Like the determination of place and direction, the calendar too is a child of the heavens. Calendar words and time words, today as at all times, go back to observations of the sky.
The short span of a month is defined by the moon in two ways: by its true circuit from star to star in 27 to 28 days, and by the cycle of its phases (from new to new or full to full) in 29 to 30 days. Our modern calendar months have nothing to do with the moon: they are divisions of the solar year.
The course of the year (representing the time taken for the earth to orbit the sun) can be fixed by the stars or the sun. The star year is reckoned from the first appearance (earliest rising) of a time constellation (such as the Pleiades) in the eastern night sky just before sunrise. This observation easily becomes inaccurate owing to haze and varying heights of horizon; also, the time when the marker constellation appears depends on the latitude of the site. A more reliable method is provided by the movement of the sunrise and sunset points along the horizon, which keep in step with the seasons. The modern astronomical year is reckoned from spring to spring, from one northward crossing of the sun over the equator to the next (equinoctial year); the Germanic year was, as we shall see, reckoned from solstice to solstice.
The old-fashioned objections to an independent development, among Germanic peoples, of the solar year, whose length (it is said) they would not have been able to calculate as a fixed number of days, fail to consider that the sky presents a different aspect to northern peoples than in the Mediterranean region, from which they are alleged to have received their calendar. The great difficulty of fixing the day of the solstice in Greece was complained of as late as 150 BC by their most notable astronomer, Hipparchos; by Greek accounts the sun stays in its solstice position for about 40 days. But in northern latitudes, with a much flatter path, this period of time is reduced to a few days, thus making it far easier to find the middle day of the solstice period. We have seen the simple fisherman Oddi Helgason in northern Iceland observing the solstice day exactly. It may also be assumed that Oddi’s method, like his reckoning of 365 days to a year, goes back to Norway, where he probably came from. Moreover, Christian traditions show that in the ancient North the solstice was universally regarded as one particular day, which governed the calendar and thus the holding of the Althing (Parliament). It is well known that the western Middle Ages, until the invasion of Graeco-Arabic astronomy, kept to the year points fixed by Julius Caesar and the Council of Nicea; Christmas for the Middle Ages fell on 25th December, and the year point on 21st December (Old Style), whereas in Iceland, long before Christianity was introduced, the shortest day was correctly placed in the middle of December.
Further, where people were anxious to hold their important festivals, which fell at full moon, at a fixed season of the year, in spring, autumn or winter, they had to reconcile their count of lunar cycles with the solar year. In this way experiments with leap years began, leading up to the bound lunar year. As mentioned in the introduction, this type of calendar is still denied to be an independent Germanic development.
The western Middle Ages, in other words the ancient Church, took over the Julian solar year from the Romans, and borrowed their bound lunar year (as used along with the solar year to decide the date of Easter) with its nineteen-year rule, from the Greeks. According to the creation poems in the Edda, the sun, moon and stars were given regular paths by the divine ones in order to measure out time; and these poems therefore tell us that the Germanic calendar was not taken from the Greeks or Romans, but directly from the divine order of the heavens: in other words it was based upon observation.
We have to investigate the historical sources still available to us. Some of these are older than all poems of the Edda, although the substance of those poems may go as far back in time as the oldest historical sources. We shall see that whereas the Middle Ages based their calendar on borrowing, and were penalized for turning their backs on Nature by a total inability to develop, and a conformity to rules that verged on the ridiculous, Germanic reckoning then and at earlier times owed its many-sided developments to independent observation of the sky.
As the constellation that marked the beginning of the star year, the Pleiades were most often used. It is possible that apart from the later name “Seven Stars” these were also known as the “Herd of Boars”. The emergence of the sky god’s sacred animal, the Yule boar and oath boar, after whom Odin’s wedge-shaped battalion, and the troops at the head of the army, were named [13], could very well serve as the start of a new year. When in the Middle Ages it was said that the cuckoo was silent as long as the Pleiades were invisible, it meant that the Seven Stars were invisible until they reappeared as a time marker in the darkness just before sunrise. Again, the expression “mother hen with twelve (not seven) chicks” appears to denote the start of a new year.
Another constellation that can be identified as a year-marker is Orion, with its Germanic names “the Rake, the Three Reapers, the Plough, Frigga’s Distaff”. By its early rising this magnificent figure told the farmer the beginning of harvest, the time to plough, and in the North the flax harvest.
There is no surviving evidence for the use of the star year in the calendar of the Germanic peoples. It must be assumed that it had already been dropped in prehistoric times, in favour of the more advanced types of calendar, the solar and lunar years.
Some 330 years before our era Pytheas (see Section II.9) asked for and received from the Norwegians in Trondheim the sun’s rising and setting points on the shortest day, along with the difference in position in higher or lower latitudes; and this was in the summer, when these phenomena were not visible. The sun’s position at the solstice was part of their stock of astronomical knowledge. Nine hundred years later, on the same Norwegian coast, but beyond the Artic circle, the sun’s return after the 40-day polar night was observed “according to custom”. This account too, written down by Prokopios about 550 AD, comes directly from the Norwegians themselves. Since the account states that the last 5 days (of the 365-day year) before the reappearance of the sun were kept as the chief festival of the year, it follows that in order not to miss this 5-day festival the days of the year would have to be counted carefully from 1 to 360. The counting of the 5 days proves the counting of the 360 days. Indeed, we have seen from the accounts already quoted how important this counting of days was to them, and that during the polar night, that is the first 35 days during which the sun stayed below the horizon, they used the transit of the moon and stars through the south point as an aid to marking off the days correctly. Here, then, we are dealing with a calendar, namely a solar year, based upon obsevation and having a length of (365 − 5 =) 360 days. (See Fig. 9.)
We come across essentially the same solar year of 360 days a few centuries later again in Iceland, settled by the Norwegians and at that time still Pagan. There it was reckoned as twelve 30-day months plus (in Icelandic style, as a result of a mistake made in the settlement period and resulting from a desire for a whole number of weeks, not five but) four extra days. Since this kind of calendar, like the Norwegian one, is altogether non-Christian and thus not an import into the North from medieval Europe, and moreover cannot have been devised in Iceland, it follows that the Icelandic calendar is probably just a corruption of an old Norwegian one, imported from Norway; whence in turn we may infer that the Norwegian solar year of 360 days was reckoned as twelve “months” of 30 nights each. Since the new year was fixed by annual observations, no intercalary day was necessary. The five remaining days, in which the chief festival of the year was held, were “leapt over” in the calculations after every 12 months (Icelandic hlaupa, that is, a leaping over); the new year began with the first day of the first month. This leaping over, instead of an extra (intercalary) day, is apparently of high antiquity.
In the Germanic region as a whole, as distinct from western and southern Europe, it appears that in very early times, perhaps even before the Roman weekday names were adopted in the 4th century, a year of thirteen 28-day months was in use, such as is found in neither Roman nor medieval Christian time reckoning. The continuous seven-day week, which was probably older in Asia Minor than in Rome, is already described in the Allemanic laws around 600 AD as “an old custom”. The well-known riddle of the year tree generally counts twelve boughs for the months, except that in the Germanic region, and the Finnish borders influenced by it, the year tree has thirteen boughs.
“A tree has thirteen boughs
and each bough has four nests
and in each nest seven young ones.”
or:
“I know a tree
high on the hills
with thirteen boughs
four twigs on each bough”, etc.
or in the Faroes:
“I know a tree
on top of the hill
with thirteen boughs.”
How deep-rooted this calendar was can be seen from the fact that in Germany Albrecht Dürer (in a letter to Jakob Heller written in 1508) was still counting 13 months to the year, as Norway did officially even in the 17th century [14]. The name “Thirteenth Day” (instead of “Twelfth Day”) for the Christian 6th January, as a prophetic day for the 13 (not 12) months, is a reminder of this.
It is certain that the Germanic regions, when they adopted the Roman god and planet names for the days of the week, did not at once adopt the Roman (that is Julian) calendar. This innovation was confined to the church’s sphere of influence, and Charlemagne’s twelve names for the months must be understood in this sense.
This calendar, in the form of 52 seven-day weeks, was introduced into Iceland about 870 AD along with the old Norwegian solar year (of twelve 30-night months and 5 days left over). However, people wished to begin not only every week, but every one of the thirteen months, and every year, with the same day of the week (Thursday). But since the 52 weeks amounted to only 364 days, the 365th had to be leapt over, just as the 5 extra Norwegian days were left over. In Asia Minor this leaping over of the 365th day was general. It must also have been the custom in Scandinavia, since the Norwegians, as we have seen, were well aware of the 365 days in the solar year. But in the confusion of the settlement period the Icelanders forgot this leap day, so that their calendar very soon gained appreciably over the seasons.
This, however, the Pagan Icelanders noticed very quickly for themselves from the sun’s movements, and it was Thorstein Surt in western Iceland who around 950 advised the Icelanders to insert a full week, every seven years at first, and then at shorter intervals as the length of the year became known more accurately. Thorstein’s extra week was based on comparison not with the Julian year but with the actual solstice. This calendar, unique in the history of time reckoning and better managed subsequently than the Julian one, lasted in Iceland until our own time.
The dependence of the new year (the first day of summer) upon the solstice is shown once again by a decision of the Pagan Althing in 999, by which the first day of summer in every year was to fall ten weeks before the Althing. In Christian times this interval was retained, when (in Julian terms) the Althing was placed on 24th June and the first day of summer ten weeks earlier on 14th April. The pre-Christian beginning of summer appears from this to be the evening when the sun’s upper edge vanishes at the WNW point, just as by other accounts the beginning of winter was defined by sunset at the WSW point, the “eykt point” (see Fig. 2 in the June issue).
The young Icelandic free state begins its time reckoning with two forms of the solar year. Of the moon, we hear no more.
Introductory note. The sun’s yearly course of 365 or 366 days includes twelve complete lunar cycles of 29½ days, which total 354 or 355 days. The solar year always begins on the same date. If (as happens only at long intervals) the solar and lunar years happen to begin on the same day, then within the first solar year there fall not only twelve complete lunar cycles, but 11 or 12 days of the thirteenth. The remaining 19 or 20 days of this last cycle overlap into the second solar year. Thus the first lunar year has 13 cycles, and the second lunar year begins 19 or 20 days after the second solar year. Now, assuming that the lunar year is required to start as close as possible to the start of the solar year, the start of the third lunar year can fall behind again by 11 or 12 days without coming ahead of the start of the solar year. Thus the second lunar year has only 12 months; it is a “common” lunar year as opposed to a “leap” year of 13 months. Such a variable lunar year, based on the attempt to reconcile it with the solar year, is often called a “bound” lunar year (as opposed to the “free” lunar year running independently of the seasons) [15].
In 725 the Anglo-Saxon church historian Bede [16] reported on the calendar of his Pagan forebears, who had migrated from the Germanic lands into Britain. They reckoned their year by the moon, and from time to time added a thirteenth month. This 13-month year used to be called Thrilidus, because in addition to the two months called Lida in the summer half year a third month of the same name was inserted.
The solar year began with “the turning of the sun towards longer days”, that is, with the longest night, which they called “mothers’ night” [17]. They divided the lunar year into winter and summer halves, and began the winter half year with the full moon Winterfyllith (winter full moon), corresponding roughly to the Julian October. Besides two summer months with the same name Lida (either “mild” or “travelling” month) they had two Yule months in winter. At the full moon that ended the first Yule month and began the second, they held their Yule festival, also called “midwinter” because it was taken as the precise midpoint of the winter half year. In relation to the solar year, this feast was movable like the full moon; but it always fell on the first full moon after the shortest day (“mothers’ night”), which in the 7th century fell on 18th December (not 25th December as Bede states).
Three full moon cycles after Yule, on the fourth full moon after the shortest day, there began the summer half year, in the middle of which the third Lida month was intercalated in lunar “leap years”. Bede has not left us any rule for intercalation, but we have a rule from Scandinavia that probably goes back to older Germanic times (see below).
The bound lunar year described by Bede is derived by him from the immigrant Germanic peoples, and thus from the ancient Saxons. The months Eostre and Hretha named by him are also attested in Germany; Yule month is current all over the Germanic North; the Goths also had two consecutive Yule months (and probably their festival at full moon). The month Thrimilki, assigned by Bede to May, in which cows could be milked three times a day (“given the fruitfulness of Britain or Germany”) can no longer be proved in Germany, but its name is still in use in Sweden and Norway.
We may conclude that this form of the bound lunar year was prevalent in the Germanic homelands before the Anglo-Saxon migrations, that is, in the first centuries of our era. In fact, it fits in with the accounts of Caesar and Tacitus at that period (see below).
Introductory note. The 25 or so little “sheep islands” (Faer Oer) in the Atlantic Ocean were settled in the 9th century by Norwegians. Amid the isolation of the Faroese people, wholly dependent on fishing for their livelihood, there developed a form of the bound lunar year that is unique on earth.
The Faroese name for the moon, which at the winter full moon in these latitudes (62° N) circles above the horizon for an average of 17 or 18 hours, was “night sun”. The months themselves were called “suns”; the new crescent “sun coming”; the full moon (when the disk begins to shrink) was “sun turning” or “full sun”. Like all Germanic peoples they distinguished winter and summer halves of their lunar year, which, like the Angles and Saxon and (as we shall see) the Norwegians, they bound to the shortest day.
Since the islanders’ livelihood depended on starting the fishing season at the right time (because the fish migrations begin regularly at the same time every year), they could not use the old form of lunar year that they had brought with them. The required insertion of a whole month every two or three years carried with it the danger of missing the shoals of fish in the main fishing month. Instead of a whole month, they therefore inserted half a month, thus beginning one year at new moon and the next at full moon.
The accounts say that “winter moon” is the one in the sky on the shortest day: the year is a common year if in addition the following summer month (“summer sun”) contains the longest day. Next year, however, “summer month” will extend over the longest day only if the required period of time has been intercalated, that is, if “winter moon” has been started not from new moon but from full moon. Through this constant alternation of new moon and full moon years things were adjusted so that the start of the main fishing month (einmánasólin) varied by only a few days each year. “Yule moon” is the moon that is in the sky on the shortest day.
Introductory note. How could people know at the start of a lunar year that during its course a thirteenth month would have to be inserted? – On which new or full moon was the new lunar year, or the summer or winter half year, to begin? – How could the beginning be fixed if the sky was overcast? – On which day of the lunar or solar year should people attend the high sacrifice or national moot? – These questions could be answered only by a definite rule for intercalation, which for every peasant in the Swedish forest, for every sailor, for every soldier and his commander, was adequate, easy to use, and reliable.
The shortfall of a 12-month lunar year from a solar year (Introductory note to 3.) amounts not to 11 but on average 11½ days; the excess of a 13-month lunar year is not 19 but 19½ days [18]. In the church calendar, which reconciles the solar and lunar years by a 19-year cycle derived from ancient Greece, the above differences are rounded down to 11 and 19 days; in the north Germanic calendar cycle of 8 years, the church’s “eleven rule” came into conflict with the rounding up to 12 and 20 days: the “twelve rule”.
The precise date of the Pagan Yule festival has not come down to us directly; we only know that in Norway in the 10th century it was shifted from its Pagan date, namely “midwinter’s day”, to the Julian 25th December. Things were different with the great Swedish sacrifice to Dis (the goddess of Fate), at a national festival in Uppsala; this, when sacrificial rites were abolished, was simply converted into the Disting Fair, keeping its old date on the full moon of Disting month. The Disting moon was an actual lunar cycle and took its name from the Pagan sacrifice to Dis. In Christian times the rule ran: Disting lies on the full moon following the first full moon after midnight of 6th/7th January.
Before the Disting moon the Yule moon invariably ran its course round the sky; the present Norwegian country rule still confirms this, when it says that the Yule moon is the one in the sky on 6th January, and then follows the Disting moon. But, just as the latter got its name from the Pagan sacrifice to Dis, so the Yule moon can only have got its name from the Pagan Yule festival. This fell on the full moon of every Yule moon and was therefore (like all Germanic sacrificial festivals) not fixed within the solar year, but a movable feast. The transfer of the Yule festival from its Pagan date to the Christian 25th December (proposed by Haakon the Good, but owing to Pagan opposition not carried out till later) was a transfer from the Pagan lunar year to the Julian-Christian solar year. The statement by Snorri Sturluson that Yule fell on the Pagan “midwinter’s day” is confirmed. Once again, the reference is to the full moon in the middle of the six-month winter half of the lunar year, as among the Anglo-Saxon and Faroese, and probably all Germanic races.
Now the 6th January mentioned in the rule is an ecclesiastical date, quite unknown in the Pagan calendar. The point in the Pagan solar and lunar calendar that later became the Church’s 6th January can be calculated by means of the Germanic “twelve rule”. The Church bases its calendar on an “eleven rule”, in that it calculates not 12 but 11 days difference between the 12-month lunar year and the solar year. The 6th of January is known in Germanic lands as the “Thirteenth Day”; on this day, 12 full days have elapsed since 25th December. Now there exists an east Finnish rule for intercalation borrowed from Scandinavia: there is to be a “leap” year, that is, 13 months must be counted in the coming year, if the first new moon falls before 6th January; and the significance of this date is that 12 full days have gone by since the mark day, here 25th December. The Finno-Scandinavian rule for leap years agrees with the Norwego-Swedish rule for Disting. We see, however, that 6th January can only be the dividing date because 12 (not 11) full days before that was 25th December, that is, the first day of Yule as a mark day, that is, start of the Julian solar year. Both these dates are purely Christian, but the calculation is inconsistent with the Church’s use of eleven, based upon Meton’s rules. However, if on the one hand this practice is not Christian, and on the other hand the Disting Fair kept to its Pagan date in Christian times, the only assumption left to us is that the Norwegian, Swedish and east Finnish twelve-rule, using 6th January, has preserved in its Christian-Julian framework the pre-Christian dividing date for the leap year rule. This conclusion can be verified quite conclusively by calculating from the date of the old Danish chief sacrifice at Lejre in Sjælland (Zealand), abolished in the year 934 AD.
As to this Danish national festival, Thietmar of Merseburg [19] reports around 1000 that it was held every 8 full years, with 99 heads of sacrificial beasts being offered each time, and this was “in January after the time when we (Christians) celebrate our Lord’s epiphany (6th January).” Thietmar is trying to express the Pagan date in Christian-Julian form. We see from this that 6th January played a decisive role even in Pagan times, which had no such way of expressing it. Interpreting this expression with the help of the Scandinavian twelve-rule, we see that it coincides with the full moon following the first new moon after the shortest day at that epoch (16th December Old Style); and that not only the east Finnish rule, but also the Pagan Danish Lejre rule around 900 resembles the later Norwegian-Danish rule for the Yule moon and Disting moon. Just as the three-day Yule festival in the North fell on the full moon following the first new moon after the shortest day, and Disting fell on the full moon following the second new moon, so the Danish festival fell, every eighth year, on the same full moon as Yule.
As with the Anglo-Saxons and the Goths, all these high sacrifices were held at full moon and were thus “movable” feasts. We understand why Thietmar did not simply express the time of the Danish sacrifice as a fixed date; but we must recognize that in saying it falls between 6th and 31st January he has described it as accurately as anyone can express, in terms of the Julian solar year, a movable date of the lunar year bound to the solar year.
The old Danish national sacrifice at Lejre, abolished in the 10th century, was held at Yule “after every 8 full years”. In Sweden there was a similar national sacrifice at Uppsala, which was also held “after every 8 full years”, admittedly not at Yule but “around the spring equinox”. The vagueness with which this time is given nevertheless allows the conclusion: the high sacrifice at Uppsala was also held on a full moon depending on the shortest day.
However, the times of these high sacrifices, both of which were 8-yearly sacrifices in the lunar calendar, must have been based upon the same rule. In the account by Adam of Bremen [20], the Swedish sacrificial heads were 72 in number, and it is clearly stated that nine of these were allotted to each of the eight days of the festival; at the Lejre sacrifice there were 99 heads. Since each 8-year interval contained 99 lunar cycles, it seems that the number of sacrificial heads, 99 every 8 years, contains a time symbol. Similarly the high sacrifice at Uppsala every eight years consists of an eight-part sacrifice with nine kinds of animal in each part. Thus the national sacrifice at Uppsala will have been fixed by the same twelve-rule as the Danish high sacrifice, the Pagan Yule festival, and the Disting at Uppsala.
This old intercalary rule, which we have already proved to be in use in the 10th century, has been handed down quite clearly, although at a late period, in Uppsala itself. At the Disting Fair in Uppsala in 1689, a well-known Swedish doctor, Olof Rudbeck, came across a grey-haired peasant with a runestave, and was told the following intercalary rule for the Disting full moon:
The moon steps 12 and 20 during Aun.
The last expression seems to be a special term for the period in which the moon’s “step” takes place. But the numbers 12 and 20 (instead of 11 and 19) are precisely those on which our rules and calculations are based, and which confirm the calendrical connexion found above between the Norwegian Yule moon, the Swedish Disting moon, the east Finnish and Scandinavian leap year, and the Danish and Swedish national sacrifices at Lejre and Uppsala.
Consequently, the rule told to Rudbeck would appear to be a deliberate statement of an eight-year rule. After an arithmetical series this rule returns to its starting point after every eight terms: if we assume that in the first year the solar and lunar years begin on the same day, then the beginning of the lunar year falls as follows:
| in | year | I | on | day | 1 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| +20 | = 13 | months | |||||||||||||
| „ | „ | II | „ | „ | 21 | ||||||||||
| −12 | = 12 | „ | |||||||||||||
| „ | „ | III | „ | „ | 9 | ||||||||||
| +20 | = 13 | „ | |||||||||||||
| „ | „ | IV | „ | „ | 29 | ||||||||||
| −12 | = 12 | „ | |||||||||||||
| „ | „ | V | „ | „ | 17 | ||||||||||
| −12 | = 12 | „ | |||||||||||||
| „ | „ | VI | „ | „ | 5 | ||||||||||
| +20 | = 13 | „ | |||||||||||||
| „ | „ | VII | „ | „ | 25 | ||||||||||
| −12 | = 12 | „ | |||||||||||||
| „ | „ | VIII | „ | „ | 13 | ||||||||||
| −12 | = 12 | „ | |||||||||||||
| „ | „ | I | „ | „ | 1 as above. | ||||||||||
With the ninth year the series begins afresh. The use of an eight-year cycle is historically confirmed by Lejre and Uppsala, but also the eight-year rule is prevalent in Swedish provincial law, in Swedish and Danish folksong, and in the old Norwegian songs of Thor; it is also attested in Germanic myth and even early Germanic verse drama.
The equating of 8 solar years to 8 lunar years plus 3 intercalary months fails at the end of a cycle by 1½ days, and after another 8-year cycle by 3 days. People would have to make a further intercalary rule, or else begin the series afresh by observation. Probably the latter was done. A general discussion and announcement of the calendar for the forthcoming year is known to us for certain from the Pagan Althing in Iceland. Incidentally the Church’s 19-year rule had fallen behind the true moon by some 2 days around the year 1000, and by 4 days in the 17th century: reason enough for the country people to hold fast to the old practical rules from Pagan times long into the Christian centuries.
The eight-year rule is in no sense a variant of the nineteen-year rule. Using the above arithmetical method, the eleven-rule does not return to its starting point until after 30 years (11 + 19). Quite apart from the error, already mentioned, of this rule in the course of centuries, agreement between the rule and the moon’s true phases every 19 years is only brought about by an artificial device (omission of a day every time, the so-called “saltus lunae”). Naturally the rule was not corrected by the Church. But just as the Norwegian and Icelandic solar year was kept in health by yearly observations, so too was the lunar year, by actual observations of the moon. The Disting Fair and the national sacrifices at Lejre and Uppsala were of necessity held at the time of true full moon, because people needed its light.
The ancient Greek eight-year cycle (oktaëteris) regulated the calendar and the sacrificial rites of ancient Greece for many centuries: the Delphic and Olympic periods were (like Lejre and Uppsala) dependent on it and its 99 lunar cycles. The twelve-rule of the Germanic North is however as little in evidence in Greece as in the Christian Middle Ages. The old Germanic eight-year cycle is of native origin.
The bound lunar year in Germanic countries probably goes back to very ancient times. In the great battles between the Romans and the Germanic Suevi in Alsatia about 58 BC, the Germanic commander Ariovistus was (according to Caesar and others) bound by religious custom not to give battle before new moon. Considering the vital importance of the coming battle, and considering that he could not transgress the law by a hairsbreadth, Ariovistus must have been in a position to fix the true occurrence of new moon (which is normally invisible) even if the sky was overcast for a week.
This was possible only through a tried and tested method of calculation. Tacitus reports explicitly on such calculations about 150 years later, when he says that the Germans “calculate” the number of days so as to take part in meetings, or begin important undertakings. From the accounts of Caesar and Tacitus, the two “tides”, new moon and full moon, must have been an integral part of a fixed calendar, no doubt a bound lunar year. The oldest high sacrifices of these West Germans – for instance the autumn festival of the goddess Tamfana [21] in Westphalia – were held at fixed seasons of the year, yet also at new or full moon. Thus the oldest Germanic calendar on record was based on a compromise between the lunar year and the sun’s course; but that is the hallmark of the bound lunar year, which only a few centuries later is seen fully fledged in the old Saxon and Anglo-Saxon territories.
If researchers have until now denied the independent observation and use of the solar year and bound lunar year, we can now recognize that the Pagan Germanic peoples had in fact developed both types of calendar in various forms, namely:
The Julian calendar was first introduced into the Germanic regions by the medieval Church; even then the Germanic calendars remained in use alongside it for many centuries, in some cases down to our own times [22].
Germanic calendars took the actual appearance of the sky as their foundation. Harmony between rules and observations was achieved in various ways, by intercalation of months, half months, or weeks.
The development of all types of Germanic calendar is based on independent observation of the sky.