Introduction

After reading *The Lord of the Rings* many people grew
fond of this wonderful book, of time and events depicted in it, so much that
they wished to replenish their red-letter day list with anniversaries of memorable
evens of the War of the Ring. At least the Fall of Sauron and Bilbo and Frodo’s
birthday became tolkienites’ favourite holidays. Good for us that Tolkien brought
together all the events in *The Tale of Years *listed by date. Thus, relying
on a clear statement, Tolkien fans celebrated the Fall of Barad-dur at March
25 and Bilbo and Frodo’s birthday at September 22 as it is written in LotR.

It is a pity that not all of them paid enough attention to Appendix D* *where
the calendars of Middle-earth are scrutinized. In fact, they should have paid
attention to these words: “In the above notes, as in the narrative, I have used
our modern names for both months and weekdays, though of course neither the
Eldar nor the Dúnedain nor the Hobbits actually did so. Translation of
the Westron names seemed to be essential to avoid confusion, while the seasonal
implications of our names are more or less the same, at any rate in the Shire.”

Those who neglected any accuracy argued: “since it is written “March 25” then
it is March 25 indeed!” and celebrated their favourite holidays in stated days
of stated months. In their own way they are right, many holidays (for example
the religious ones) do not fall on exact anniversaries of the events they commemorate.
But this example is suitable for holidays where one needs only to unite people
to commemorate something significant, not to celebrate its historically accurate
anniversary. As for the War of the Ring’s events, I hold the opinion that if
Tolkien struggled to write down how to correlate Middle-earth and our calendars
his work must not sink into oblivion. Thus my credo is “the 25^{th}
of Rethe have no more in common with the 25^{th} of March than with
the 2^{nd} of April”. Read on and you will learn why. By the way, below
I will use original names for middle-earth months “to avoid confusion”.

But I hope that the readers became interested in how exactly middle-earth and
our months contemporise. In the abovementioned paragraph Tolkien wrote that
our New Year’s Day (that is the 1^{st} of January) corresponds to the
9^{th} of January (that is the 9^{th} of Rethe) in the Shire
calendar. Here comes the most interesting part – the calculations. I hope that
you have already read Appendix D* *and are able to follow the thread of
my thoughts. Let’s begin with the Shire calendar, since all the dates in LotR
(months and days) are given in this reckoning.

The Calendar of the Shire

A short note about Sh.c.: its year was equal to our one by its length, all
the month were 30 days long but there were five days outside the months – three
in the middle of the year (1 Lithe, the Mid-year’s Day and 2 Lithe), one at
the end (1 Yule) and one at the beginning of the year (2 Yule). In every fourth
(leap) year, except the last year of a century, one additional day outside the
months, the Overlithe, was added. The reckoning in the Shire began in the year
1600 of the Third Age, but “to avoid confusion” I will use the reckoning of
Gondor, as in *The Tale of Years* (that is 3018 instead of 1418).

So the Shire year starts not from Afteryule 1 but from a day outside the months, 2 Yule – that is virtually Afteryule 0. Thus Afteryule 9 is the tenth day of the year. Moreover we know that the last day of the year is 1 Yule, a day outside the months too. Thus a new year in the Shire calendar (Sh.c.) starts at December 23 in Gregorian calendar (G.c.). See the table:

Shire | DOM | Afteryule | |||||||||

Yule 2 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | |

Greg. | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 1 | 2 |

December | January |

But you may object - Tolkien wrote that the Shire calendar was in advance of our one by ten days, not eight! “…By some ten days”, to be exact. Why “some”? Read on.

Next problem we encounter trying to compare our calendars is that the length of our months is different. A lot of people used the abovementioned correspondence forgetting that already the second month, Solmath, has 30 days, unlike 28 days in our February in G.c., and that way the advance will amount to nine days; the third month, Rethe, will be one day shorter than our March, the advance will amount to seven days. And so forth, thusly in each month the correspondence will differ. Month ‘joints’ are shown in the table below (for non-leap years):

Shire | DOM | Afteryule | Solmath | Rethe | |||||||||||||||

2 Yule | 1 | ... | 8 | 9 | ... | 30 | 1 | ... | 9 | 10 | ... | 30 | 1 | ... | 7 | 8 | ... | 30 | |

Greg. | 23 | 24 | ... | 31 | 1 | ... | 22 | 23 | ... | 31 | 1 | ... | 21 | 22 | ... | 28 | 1 | ... | 23 |

December | January | February | March |

Shire | Astron | Thrimidge | Forelithe | DOM | |||||||||||||||

1 | ... | 8 | 9 | ... | 30 | 1 | ... | 8 | 9 | ... | 30 | 1 | ... | 9 | 10 | ... | 30 | 1 Lithe | |

Greg. | 24 | ... | 31 | 1 | ... | 22 | 23 | ... | 30 | 1 | ... | 22 | 23 | ... | 31 | 1 | ... | 21 | 22 |

March | April | May | June |

Shire | DOM | Afterlithe | Wedmath | Halimath | ||||||||||||||||

MYD | 2 Lithe | 1 | ... | 6 | 7 | ... | 30 | 1 | ... | 7 | 8 | ... | 30 | 1 | ... | 8 | 9 | ... | 30 | |

Greg. | 23 | 24 | 25 | ... | 30 | 1 | ... | 24 | 25 | ... | 31 | 1 | ... | 23 | 24 | ... | 31 | 1 | ... | 22 |

June | July | August | September |

Shire | Winterfilth | Blomath | Foreyule | DOM | |||||||||||||||

1 | ... | 8 | 9 | ... | 30 | 1 | ... | 9 | 10 | ... | 30 | 1 | ... | 9 | 10 | ... | 30 | 1 Yule | |

Greg. | 23 | ... | 30 | 1 | ... | 22 | 23 | ... | 31 | 1 | ... | 21 | 22 | ... | 30 | 1 | ... | 21 | 22 |

September | October | November | December |

In the leap years of the Shire calendar (in 3020, for example) its dates advance one day forward relating to our ones starting with Overlithe; in the leap years of the Gregorian calendar the dates of S.c. recede one day back starting with Rethe 8.

As our Japanese friends (namely Makoto Takahashi) noticed, the Mid-year's Day
which is according to Appendix D must correspond as nearly as possible to the
summer solstice (June 21) actually fall on June 23. But our Japanese colleagues
therefore suggested to shift the beginning of the year in the calendar pf the
Shire one day forward or to throw off every 16^{th} Overlithe (that
is in every fourth leap year), but I regard such extreme measures as unjustified,
since (a) Tolkien wrote "as nearly as possible to the summer solstice",
not "exactly to the summer solstice", (b) Sh.c. was duplicated from
the King's Reckoning in year 1600 and haven't been adjusted since then, while
Stewards Mardil and Hador added three days in total to correct the Deficit.
It turns out that the Shire calendar cannot be absolutely correct and its discrepancy
between the Mid-year's Day's and the summer solstice is not to be wondered.

The Shire calendar is not suitable for reckoning our time at all. In comparison with our calendar it is rather inaccurate – a century in Sh.c. is one day shorter than in G.c. and we have no knowledge that this deficit was corrected somehow. In the Third Age it was small enough, but on a scale of 2000 years the calendar of the Shire will retard up to seven days and the gap will be increasing. But in the Third Age all the dates are correct, don't worry!

That's all, now you may easily celebrate the Fall of Sauron on March 18 and
Bilbo and Frodo's birthday on September 14 in the firm belief that you celebrate
the *anniversary*.

The Calendar of Imladris

Much more mysterious and fair was the elven calendar that Tolkien revealed to us…

A short note about E.c.: as Samwise remarked, the Elves who had more time at
their disposal reckoned it in longer periods. An elven 'year' – *yén
– *actually* *means 144 of our years, so we better call it a century.
But they also observed a 'normal' solar year – *coranar *or *loa – *of*
*365 days, Such a year they divided into six seasons – four of 54 and two
(the second and the fifth ones) of 72 days. See Appendix D for the details.
They had five days outside the months, just as the Hobbits did (but the latter
did not invent them, of course), once in twelve years three days in the middle
of the year – *enderi – *were doubled. In the last year of every third
*yén *such doubling did not occur. One *yén *consisted
of 52596 days. For the convenience I will use year numbers in the Gregorian
reckoning (though an elven year starts approximately three months later). And
remember that *yestarë *is the first day of the year, and with it
shifts the whole year (and all the following years, until another leap-year
happens – in G.c. or in E.c.).

In Appendix D Tolkien wrote that the Elves' New Year falls on the sixth day of Astron. Now we can easily get the corresponding date in the Gregorian calendar – it is March 29. But, as you know, that is a date for non-leap years.

In nowadays more and more of those who associate themselves with the Elves want to reckon the time in their way. But all of them came across one logical obstacle: we have no starting point. In his letter 211 Tolkien wrote about how long time ago did Barad-dur fell: "I imagine the gap to be about 6000 years: that is we are now at the end of the Fifth Age, if the Ages were of about the same length as S.A. and T.A. But they have, I think, quickened; and I imagine we are actually at the end of the Sixth Age, or in the Seventh." So it is the Seventh Age, all right.

But the problem is deeper – from when we are to calculate leap circles. Because of the difference betweem leap circles in E.c. and G.c. these calendars periodically shift from one to three days relatively to each other. So for accurate calculation we must determine when did the Seventh Age actually began.

Here only the common sence can help us. We know that each Age began and ended with events of global scale which importance for the whole Middle-earth deserved starting reckoning a brand new Age. Such events were the First Sunrise, the Fall of Morgoth, and the Fall of Sauron etc. Alas, we do not know for certain what did Tolkien thought about the Seventh Age, but I will proceed from the assumption that such an even can only be the Birth of Christ.

Tolkien was a roman catholic, and the meaning of Christmas for him cannot be overestimated. I am sure too that there will be no event of greater importance "tenn' Ambar-metta" and it deserves to call the Ages before it "the Elder Days". I am sure that Tolkien would agree.

Anyway, alternative events like the end of WWII and Tolkien's birthday are no good at all.

So let's assume that the Seventh Age began 2000 years ago. 13 Elven *yéni
* have passed since then: 52596*13 - 4*3 = 683736 days, where 52596 is the
number of days in one *yén*, 13 is the number of *yéni
*passed since 1 A.D., 3 is the number of *enderi *rejected from the
last year of every third *yén*, 4 is the number of such years since
1 A.D.

In the Gregorian calendar this gives approximately January 8 of 1873: 36524*18 + (365*72 + 17) + 7 = 683736 days, where 36524 is the number of days in one Gregorian year, 17 is the number of bissextile days added in 72 Gregorian leap-years since 1 A.D.

Remember that our task an this stage was to determine current *yén*,
so accuracy of *days *may be disregarded (to a certain extent): do not
consider the fact that 1700 and 1800 were non-leap years, Earth axis' displacement,
certain inaccuracy of the Elven calendar, relying entirely on the Gregorian
calendar's sufficient accuracy. So, 13 *yéni *have passed and now
we're in the 14^{th} *yén*!

Greg. years: |
1 | 145 | 289 | 433 | 577 | 721 | 865 | 1009 | 1153 | 1297 | 1441 | 1585 | 1729 | 1873 | 2017 |

Elven: yéni |
| 1 | | 2 | | 3 | | 4 | | 5 | | 6 | | 7 | | 8 | | 9 | | 10 | | 11 | | 12 | | 13 | | 14 | | 15 |

Now without much wisdom we can calculate that 128 years have passed since 1873.
So now we're in the 129^{th} *loa *of the 14^{th} *yén*.

Since the first *yén *began on March 29 of year 1 A.D., the 14^{th}
*yén *began on March 29 of year 1873. This can be calculated with
great accuracy: century length is equal in both calendars, in G.c. 10 days were
omitted (from October 5 to October 14) in 1582 and February 29 was not added
in 1700 and 1800, in total giving 12 days; in E.c. *enderi *were not doubled
four times (in the last year of every third *yén*), in total giving
12 days too. It means that at the beginning of the 14^{th} *yén
*these two calendars (and their leap circles – at that moment) were perfectly
synchronised.

1876 was a leap year in the Gregorian calendar, it means that February 29 was
added. It means that G.c. outrunned E.c. for one day. So *yestarë*
retarded from March 29 for one day. One more day was added in 1880 – and *yestarë*
became two days behind. In 1884 this shift was increased for one more day, thus
archieving three days. But already in 1885 (*loa *12) elven leap circle
was accomplished adding three more days in the middle of *loa*, so in 1886
(*loa *13) E.c. made up for lost time, catching up with G.c. so that *yestarë
*fell on March 29 again.*

**We need to remember that the bissextile shift occurred in the middle of
the year, so all the dates after enderi fell into place immediately in that
leap year, but the dates before enderi (such as yestare) did so only in the
next Gregorean year.*

*And adding the bissextile day in G.c. affected all the dates in E.c. that
followed February 28 in G.c. – that is the end and of the previous an all the
next elven year (and so on).*

These are the periodical shifts I was talking about. It happens during each
twelve-year period, that is why elven dates are not rigidly fixed in our calendar.
Such shift we are to take into account in our calculations, since elven dates
like *yestarë *admittedly fall into place only once in twelve years.

We also need to remember about, so to speak, anti-leap circles: in E.c., as
we know, *enderi*, are not doubled in every last year of every third *yén*,
and in G.c. February 29 is not added in century years wich are not divisible
by 400 (that is in 1700, 1800, 1900, 2100 A.D. etc).

Since 1872 in G.c. elven leap years were 1884 (*loa *12) , 1896 (*loa
*24), … 1992 (*loa* 120). But in 1900 in G.c. there were no 'planned'
February 29, so E.c. outrunned G.c. for one day forward after three additional
*enderi* of 1908, so the *yestarë *of 1909 fell on March 30…

Gregorian years: |
1884 | 1896 | 1908 | 1920 | 1932 | 1944 | 1956 | 1968 | 1980 | 1992 | 2004 |

Elven leap :loar |
12 | 24 | 36 | 48 | 60 | 72 | 84 | 96 | 108 | 120 | 132 |

In 1994 *yestarë *again fell on March 30. In 1996 (which was a leap
year in G.c.) *yestarë *occurred one day earlier – on March 29, in
2000 and 2001 (for the same reason) – on March 28. It also can be calculated
with great accuracy: before 1873, as I have said, two calendars ran evenly,
from 1873 to 1993 in G.c. 365*120 + 30*1 (without February 29 of 1900) days
passed and in E.c. - 365*120 + 8*4, so everything is all right, the shift amounted
two days.

So on March 28 of year 2001 A.D. we entered the 129^{th} *loa *of
the 14^{th} *yén *of the Seventh Age.

Greg. years: |
1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 |
2002 | 2003 | 2004 |

Elven :loar |
120 | 121 | 122 | 123 | 124 | 125 | 126 | 127 | 128 | 129 |
130 | 131 | 132 |

Addendum for our descendants and others

An instruction for calculating *yestarë *in this century: 2004 (*loa
*132)* *will be a leap year in both G.c. and E.c., that is *yestarë
*will fall on March 27, but all the dates in E.c. after doubled *enderi
*will be shifted forward for three days. Hence in 2005 *yestare *will
fall on March 30, in 2008 – on March 29, and in 2012 – on March 28. 2016, which
will be the last year of the 14^{th} *yén*, will be a leap
year in both G.c. and E.c., so the situation will be identical to 2004. And
2017 will be the first year of the 15^{th} *yén*, its *yestarë
*will fall of March 30 – these are the consequences of February 29 absent
in 1900.

The Calendar of Gondor

Third calendar we are interested in is the New Reckoning, established in Gondor in the Fourth Age. What is known about it? As it was before, our main source is Appendix D: New Reckoning was the revived King's Reckoning (used alredy in Númenor).

A year consisted of 365 days and was divided into twelve months: *Narvinyë*,*
Nénimë, Súlìmë, Víressë*, *Lótessë*,*
Náríë, Cermië, Urimë,* *Yavannië*,*
Narquelië,* *Hísimë*, *Ringarë* in Quenya
or *Narwain, Nínui,* *Gwaeron*, *Gwirith, Lothron,* *Nórui*,
*Cerveth*, *Urui*, *Ivanneth*,* Narbeleth,* *Hithui*,*
Girithron *in Sindarin, which approximately corresponded to our January,
February etc. (but a year began from Víressë, that is Rethe 25 in
Sh.c. – the Fall of Sauron) All the months had 30 days, each year began and
ended with days outside the months – *yestar**ë** *and *mettar**ë**
– *also there were three *enderi* or Middle-days (of which the second
was called *Loëndë*) between Yavannië and Narquelië,
that corresponded with Rethe 23, 24. 25 in Sh.c. In honour of Frodo Yavannië
30, which corresponded with Halimath 22 in Sh.c., his birthday, was made a festival,
and the leap-year was provided for by doubling this feast, called *Cormarë*
or Ringday.

Cormarë was not doubled in the last year of every century. The Deficit was adjusted in the last year of a millenium by adding several days, leaving a millenium deficit of 4 hours, 46 minutes and 40 seconds, but the exact anount of these days is not stated. Only in Tolkien's letter ¹176 it is written that in average a Númenorean year was 17.2 seconds slower that the tropical year (365.2422 days), which is even more accurate than the Gregorean calendar (which is 26 seconds faster). 17.2 seconds = 0.28(6) minutes = 0.004(7) ˜ 0.0002 days. And to make a regular Númenorean year (365.24) be 0.0002 days slower than the tropical year, one needs to add two days each 1000 years: 365.24 + 2 / 1000 = 365.2420

So the Deficit was adjusted by adding two days. We suppose that the Seventh Age began 2000 years ago. Comparing with the Gregorean calendar (which "adds" 2.5 days each 1000 years), during these 2000 years the Calendar of Gondor retarded from the Gregorean calendar for one day. According to that a table can be drawn (for non-leap Gondorean years from 2001 to 2099):

Gond. | DOM | Víressë | Lótessë | Nárië | |||||||||||||||

Yestarë | 1 | ... | 14 | 15 | ... | 30 | 1 | ... | 14 | 15 | ... | 30 | 1 | ... | 15 | 16 | ... | 30 | |

Greg. | 17 | 18 | ... | 31 | 1 | ... | 16 | 17 | ... | 30 | 1 | ... | 16 | 17 | ... | 31 | 1 | ... | 15 |

March | April | May | June |

Gond. | Cermië | Úrimë | Yavannië | DOM | |||||||||||||||

1 | ... | 15 | 16 | ... | 30 | 1 | ... | 16 | 17 | ... | 30 | 1 | ... | 17 | 18 | ... | 30 | Enderë | |

Greg. | 16 | ... | 30 | 1 | ... | 15 | 16 | ... | 31 | 1 | ... | 14 | 15 | ... | 31 | 1 | ... | 13 | 14 |

June | July | August | September |

Gond. | DOM | Narquelië | Hísimë | Ringarë | ||||||||||||||||

Loëndë | Enderë | 1 | ... | 14 | 15 | ... | 30 | 1 | ... | 15 | 16 | ... | 30 | 1 | ... | 15 | 16 | ... | 30 | |

Greg. | 15 | 16 | 17 | ... | 30 | 1 | ... | 16 | 17 | ... | 31 | 1 | ... | 15 | 16 | ... | 30 | 1 | ... | 15 |

September | October | November | December |

Gond. | Narvinyë | Nénimë | Súlimë | DOM | |||||||||||||||

1 | ... | 16 | 17 | ... | 30 | 1 | ... | 17 | 18 | ... | 30 | 1 | ... | 15 | 16 | ... | 30 | Mettarë | |

Greg. | 16 | ... | 31 | 1 | ... | 14 | 15 | ... | 31 | 1 | ... | 13 | 14 | ... | 28 | 1 | ... | 15 | 16 |

December | January | February | March |

In leap years February 29 is added in the Gregorean calendar, which shifts it one day forward until September 14 (September 15 in regular years) of that year, which is Yavannie 30 in the Calendar of Gondor. And since their leap circles coincide, this day, Cormare, doubles, recifying the shift.

© Elenhil Laiquendo, 2000-2003