Little Almanac

The Coligny Calendar

The Coligny calendar is a bronze plaque, found in 1897 at Coligny in the department of Ain in eastern France, surviving in 73 fragments that represent roughly two-fifths of an original about 1.48 metres wide. It is written in Gaulish, using Latin capitals and Roman numerals, and is dated on its letterforms to the late second century of the common era. Eight fragments of a closely comparable calendar were found at Villards-d’Héria in the Jura 1.

It is the only surviving calendrical document from a pre-Christian Celtic-speaking people, and that sentence needs exactly those words. Everything else known about Celtic time-reckoning comes from Roman outsiders or from Irish writers working centuries after conversion. This is an object, made by the people who used the system, recording the system itself.

What it shows

A lunisolar cycle of five years and 62 months: sixty regular months plus two intercalary months, inserted to keep lunar months attached to the solar year. Reconstructions of the total day count differ, and they differ because the plaque is fragmentary 12.

Twelve month names: Samonios, Dumannios, Rivros, Anagantios, Ogronios, Cutios, Giamonios, Simivisonnios, Equos, Elembivios, Edrinios, Cantlos. The names of the intercalary months are read variously and the readings are insecure.

Months marked MAT or ANM, conventionally read as good and not good. MAT months have thirty days and ANM months twenty-nine, so the marking may be primarily a statement about length with the auspicious reading secondary, or the reverse. Both readings should be given, and the expansion of ANM is itself an inference.

A division of every month at ATENOUX, usually read as returning or renewed night, splitting each month into two halves. This is the single most important structural feature on the plaque. A native half-month reckoning, closer to a fortnight than to anything in the Roman calendar, is exactly what a Roman-influenced object recording an imported system would not show, and it is the strongest argument that the calendar records an indigenous scheme 3.

A named festival: TRINOX[TION] SAMO[NI] SINDIU, “the three nights of Samonios today,” at the seventeenth of Samonios. A three-night festival, named in a Celtic language, on a Celtic calendar. It is close to the only thing known about Celtic festival practice from a Celtic-language source.

What it does not show

  • Any of the four Gaelic quarter-day names, with the disputed exception of Samonios.
  • Any solstice or equinox marking, of any kind.
  • Any zodiac, any planetary reference, any astrological content whatever.
  • Any druid. Nothing on the plaque mentions them.

Correction

Commonly saidThe Coligny calendar proves the Celtic year began at Samhain on 1 November.

The argument runs: Gaulish Samonios is the same word as Old Irish Samain; Samain opens the Irish year at the beginning of November; the Gaulish calendar therefore begins there too.

The first step is where it fails. Gaulish Samonios contains the Celtic root for summer, and Giamonios, from the root for winter, sits six months away from it. A pair of month names built on summer and winter, half a year apart, is what would be expected if they marked the openings of the two halves of the year, and it places Samonios in summer. That is where a major reconstruction of the cycle puts it 2.

The counter-argument requires Old Irish samain to derive from a compound meaning summer’s end. That derivation appears as a medieval Irish etymological gloss and is linguistically doubtful, and medieval Irish glossators produce etymologies elsewhere that are demonstrably invented 4.

So the identification of Samonios with Samain, and with it the Celtic new year at 1 November, is an inference and a disputed one. Nearly every popular account states it as settled fact.

What is handed back is larger than what is taken. The plaque attests a genuine indigenous lunisolar system, a half-month division found nowhere in the Roman calendar, an intercalation scheme, and a named three-night festival, all in a Celtic language on a dated object. That is a great deal more than a new year’s date, and none of it is inferred.

Reading it as a calendar today

A converter that turns a modern date into a Coligny date is an attractive idea and a dangerous one. There is no agreed epoch and no agreed correspondence to the Julian calendar, so any conversion is a choice of reconstruction. If such a thing is built at all it has to be labelled with the name of the scholar whose reconstruction it implements and with the names of the competing ones, and it must never be presented as a person’s Celtic date.

This site therefore treats Coligny as a document to be read, and runs no calendar from it. For the fourfold Irish scheme this calendar is so often used to authenticate, see Samhain and Cross-Quarter Days; for a European calendar that can be run, because its rules survive in the legal texts of the people who used it, see The Old Icelandic Misseri Calendar.

Gap in the evidence

The epoch is unknown. Without a dated correspondence between a position in the cycle and a Julian date, the reconstructions cannot be tested against anything external, and they disagree.

Whether the calendar was reset periodically against observation is likewise open. The five-year cycle is a worse long-run approximation than the nineteen-year Metonic rule known to Greek astronomy at the time, which raises the question of how the drift was handled, and the plaque does not answer it.

Nor does anything reconcile the plaque with the one classical notice of Gaulish lunar reckoning, which describes a rite performed on the sixth day of the moon and calls that day the beginning of their months and years. That statement and this object are hard to read as descriptions of one system, and the discrepancy is one of the more interesting open problems in the field.

References

  1. Paul-Marie Duval, Georges Pinault. Recueil des inscriptions gauloises, vol. III: Les calendriers (Coligny, Villards d'Héria). XLVe supplément à Gallia. Paris: Éditions du Centre National de la Recherche Scientifique, 1986. find a copy (catalogue search) recordprimary source

    The standard epigraphic publication of the Coligny and Villards-d'Héria calendar fragments. Author, title, publisher, place and year confirmed from the review notice in Revue des Études Anciennes; no ISBN was seen, so the catalogue search asserts nothing in its place.

  2. Garrett S. Olmsted. The Gaulish Calendar: A Reconstruction from the Bronze Fragments from Coligny. Bonn: Rudolf Habelt, 1992. find a copy (catalogue search) recordscholarly

    A major and contested reconstruction of the Coligny cycle, which places Samonios at midsummer. The title page carries a much longer descriptive subtitle than the short form used here. No edition record was loaded directly, so a catalogue search stands in place of an ISBN rather than asserting one.

  3. Stephen C. McCluskey. Astronomies and Cultures in Early Medieval Europe. New York: Cambridge University Press, 1998. find a copy (catalogue search) recordpeer reviewed

    Treats the Coligny calendar and insular computistical astronomy. Author, title, publisher, place and year confirmed from the review notice in the American Historical Review 104/5; no edition record was loaded directly, so the catalogue search stands in place of an ISBN rather than asserting one.

  4. Kim McCone. Pagan Past and Christian Present in Early Irish Literature. Maynooth Monographs 3. Maynooth: An Sagart, 1990. ISBN 9781870684101 recordpeer reviewed

    Argues that early Irish literature is the product of a monastic culture constructing a native past on biblical and patristic models, rather than a transcript of oral pre-Christian tradition.