m. c. de marco: To invent new life and new civilizations...

Durin’s Days are coming soon, many thanks to Alan Eliasen and Frink. This entry is just a note on the interpretation of “autumn” in the definition of Durin’s Day and of the dwarves' New Year in general. To recap, in The Hobbit, the dwarves' New Year’s Day is said to fall on the last new moon of autumn, where autumn is unfortunately not defined.

As I see it, we have several choices for defining autumn, listed for clarity in reverse order of their start dates:

1. The astronomical season running from the autumnal equinox (around September 22nd) to the winter solstice.
2. The meteorological season running from September 1st to December 1st.
3. The traditional season of approximately 90 days running from the cross-quarter day before the autumnal equinox (around August 7th) to the one afterward.
4. The Elvish season yavie, translated as “autumn” or “harvest” in Appendix D of The Return of the King, which, depending on your interpretation of the preceding season, may start as early as July 27th or as late September 1st.

To expand on #4, if we assume that the 72-day summer season laire that precedes yavie is centered around the solstice, then yavie itself begins around July 27th [#4a]. If summer begins on June 1 (as the Encyclopedia claims), yavie would begin on August 12th [#4b]. If summer begins at the solstice, then yavie begins 21 days later, around September 1st [#4c]. In the latter cases, yavie is still not equivalent to the meteorological or traditional seasons because both it and the second fall season following it were only 54 days long. Since the dwarves' new year depends on the date when “autumn” ends, #4 remains the earliest autumn on the list for our purposes regardless of interpretation.

The source material does not obviously favor any of the four options, but Graham Lockwood describes the ample but subtle evidence for #3. Last night, however, I did not have that link and we were forced to guess at an autumn that might make some sense out of the Encyclopedia of Arda’s crazy dates for 2001–2011.

Autumn #1 pushes the New Year well into December, which was out of the question when comparing to the Encyclopedia. (It puts New Year’s day on the autumnal equinox itself for 2006.) Autumn #2 did not occur to us at the time, but it always puts the New Year in November, where the Encyclopedia never puts it.

Alan came up with Autumn #3 independently and I concurred, being a fan of cross-quarter days and seasons. This option, however, does not quite produce the dates given in the Encyclopedia, even when adjusting the definition of “New Year’s Day” to require that the entire month (not just the new moon) fall within “autumn.” The approach seemed promising at the time, though.

I found Autumns #4 in Appendix D this morning; it was also unavailable last night. When I set Frink to calculate New Years based on #4a, #4b, and #4c tonight, I finally found the source of the Encyclopedia’s madness: #4c. My calculated values for #4c match their New Years. (For the record, #4a and #4b, the more promising interpretations, do not match.)

[Addendum: Starting from the Encyclopedia’s otherwise inexplicable claim (A) that the last new moon of autumn occurs within two weeks of October 6th, we get autumn ending on October 20th, and therefore beginning 54 days earlier, around August 27th, only a few days off from option #4c, September 1st—possibly because of rounding errors and the elves' interpolated days (which I ignored). There’s a two-week difference between the end of autumn as defined by (A) and the Encyclopedia’s own dates for yavie of August 12th through October 4th. I think this represents a misunderstanding of the dwarven New Year definition rather than a hole in my yavie theory, since the date of October 6th is otherwise too much of a coincidence.]

It seems quite unlikely that the dwarves would use the elves' (anything, never mind their wacky) six solar seasons, when four solar seasons were the norm for man and hobbit alike. Certainly the dwarves are more likely to do their own thing in cultural matters rather than haring after the elves like a bunch of Numenoreans.

I’m not saying that this is a well-defined lunar calendar into which Tolkien put his dwarf-loving heart and soul. I doubt he paid it much attention. It is deeply misguided to announce the lunar New Year based on a future solar event (the one heralding winter) rather than a past one. But perhaps we can excuse the dwarves by saying they counted a certain number of moons after observing the equinox, and the last one of the count was New Year’s Day. If we’re feeling generous. It’s certainly more believable than the idea that Tolkien had some elaborate dwarven calendar in his head that he somehow never wrote down, unlike all his other microscopically-documented ideas.

Next up: determining whether a given New Year’s Day is also Durin’s Day for your location.