There are a number of examples online regarding a 28-day 13 month lunisolar calendar with an extra day. This post will explain why this calendar is a modern invention, and has no historical basis.

Before we begin, let’s take a look at some of the claims, perpetuating the fake calendar.

Here is a comment I saw on facebook, posted by someone who says he is half Native American.

Note the key points here. Lunar phases and 13 months in a solar year (we know it is solar as he is adding an extra day to make up the total days to 365. The same length as the average Gregorian year).

Here is a quote from (which as you will come to see is such an ironic name!)

“The 13-month, 28 day alternative has been in use on this planet for more than 6000 years. In prehistoric Indi and China, and throughout South America it was the standard time-keeping system. The Essenes, Egyptian, Polynesian, Maya, Inca, Lakota, and Cherokee used a 13-month, 28 day calendar. the Celtic knowledge of the Druids is based on the tree Calendar, also a 13-month, 28 day calendar. Today many cultures are still using their traditional 13-month calendar system.”

The real clue to just how untruthful this paragraph is, is the reference to Robert Graves’ “Celtic Tree Calendar”, a system long known to be the invention of Robert’s in his book “The white goddess”. In the book he analyses European mythology and the Irish Ogham alphabet to propose that Celts had a secret hidden lunar calendar of 13 months of 28 days, with an extra day to reconcile it with the solar year. He “proves” this by liberally re-arranging both the Ogham alphabet and many of the Ancient poems of Ireland and Wales.

The truth is that his calendar is pure fantasy. The Celts had no such calendar. Yet the myth is continually perpetuated by the ill-informed. Not only did the Celts never use a 28-day 13-month calendar, but I would venture that no culture in history has ever used this system for one very obvious and simple reason. The maths simply doesn’t work. Yet, as I wake up this morning, there it is again being shared by another poor soul.

All calendars were created as a way to measure the passage of time by observing the movements of the sun and moon. A year is simply the Earth’s orbit around the Sun. By creating a calendar, you are creating a way for a culture to align itself with the rhythms of the environment. The changing seasons and the movements in the havens. So, a fundamental prerequisite of any calendar is that it must work. Also remember that no calendar is perfect. Every single one has a margin of error. And in general, this error is corrected by introducing leap days or leap months. Therefore any calendar that wishes to remain in sync with what is observable in the sky must make corrections every so often. Finally, calendars can either follow only the sun, only the moon or both. i.e calendars can be solar, lunar or lunisolar.

With all that in mind, let’s take another look at the claim

The claim is:
That there is a calendar that allows you to follow the “phases of the moon”. That it has months of 28 days. That there are 13 months in a year (13 x 28 = 364). And that at the end of the year there is a single extra day (364 + 1 = 365 – the same length as the average Gregorian year). Importantly there is no suggestion of any other corrections, other than this extra day each year.

In essence, that it is possible to have a 28 x 13 + 1 lunisolar calendar with no further corrections.

Phases of the Moon.

Discounting any association with the sun, first let’s address this lunar phase claim and see if this works as a lunar calendar.

There are four main observable cycles of the moon. The first is the synodic month, or lunation cycle. This is the time it takes from full moon to full moon, and results in what we know as phases of the moon.

The apparent full, waxing and waning of the moon is a result of this cycle, and it takes 29.5 days. The second cycle is the sidereal month. This is the time it takes the moon to orbit the earth once and it takes 27.32 days. The third cycle is lunar nodal period. This is the time it takes for the moon’s rising points to travel up and down the horizon, from the southeast to the northeast and back again. This cycle takes 27.21 days. The final cycle is the anomalistic month. This is the time it takes for the moon to travel from its closest point to earth (perigee) when it appears biggest, to its farthest point (apogee) where it appears smallest, and then back again. This cycle takes 27.5 days resulting in an increase and decrease in the moons size of around 14%)

The claim is that the 28-day calendar follows the phases of the moon. As stated above, this cycle actually takes 29.5 days. Anyone who tried to follow a 28-day calendar that aligned to the phases of the moon would quickly notice this discrepancy of 1.5 days per month. If you were to start counting from a full moon, after only 4 months you would notice that the full moon occurred 6 days after you expected it. Over the course of the year (13 months) you would accumulate a discrepancy of 19.5 days. i.e. the 13th full moon would occur 19.5 days after you expected it. Add on the extra day and you are now 20.5 days out of sync. This means that the idea of a 28-day month that follows the “moon phases” is utter rubbish. It is simply not possible.

Perhaps instead, the “moon phases” being referred to is actually a mistaken reference to another lunar cycle?

Perhaps the anomalistic month, which at 27.5 days is the closest number to the alleged 28 day “phases”. Well there are a number of problems with this theory. The anomalistic month first and foremost has never been known as “moon phases”. It is also rather difficult (but not impossible) to measure the apparent size of the moon. And that is not to mention that 27.5 is still not 28 and you lose half a day each month, resulting in 6.5 days lost over a 13 month year.

What about the sidereal month? This is the time it takes the moon to complete an orbit of the earth. The simplest way to measure this is to select a fixed point in the sky and time how long it takes for the moon to pass this point. The first problem you have is selecting a fixed point. It can only be a star, but it has to be a star that the moon can “pass”. This means that you have to pick a star somewhere along the ecliptic. The Zodiac. The issue here is that for a good proportion of the year, any of these stars will be over head during the day light. So, you could not use only one star, but several of them around the ecliptic. The second issue is the third cycle of the moon, the lunar nodal period. This results in the moon appearing anywhere up to plus or minus 28.5 degrees of the ecliptic. So, the moon could pass far above or below any selected fixed point. Throw in the fact that some nights it will be cloudy and you can’t see the stars, and you have a cycle of the moon that is particularly difficult (but not impossible) to observe with the eye.

But how does the maths stack up? Well the cycle is 0.68, let’s say 0.7 days shorter than the supposed 28-day month. After only one year (13 months) the discrepancy is 9.1 days. This means that the if you select a point in the sky that you expect the moon to pass, after 13 months the moon will pass that point 9.1 days before you expect it to. Add on the extra day and you are now 10.1 days out of sync. This again is an unacceptable discrepancy for any astronomer dreaming up a calendar and actually watching the sky in the ancient world, if their objective is to create either a lunar or lunisolar calendar.

The maths clearly highlights that there is no way to have a calendar where each month is 28 days long and remain in sync with any cycle of the moon for even one year.

Lunar Months

What about the claim of 13 lunar months in a year? Well full moon to full moon (synodic month) is 29.5 days. 29.5 x 13 = 383.5. that is 18.25 days longer than the solar year (365.25). If you started your year with a full moon on January the 1st. Then the next year (13 months later) you would be starting your next year on January 19th. The year after that, February the 8th. If the lunar cycle that defines a month is the moon phases (synodic month) then there is no way you can have 13 full months in a year.

Of course, if you have 12 lunar months, your year is too short, and the next year you will start a bit earlier. After approximately 2 and a half years the discrepancy will have accumulated enough to include a 13th full moon (not an entire synodic month) in the calendar. This is the definition of a Blue Moon. Each season has 3 full moons (3 x 4 = 12) and every two and a half years a season will have four full moons. The third in the sequence is known as a Blue Moon.

The fact is, the 28 x 13 + 1 calendar simply cannot be a lunar calendar, and it certainly cannot reconcile any lunar pattern with the solar year within the space of a single year as it seems to claim. The fact is the moon and the sun engage in a 19-year dance through the sky known as the Metonic cycle. The metonic cycle was first discovered in ancient Babylonia, and then rediscovered by Meton of Greece. He noticed that if he looked at the sun and moon, on a given date, they would be in the same place in the sky on the same date 19 years later. If you want to make a calendar that reconciles the sun and the moon, then the Metonic system is the ONLY way you can do it. With that in mind, let’s forget about the moon, and see if this alleged calendar works as a solar only calendar.

Solar year

Well it has the correct number of days. 28 x 13 = 364 + 1 = 365. So, for a couple of years this calendar will stay in sync with the Gregorian calendar. After that, however, there are no further rules around leap years. Every 4 years you would have to include a second non-day, just like the Gregorian calendar does. In fact, you would have to follow the exact same rules as the Gregorian calendar with the inclusion of a leap day every four years except where the year is divisible by 100, unless the year is also divisible by 400. Although this would allow the 28 x 13 + 1 calendar to maintain accuracy with the solar year, I have never seen any such suggestion applied to the Celtic tree calendar or any other alleged 28 x 13 + 1 calendar. So, without these rules in place, the calendar loses 1 day every 4 years, and no rules are in place to correct that. Therefore the calendar doesn’t work as a solar calendar either.

Historical claims.

Given that we have now clearly shown that this alleged lunisolar calendar simply doesn’t work for even one year, it is important to address the claimed antiquity of the system. Let’s look again at that quote from

The 13-month, 28 day alternative has been in use on this planet for more than 6000 years. In prehistoric India and China, and throughout South America it was the standard time-keeping system. The Essenes, Egyptian, Polynesian, Maya, Inca, Lakota, and Cherokee used a 13-month, 28 day calendar. the Celtic knowledge of the Druids is based on the tree Calendar, also a 13-month, 28 day calendar. Today many cultures are still using their traditional 13-month calendar system.

Well I am not going to waste any more time rubbishing the tree calendar, but what I will say is that surviving evidence such as Stonehenge and the Coligny Calendar both display knowledge of the Metonic system which uses alternating 29- and 30-day months (not 28). This is the only evidence we have of Celtic or proto-Celtic calendars.

Next let’s look at the Cherokee and Lakota calendars. Well some examples show 12 months, others show 13. And they are clearly naming for the lunation cycles which last 29,5 days. I suspect that all of these calendars (because they were observational) would have been 12-month lunar calendars with a 13th month added every two and a half years (blue moon). I see no evidence for 28-day month length or reconciliation with the solar year. They are Lunar calendars only.

Ancient India and China both discovered the Metonic system, quite possibly both inheriting it (like the Greeks did) from Babylonia.

The Mayan calendar is extremely complex and includes a number of cycles. In order to do this, it measures both the synodic and sidereal months, but at no time does it denote a uniform 28-day 13-month year. It is a bit of an insult to the Mayan calendar to suggest it is so simple.

The Inca had two separate calendars. A solar calendar of 365 days and a lunar calendar of 328. They did not have a 28-day 13-month lunisolar calendar.

The Egyptians were not very good at calendars it would seem. They had a number of tries at it, and in the end, they ended up with a 365-day solar calendar. The Greeks pointed out that they needed to add a leap day, but the Egyptian priesthood resisted fearing the chaos that it would cause. It was not until the Romans took over that they finally submitted, added the leap day and fixed their calendar which resulted in Julius Cesar taking their calendar and creating his own from it, the Julian Calendar which was later tweaked to become the Gregorian Calendar we use today, which like its Egyptian ancestor is solar.

The Polynesian calendar is actually a mechanism for tracking the appearance of certain constellations in the sky, and has very little to do with the sun and the moon. A very sensible thing to do really, as the sun and the moon are all over the place, and reconciling the patterns is hard. Why bother? The sky rotates at a nice steady pace year on year. I guess they thought they would just measure that.

As for the Essenes calendar, well the quote is almost true. The year is 364 +1 just like the calendar in question but it at no point pretends that it is a lunar or lunisolar calendar. It is solar only. It understands that 28 days and the lunar cycles cannot be reconciled without constant correction. And it is likely why Judaism doesn’t use this calendar today, and instead uses the Metonic system.

So why does the Druidcraft Calendar solve all of these problems?

Like all calendars that reconcile the movements of the sun and moon, the Druidcraft Calendar uses the Metonic system. This is the historical basis of the Druidcraft Calendar. The Gallo-Roman Coligny calendar that may reflect an earlier Celtic tradition is Metoinc, though slightly different to the one employed on the Druidcraft Calendar. The Aubrey holes at Stonehenge reflect a Metonic system too.

The Druidcraft calendar tracks both the sidereal and synodic lunar month. The Moon ring (3rd one in) has 56 holes, and moves 2 holes per day. So it takes 28 days for the peg to complete one cycle around the calendar. That’s correct. 28 and not 27.3. If you are about to shout “hypocrite!” at your screen wait for the “but”. The calendar also contains a rule for correction, as well as there being a variable involved. First, the ring is intended to actually show the positions of the sun and the moon in the sky in relation to each other in order to predict eclipses. In the time it takes for the moon peg to get around the calendar, the sun peg will have moved at least 2 holes. So the relationship between the position of the sun and the moon is preserved. Second there is a rule for correction involving the second ring that tracks the synodic month. The Lunation ring.

The Lunation ring is actually two rings; one of 30 holes and one of 29. This is a Druidcraft month, 29 or 30 days. The peg travels around each ring alternately. This results in a 2-month period of 59-days which divided by 2 is 29.5, the same time it takes for a lunation cycle. This ring remains accurate. So, the correction rule is simple. On the first of each month (new moon) ensure that the moon peg is in the same hole as the sun peg (that the sun and the moon are in the same place in the sky), by correcting the moon peg. Most of the time you do not need to do this, but after a few months the discrepancy builds to a point where the moon peg is a hole or two off and so you correct it. This ensures that the calendar will continue to accurately predict lunar events.


I hope that this article has once and for all shown that the evidence for the existence of an ancient 28 x 13 + 1 lunisolar calendar is non-existent, because it is impossible. Modern attempts to push such a calendar lack any credible mathematical support. If we accept that the calendar is not lunar and apply Gregorian leap year rules to this calendar then it would function quite well as a solar calendar. I believe I have further shown that many lunar only calendars are also mistaken for lunisolar, such as the North American tribal ones. There is much misinformation in this area, and people using them must remember that the 12/13 months have to be variable and follow blue moons (resulting in a Metonic calendar). Some of these calendars could well be half recorded Metonic systems, but up to this point, I have not seen that suggested or documented and they certainly couldn’t have used 28-day months. The claims of our half Native American commenter are either very misinformed and confused, or out and out lies. Likewise, the clickbait content of Very simply poorly researched material that 2 minutes with a calculator can rip apart. And as for the so called “Celtic tree calendar”, well, that is one of the most prolific pieces of fake lore there is, and it simply refuses to die. It has absolutely no historical basis, no mathematical support, and no astronomical support. If you want a real ritual calendar with genuine thought put into it for contemporary pagans, well you are in the right place.