, Helen T. "The Coligny Calendar as a Metonic Lunar Calendar." Études celtiques vol. 42, 2016, pp. 95–121, https://www.persee.fr/doc/ecelt_0373-1928_2016_num_42_1_2471

- ---. " Defining the Systematic Patterns for the Triple Marks of the Coligny Calendar." Études celtiques, vol. 44, 2018, pp. 91–118, https://www.persee.fr/doc/ecelt_0373-1928_2018_num_44_1_2182

- ---. "Identifying the Seasonal Placement of the Coligny Calendar." Études celtiques, vol. 50, 2024, pp. 37–75.

- Olmsted, Garrett S. "A Definitive Reconstructed Text of the Coligny Calendar." Journal of Indo-European Studies, vol. Monograph 39, Institute for the Study of Man Inc., 2001.

- ---. "The Use of Ordinal Numerals on the Gaulish Coligny Calendar." Journal of Indo-European Studies, vol. 16, no. 3-4, 1988, pp. 267–339

- Ossendrijver, Mathieu. " Babylonian Scholarship and the Calendar during the Reign of Xerxes." Xerxes and Babylonia : The Cuneiform Evidence, edited by Caroline Waerzeggers and Maarja Seire, Peeters, 2018, pp. 135–64. https://directory.doabooks.org/handle/20.500.12854/34384

- Parisot, Jean-Paul. "On the Origin of the 5-Years Cycle in the Celtic Calendar." Études celtiques vol. 29, 1992, pp. 343–54, https://www.persee.fr/doc/ecelt_0373-1928_1992_num_29_1_2017

- Pliny, the Elder. "Natural History with an English Translation in Ten Volumes." The Loeb Classical Library, edited by T.E. Page, vol. IV, William Heinemann Ltd/, 1938. general editor, Harris Rackham, https://archive.org/details/naturalhistory04plinuoft/mode/2up

- Swift, Cathy. "Celts, Romans and the Coligny Calendar " Theoretical Roman Archaeology Journal, vol. 2001, 2002, pp. 83–95, doi: https://doi.org/10.16995/TRAC2001_83_95

This article is part of the Staring into Space project.

Questions? Comments? Suggestions? Contact me at