The Days of the Week

Diês dictî sunt â deîs quôrum nômina Rômânî quibusdam stêllîs dêdicâvêrunt. Prîmum enim diem â Sôle appellâvêrunt, quî prînceps est omnium stêllârum ut îdem diês caput est omnium diêrum. Secundum diem â Lûnâ appelâvêrunt, quae ex Sôle lûcem accêpit. Tertium ab stêllâ Mârtis, quae vesper appellâtur. Quârtum ab stêllâ Mercuriî. Quîntum ab stêllâ Iovis. Sextum â Veneris stêllâ, quam Lûciferum appellâvêrunt, quae inter omnês stêllâs plûrimun lûcis habet. Septimus ab stêllâ Sâturnî, quae dîcitur cursum suum trîginta annîs explêre. Apud Hebraeôs autem diês prîmus dîcitur ûnus diês sabbatî, quî apud nôs diês dominicus est, quem pâgânî Sôlî dêdicâvêrunt. Sabbatum autem septimus diês a dominicô est, quem pâgânî Sâturnô dêdicâvêrunt.

St. Isidore of Seville (c.560-636), Orîginês 5:30

In Hebrew, the days of the week are simply numbered, except for the 7th, which is the Sabbath (Shabbat). In Arabic, the days of the week are also numbered, and the 7th day is still the Sabbath (asSabt), but the 6th day is now alJum'ah, the day of "gathering" (jum'), when Muslims pray at the mosque (jâmi'). In Modern Greek, the days are also numbered, and the 7th is still the Sabbath (Savvato), but the 1st day is now Kyriakê, the Lord's day. Interestingly, the 6th day in Greek is Paraskeuê, the "Preparation." This is actually the Jewish term, as preparation is made on the 6th day for the Sabbath -- preserved in a language today almost entirely of Christians, but formerly of many Jews in the Hellenistic and Roman periods.

While the Eastern Mediterranean languages reflect variations on a simple numbering of the days of the week, the languages of Western Europe all (except Portuguese) reflect names based on the names of the naked eye planets, which included the Sun and the Moon, either in a Latin version or a corresponding Germanic version. The 1st day in Latin is named after the Sun (Solis dies), but Christians also consider that the Lord's day (Dominicus dies), as in Greek. The 2nd day is named after the Moon (Lunae dies). The 3rd day is named after Mars (Martis dies). The 4th day is named after Mercury (Mercurii dies). The 5th day is named after Jupiter, or Jove (Jovis dies). The 6th day is named after Venus (Veneris dies). And the 7th day is named after Saturn (Saturni dies), though it can still be called the Sabbath (Sabbatum or Sabbati dies).

The Planets
SumerianBabylonianGreekLatinEnglish
NannaSinSelenêLunaMoon
EnkiNabûHermesMercuriusMercury
InannaIshtarAphroditêVenusVenus
UtuShamashHeliosSôlSun
GugalannaNergalAresMarsMars
EnlilMardukZeusIuppiterJupiter
NinurtaNinurtaKronosSaturnusSaturn
The Latin names of the Planets were simple translations of the Greek names, which in turn were translations of the Babylonian names, which go back to the Sumerians. The Egyptians had different assignments. Some interpretation was required for the Greek, and even for the Babylonian, translations, however. Nergal, for instance, was the god of war but also of pestilence and, especially, the Underworld -- overlapping with the Greek Hades. While Kronos was the father of Zeus, Ninurta was the son of Enlil. The Babylonians replaced the Sumerian national gods Enlil and Enki with the patron god of Babylon, Marduk, and his son, Nabû -- though Marduk was actually taken to be the son of Enki (called "Ea" in Babylonian). Ninurta, an obscure god inherited by the Babylonians, may have been identified with Saturn, the slowest moving planet, because, at least in one story, he was identified with the turtle. [note]

The Germanic version of the Latin day names has some correspondences and some differences. In English, the 1st, 2nd, and 7th days are still named after the Sun, Moon, and Saturn, respectively. The 6th day, Friday, looks like the name (Fria or Freya) of a Germanic love goddess, which would correspond to Venus (fri-, as in "friend," is a cognate of philein, "to love," in Greek), though the day is also said to be named after the goddess Frigg, who is also a goddess of love, and of the hearth (which would be Vesta rather than Venus in Rome). The 5th day, Thursday, named after Jupiter, who is a thunder god, in Latin, is named after a Germanic thunder god, well known as "Thor" in Norse mythology. Tuesday is named after Tiw, a god of law, but also said to be a god of war, which would match up to Mars. Wednesday is named after the king of the gods, who was Wotan in ancient German and Odin in Norse mythology. This has no obvious correspondence to Mercury, though Odin as a god of wisdom might suggest the role of Mercury in association with learning, and in Late Antiquity with esoteric knowledge.

The curious thing about the Latin names, clearly using the planets, is that the ancient order of the planets, rising from the Earth to the Fixed Stars, can be read off by starting with Monday and jumping every other day for two weeks:  Monday (Moon), Wednesday (Mercury), Friday (Venus), Sunday (Sun), Tuesday (Mars), Thursday (Jupiter), and Saturday (Saturn). One is left with the impression that the names were assigned in a kind of code, so that the Sun would come first in the week, but then the true order of the planets could be read off nevertheless. Saturn comes both at the end of the week and at the end of planets. The day that many people consider to the the 1st day of the week, Monday, is the first planet and does begin the sequence of planets.

While it is common to explain this peculiar sequence as going back to the Babylonian assignment of the planets to different hours of the day (e.g. David Ewing Duncan, Calendar, p. 47), I am suspicious that such an astrological mechanism actually does not go back all the way to the Babylonians. I have not found this explanation in critical sources about the Babylonians but largely in popular accounts which give no references and seem to assume that everything in astrology is originally Babylonian. We even have the testimony of Anna Comnena (1083-1153), who probably had access to texts that are now lost, that astrology didn't exist in Greek/Hellenistic times at all:

The art of divination is a rather recent discovery, unknown to the ancient world. In the time of Eudoxus [c.408-355 BC], the distinguished astronomer, the rules for it did not exist, and Plato had no knowledge of the science; even Manetho the astrologer [c.280 BC] had no accurate information on the subject. In their attempts to prophesy they lacked the horoscope and the fixing of cardinal points; they did not know how to observe the position of the stars at one's nativity and all the other things that the inventor of this system has bequeathed to posterity, things intelligible to the devotees of such nonsense. I myself once dabbled a little in the art, not in order to make use of any such knowledge (Heaven forbid!) but so that being better informed about its futile jargon I might confound the experts. [The Alexiad of Anna Comnena, translated by E.R.A. Sewter, Penguin Classics, 1969, pp.193-194]

The astrological account, especially, has two historical problems:  (1) the Babylonians used a 12 hour day, not a 24 hour day (cf. Georges Roux Ancient Iraq [Penguin, 1992], p. 364), and (2) the sequence of the planets was based, as Duncan says, on their "correct cosmological order," but the "correct" order of the planets was still a subject of dispute even among the Greeks for some time. The whole idea that there was an ascending sequence of planets up through the "spheres" of the heavens is probably entirely an artifact of Greek astronomy -- Pythagoras's "harmony of the spheres." Babylonian astronomy, like Egyptian, saw the sky as one dimensional, not three dimensional. My suspicion, therefore, is that numbering of the days is Middle Eastern and that the assignment of the planets to the days, as part of a larger development of astrology, is Hellenistic or, as Comnena says, even later. It definitely antedates Constantine's official introduction of the seven day week, given Duncan's example of a graffito diagram from Pompeii (p. 47), shown at right, which traces the sequence of days within a heptagon, around which the planets are listed in their celestial sequence, as this had been agreed upon by that time.

Now, however, I have found a scholarly source with a reasonable account. This is The Oxford Companion to the Year, An exploration of calendar customs and time-reckoning, by Bonnie Blackburn & Leofranc Holford-Strevens [Oxford University Press, 1999, 2003]. According to this source, the formulation of the planetary week goes back to Ptolemaic Egypt [pp.566-568]. Its first historical description is by the Roman historian Cassius Dio (Consul in 229). There, the hours of each day are associated with a planet in descending order, beginning with Saturn. Each day is ruled by its first hour, so the week begins with Saturday. Since there are 24 hours in the day, each day runs through all the planets three times (3x7), with a remainder of three (3x7+3 =24). Thus the first hour of the day following Saturday is ruled by the Sun, and becomes Sunday. Consequently, it is simply an artifact of the sequence of days generated in this way that the planets in ascending order can by read off every other day. Although the pagan planetary seven day week would thus have a pagan origin, it is noteworthy that the notion of a seven day week nevertheless might still be Jewish, since the example of the significant Jewish community of Ptolemaic Alexandria, already using a seven day week, could well have suggested the use of the seven planets for a similar cycle. It could be a nice instance of Hellenistic cultural syncretism. The status of Sunday as the first day of the planetary week, rather than Saturday, could be due either to Jewish and then Christian influence or to the circumstance that the Sun became the principal Roman state god as Sôl Invictus or Mithras under Aurelian and the Tetrarchy. Thus, the Oxford Companion mentions [p.567] that Constantine did not refer to Sunday as the "Lord's Day" or the "Sabbath," but as "the day celebrated by veneration of the sun itself" (diem solis veneratione sui celebrem). This is consistent with other indications that Constantine did not abruptly abandon the solar cult despite his establishment of Christianity.

The convention, becoming more common, to start calendar weeks on Monday, is a result of the Western European names, especially the German ones, which do not call Saturday the Sabbath -- or do not do so anymore in a recognizable way. Since Christians, especially Protestants, think of Sunday as the "Sabbath," the tendency is to number it as the 7th, rather than as the 1st, day. Familiarity with Greek or Arabic, or several Romance languages, however, would inform one that Saturday remained the Sabbath, as in Hebrew, even for Christians and Muslims.

The chart at right illustrates the two week period that counts off the classic ascending order of the planets, from left to right. It gives the names of the planets, in light blue, the names of the metals that alchemy associated with each of the planets (in roughly the appropriate color), the symbols for the planets and metals used in alchemy and still in astronomy, and the names of the days of the week, from top to bottom, in Latin, Welsh, French, Spanish, Italian, English, and German. The Welsh names are of great interest, since Welsh is the remaining language of Roman Britain. It preserves the Latin names of the planets more faithfully than even the Romance languages, which are actually descended from Latin, as Welsh is not.

The name of Saturday in Spanish and Italian is clearly derived from Sabbatum. The French name, Samedi, has the same origin. In Old French it was Sambedi, from Vulgar Latin Sambati dies. There are many cases where a b has become an m in "Sabbath," including Romanian Sambata, Hungarian Szombat, and even Persian Shambe (written Shanbe). In French, we have the interesting development that the final t has become confused with the -di (for dies) ending of all the other days of the week. The German name of Saturday, Samstag then looks like nothing less than a transformation of the French name into German, with German Tag, "day," subsituted for the -di element, and a genitive -s added to the root. In German, however, the -s may not be a genitive but a remnant of the final t, since the word in Old High German was Sambaz tac (compare English that, ending in t, with German das, ending in s). In the oldest attested Germanic language, Gothic, Saturday was Sabbato dags [Winfred P. Lehmann, A Gothic Etymological Dictionary, E.J. Brill, Leiden, 1986, p. 289-290].

German has otherwise dispensed with the traditional names of Wednesday altogether. Mittwoch would be "Midweek" in English

The gender of Dominius has remained masculine in Spanish and French but turned feminine in Italian. This may be because dies was occasionally used in the feminine in Latin, which ended up getting generalized in Italian. The vowels have gotten a bit scrambled in French Dimanche.

Welsh and Italian show parallel use of g's for the glides w/v and y/j, since Welsh reproduces Veneris as Gwener and Italian Jovis dies as Giovedi.

It is Welsh and English, curiously, that emerge without the use of either religious day name, neither Dominicus nor Sabbatum. It must be something about Britain. Indeed, the Oxford Companion opines that Britain was simply the furthest from the Middle Eastern origin of the Christian names.

Philosophy of Science

Philosophy of Religion

Philosophy of History

Home Page

Copyright (c) 1999, 2000, 2002, 2005 Kelley L. Ross, Ph.D. All Rights Reserved

The Days of the Week, Note

I have experienced some frustration over the years assembling the Sumerian and Babylonian names of the planets. Georges Roux (Ancient Iraq, Penguin, 1992), although very thorough about almost everything, is almost no help at all. N.M. Swerdlow's The Babylonian Theory of the Planets (Princeton, 1998), although the kind of book that could be expected to tell the whole story, actually doesn't give any Babylonian or Sumerian names for the planets. Only David Ewing Duncan's Calendar (Avon, 1998 [note]) gives the complete list of Babylonian names (p. 45), but no Sumerian ones. Giving the Sumerian equivalents of the Babylonian gods is then not always easy. I have not seen Gugalanna as the Sumerian equivalent of Nergal anywhere except in Diane Wolkstein and Samuel Noah Kramer's Inanna, Queen of Heaven and Earth (Harper & Row, 1983). It is Wolkstein and Kramer who also mention the story of the "Ninurta-Turtle" (p. 142). And then there is the awkward circumstance that Marduk and Nabû are not derived from Sumerian gods at all. Marduk, however, pretty clearly replaces Enlil, and his son Nabû has much the same function as a patron of learning as Enki/Ea. "Patron of learning," indeed, is the salient characteristic, since that seems to be the basis of the identification of Nabû with Hermes, and later of the identification of the Germanic Wotan with Roman Mercury.

Doubtless there are heavy duty sources that give the primary texts for all the Babylonian and Sumerian planet names, but I have not really wanted to turn this research into a full time job. That so many secondary sources fail to even list the equivalents (I think Swerdlow's oversight is unforgivable) is peculiar -- a peculiarity I can at least try to remedy.

Return to text

The Days of the Week,
David Ewing Duncan's Calendar [Avon, 1998]

The usefulness of Duncan's book is compromised by many, many errors, sometimes egregious ones. His treatment of luni-solar calendars is a disaster. The Babylonians did not use "seven years of thirteen months followed by twelve years of twelve lunar months" [p. 14]. The years with thirteen months were the 3rd, 6th, 8th, 11th, 14th, 17th, & 19th of a cycle, which is an even mathematical distribution, and a rule still used by the Jewish calendar, which Duncan says "intercalates a month every three years...though this system still leads to a gradual [!] drift that requires a second extra month to be added now and then by Jewish elders" [p. 14]. Where in the world did Duncan read this nonsense? What "Jewish elders" would that be? The "Elders of Zion"? Very shocking. Duncan also says that the Metonic year, with the Babylonian intercalation, runs "several hours fast" [p. 14]. No. It is off a day in only 219 years, which is more accurate than the Julian calendar (off a day in 128 years). This means it is only off a tenth of an hour (6.5 minutes) every year. Also, the Metonic year is longer than the tropical year (365.2467463 days against 365.2421988 days), which means that it runs slow, not fast, a confusion Duncan also has with the Julian calendar. Out of many other examples, just one more will do. Duncan says that "the Julian calendar the conquistadors brought with them was less precise" than the Mayan calendar they found [p. 18]. However, Duncan then describes the Mayan/Aztec calendar, which was no more accurate than the Egyptian calendar (a 365 day year) and considerably less accurate than the Julian. Why didn't he notice the problem with his own description? It is very sad to see a book published for popular consumption so confused and mistaken.

Return to Note

Return to text