Ancient Greek Calendars

Overview

The classical period of Greece lasted from the 5th until the middle of the 4th century BCE. During this time, the Greeks secured their independence by defeating the Persians. But conflicts between the Greek states never really disappeared. The long rivalry between Athens and Sparta finally led to the Peloponnesian War, after which the decline of classical Greece began.

In the latter half of the 4th century BCE, the northern kingdom of Macedonia dominated Greece, and the Hellenistic period began. The Macedonian king Alexander the Great conquered Persia and even invaded India, thus spreading the Hellenistic culture over vast areas.

Later, in the 2nd century BCE, the Romans interfered, claiming to reestablish the freedom of Greece from Macedonian domination. In the end, Macedonia and all of Greece became part of the Roman empire.

Calendar Cycles

In classical Greece, an eight-year cycle called ‘oktaeteris’ was known. It approximated the length of the tropical year with (365 + ¹/₄) d and the lunar year (i. e. 12 synodic months) with 354 days. Thus, eight lunar years have 8 · 354 d = 2832 d, and eight tropical years have 8 · (365 + ¹/₄) d = 2922 d. The difference between those two lengths is 90 days, or three 30-day lunar months. So in a period of eight years, a 30-day month would have to be intercalated three times to reconcile the lengths of lunar and tropical year.

Soon it was discovered that the assumed length of the lunar year was not correct. A more precise approximation of the lunar year (354 + ¹/₃) d resulted in a difference of (87 + ¹/₃) d between eight tropical and eight lunar years. In eight years, the ‘oktaeteris’ would thus be out of step by more than two days.

A more precise cycle is attributed to Meton. This cycle assumes 19 tropical years to have 6940 days, as well as 235 lunar months (110 of them ‘hollow’ and 125 ‘full’). Since 12 · 12 + 7 · 13 = 235, seven years of the 19-year cycle would have to have 13 months, the other years 12 months. The assumed length of a lunar month of (29 + ²⁵/₄₇) d is only about 2 min longer than the actual synodic month. The tropical year however was approximated with (365 + ⁵/₁₉) d which is more than 30 min too long and thus less precise than the value of (365 + ¹/₄ )d.

The length of the tropical year assumed for the 19-year cycle was about ¹/₇₆ day too long. Callippus multiplied the Metonic cycle by four (4 · 19 a = 76 a)and removed one day so that 76 years had 6940 d · 4 − 1 d = 27759 d. Thus, a tropical year of (365 + ¹/₄) d and a lunar month of (29 + ⁴⁹⁹/₉₄₀) d were assumed. The assumption for the lunar month is only about 22 seconds longer than the actual length of a synodic month.

Interestingly, these cycles never seemed to have been used for the civil calendars in classical Greece.

Greek Civil Calendars in the Classical Period

Virtually every Greek state used its own calendar with different month names, beginnings of the year, and intercalations. However, most of the calendars followed the same principles.

The Greeks used lunisolar calendars with years of 12 or 13 months. A month could be ‘hollow’ or ‘full’ having 29 or 30 days respectively. Intercalations seem to have been done as needed and sometimes arbitrarily. In any case there was no fixed pattern although several cycles were known to the Greeks.

The years were named after the holder of a certain office whose term lasted one year. In Athens this was one of the archons, in Sparta one of the ephors. The counting of years according to an era only became widely used after the Roman conquest.

The twelve months of a common year bore different names from city to city. For most of the states, neither all the names nor the exact sequence of the months are known, as are often the beginning of the year and the intercalary month (i. e. the month repeated in leap years). The Macedonian calendar in particular is very poorly attested in the time before its adaptation in several regions during and after Alexander's conquest.

The description of a few selected calendars of which we know more than one or two month names may suffice to give an impression of the diversity in Greek calendars.

Athens

There were three calendars in use in Athens. The most important was the civil calendar, a lunisolar calendar with years of 12 or 13 lunar months. The beginning of each month in theory was determined by observation, i. e. the day of the first visibility of the crescent in the evening became the first day of the month. This calendar was subject to manipulations of the archons. Thus, due to intercalations for political rather than astronomical reasons, this calendar could be out of step with the actual seasons. There is even evidence of single days having been repeated several times. Dates of this calendar are sometimes marked ‘kat archonta’.

Another calendar with the same month names but controlled purely by astronomical observation was also used, its dates having been called ‘kata theon’ dates. Dates in this calendar appear only rarely in inscriptions or manuscripts.

The third calendar was the prytany calendar and regulated the execution of certain offices by representatives of one of the so-called ‘tribes’.

Civil Calendar

The months of a common year of the civil calendar in Athens are compiled in the following list. In a leap year, Poseideon (marked with an asterisk) was repeated as intercalary month in 13-month years.

• Hekatombaion
• Metageitnion
• Boedromion
• Pyanopsion
• Maimakterion
• Poseideon^∗
• Gamelion
• Anthesterion
• Elaphebolion
• Mounychion
• Thargelion
• Skirophorion

The year began with the first visibility of the crescent after the first new moon after summer solstice. Intercalation followed no fixed pattern, although several cycles were well known in Greece.

The days within a month were not counted straightforward. The months seems to have been divided into three decades. The first day was named ‘New Moon’ (noumenia). The next 9 days were counted from 2 to 10 ‘of the waxing moon’ (histamenou). Days 11 and 12 were simply numbered, whereas days 13 to 19 were designated three to nine ‘over ten’ (epi deka). Day 20 could simply be called the 20th, but also the ‘first tenth’ (dekati protera). Days 21 to 28 were counted backwards from 10th to 3rd ‘of the waning moon’ (phthinontos). In a full month, the 29th day was the ‘2nd of the waning moon’. The last day of each month (full or hollow) was named ‘Old and New’ (heni kai nea).

Kata Theon Dates and the Prytany Calendar

Dates according to the observation-based lunisolar calendar sometimes appear. The calendar had the same internal structure as the civil calendar, i. e. the same month names etc. Its year also began with 1 Hekatombaion on the day of the first visibility of the crescent after the new moon after summer solstice.

The prytany calendar in the 5th century BCE seems to have had a 366-day year divided into ten periods. However, for the 4th century BCE, Aristotle describes it as lunisolar, and there is evidence that it, too, began its year on 1 Hekatombaion. The number of tribes increased from ten to 13 over the time.

Delphi

We don't know as much about the calendar of Delphi as we do about the Athenian one. There may have been more than one calendar systems, as in Athens. The names and sequence of Delphian months in the 4th century BCE is known, however.

The year was divided into two semesters, indicated by their number: first semester (prote examenos) and second semester (deutera examenos). Each semester consisted of six months, see the following list. Again, the month repeated in leap years is marked with an asterisk.

1st semester

• Apellaios
• Boukatios
• Boathoos
• Heraios
• Dadaphorios
• Poitropios^∗

2nd semester

• Amalios
• Bysios
• Theoxenios
• Endyspoitropios
• Herakleios
• Ilaios

The year began with the first visibility of the crescent after the first new moon after summer solstice, as in Athens. The designation of days within the month seems to have followed the Athenian scheme. However, original sources are only available for days 1 to 20.

Boeotia

The names and sequence of months in Boeotia from the 5th century on are shown in the following list. The last month, Alalkomenios, is reported to have been repeated in leap years in the 3rd century BCE.

• Boukatios
• Hermaios
• Prostaterios
• Agrionios
• Thiouios
• Homoloios
• Theilouthios
• Hippodromios
• Panamos
• Pamboiotios
• Damatrios
• Alalkomenios^∗

The year began around the winter solstice. Sources state that Prostaterios is the Athenian Anthesterion, and that Hippodromios equals Hekatombaion.

Days were presumably counted in a way similar to that of Athens. However, there are sources for 19 days only.

A parallel observation-based calendar resembling the Athenian kata theon also seems to have been employed in Boeotia.

Milet

The sequence and names of the months in Milet until the end of the 4th century BCE are shown in the following list.

• Boedromion
• Pyanopsion
• Apatourion
• Poseideon
• Lenaion
• Anthesterion
• Artemision
• Taureon
• Thargelion
• Kalamaion
• Panemos
• Metageitnion

The year began after the autumnal equinox but was shifted to the spring equinox around the end of the 4th century BCE. After that, the year began with the month Taureon. There is no certainty regarding the intercalary month. After the shift of the beginning of the year it might have been Artemision.

The count of days within a month seems to have followed the Athenian pattern, although it is not clear whether the days of the last decade were counted forward or backward.

Laconia (Sparta)

Although our knowledge about the Spartan calendar is very limited, the months known are listed here. For Sparta itself we know the names of nine months, the sequence of which has not yet been established:

• Agrianios
• Artemisios
• Geraistios
• Hekatombeus
• Heraios
• Karneios
• Yakinthios
• Phliasios
• Phlious

The following months are attested for Laconia (the portions in [] are reconstructed):

• Hekatombeio[s]
• [Kar]neiou
• Laphrioi
• Lykeiou
• [Yaki]nthiou

The similarity of some of the names suggests that there was a single calendar valid in Sparta as well as all Laconia. There are some hints on the sequence of the months, but so far it is not possible to reconstruct a complete sequence, or the beginning of the year, or the mode of intercalation. The latter lay in the hands of the ephors.

Hellenistic Calendars

In 336 BCE, Alexander the Great becam Macedonian king. Within the next 13 years he conquered Persia, Egypt and parts of India. With him he brought the Macedonian names of the months. Soon in many cities calendars with these names were established. A fixed 19-year leap year cycle in use in Babylonia was quickly taken over by many cities. After Alexander's death his empire disintegrated into several states.

Calendar in the Seleucid Empire

One of the states created by the disintegration of Alexander's empire was the Seleucid empire named after Seleukos I Nikator (r 312/305 BCE - 281 BCE), its first ruler. It first comprised only of Babylonia, where a luni-solar calendar with a fixed leap year pattern had been in use since the middle of the 4th century BCE, although the beginning of each month was still determined by observation.

Probably shortly after Babylonia had been conquered, the Babylonian calendar was adapted by simply renaming the months. The year began around the spring equinox.

Officially, the year began with Dios, which equalled the seventh month of the Babylonian year. In fact, often the year was begun with the first month of the Babylonian year, i. e. Artemisios = Nisanu. Intercalation followed the Metonic cycle, i. e. there were 12 common years of 12 lunar months each and seven leap years of 13 months. In six of these leap years, Xandikos was repeated. In the remaining leap year Hyperberetaios was repeated.

Within the first half of the first century CE a shift by one month took place in the relation between the Macedonian and Babylonian months. Now, Nisanu equalled Xandikos, resulting in the following table.

The shift presumably also affected the intercalary month. A second Dystros is attested in the second century CE and we may assume that Dystros and Gorpiaios have been the intercalary months after the shift.

The days were counted straightforward within a month, from 1 to 30 or from 1 to 29 in ‘full’ and ‘hollow’ months, respectively.

The years were counted according to the Seleucid era, see Seleucid Era (Epochs and Eras).

Ptolemaic Egypt

When Alexander conquered Egypt there had existed the Egyptian Calendar for centuries already. The invaders brought the Macedonian calendar with them. This calendar was modified by adopting a fixed leap year cycle, albeit a quite crude one, from ca 280 BCE on. In the latter half of the third century BCE the cycle seems not to have been used anymore, and from ca 200 BCE on the Egyptian calendar was used with Macedonian month names. For this calendar, see the section on the Calendar in Ptolemaic Egypt.

Calendars in Greek Cities after the Roman Conquest

In many Greek cities now part of the Roman empire Greek or Macedonian month names survived. But with the Julian reform (see Julian calendar) many calendars became aligned with the Julian calendar. However, the calendar was probably not completely aligned to the Julian calendar, particularly in large cities like Miletus, Ephesus, or Pergamum. Obviously the civil calendar deviated from the official one even in Roman times. The calendars described here are the official calendars.

In some calendars the names of the months of the Julian calendar were simply substituted with Macedonian or Greek names, as can be seen in the following table.

The days were counted straightforward.

Other cities felt obliged to let their year start with Augustus' birthday, which was a. d. IX Kal. Oct. = 23 September (1). The lengths of the months were adapted so as to start always on a. d. IX Kal. Such calendars were in use in the province of Asia, on Crete, in Bithynia and elsewhere. Some of them are listed in the table below.

A further group of calendars took its month lengths from the Egyptian calendar. These calendars have twelve months of 30 days each, followed by five epagomenal days. In leap years there were presumably six epagomenal days, as in the reformed Egyptian calendar.

The following table shows the correlation between these calendars and the Julian calendar in common years. For leap years of the calendars that have been aligned to the reformed Egyptian calendar, one can refer to the table given in the Egyptian calendar section. But there were also calendars with different beginnings of the year. The following table shows the correlation between some of these calendars and the Julian calendar. The first month of the year is the one that follows the epagomenal days.

There were also calendars that employed months of 30 or 31 days, without a short 28-day month or epagomenal days. These can be found in Tyre and Heliopolis.

Remarks

 ↑  top of page

http://www.ortelius.de/kalender/greek_en.php   © Holger Oertel 2000-2008; last change: 30 May 2009