Another way of reconciling solar and lunar cycles

While I was poking around doing research for this post (specifically, trying to find evidence of Native North American cultures having ecologically-named and possibly ecologically-based months on a lunisolar calendar), I stumbled across a way of reconciling the cycles of solar year and lunar month that I hadn't heard of before. I came across it in the Musqueam reference grammar by Wayne P. Suttles. The evidence for how the Musqueam calendar worked is kind of thin, but other Salishan languages use the system described (p. 517):

One way of adjusting a lunar calendar to the solar year is to insert a thirteenth month every four years or so, as the Jewish and Chinese calendars do. Another, probably more common way is to begin counting moons with some event that is determined by the solar year, count up to, say, ten moons, stop counting, and then begin again when the event that you started with comes again. In this way a "Sandhill Crane Moon" will correspond to the coming of the sandhill cranes.

Counting moons, leaving a gap, and beginning the count again with the observation of something determined by the solar year is just what was done by people upriver from the Musqueam. According to Diamond Jenness (1955, 7-9), the Katzie, who spoke a Downriver Halkomelem dialect very close to that of the Musqueam, counted ten months beginning with the arrival of the sockeye in August of the Gregorian calendar, leaving a period covering June and July with two names but not regarded as part of the count. The Chehalis, an Upriver Halkomelem-speaking peoople, as reported by Charles Hill-Tout (1904, 334-35), began a count of moons with the chinook salmon spawning in October, counted ten moons, and stopped counting in July, leaving a period called by a term said to refer to the coming together of the ends of the year. Neither Jenness nor Hill-Tout gave any reason for this uncounted period. But James Teit (1900, 339), describing the practice of the Ntlakapmux (Thompson), is very clear about it. The Ntlakapmux began a moon count with the rutting of the deer or some other animal in the fall, counted eleven months, and left a period as "the rest of the year." "This indefinite period of unnamed months," Teit explains, "enabled the Indians to bring the lunar and solar years into harmony".

This isn't relevant for my blind society, but is worth being aware of for more general calendar-making, as another way things can be done.

Blind society calendar system

I think all that research and brainstorming was enough to give me a good idea for how timekeeping works in my blind society (I really should come up with enough of their language to give them a name...). I think there's a chronological development over time.

In an early period, before the time that I'm focusing on, they marked parts of the year according to formal ecological timekeeping, and also had a 7-day week cycle. By the time I'm focusing on (but maybe not many centuries in the past), they had developed a 4-week unit, which I may as well call in English a month. Months were ecologically-named, and there was an ecologically-determined year start. They knew that a year normally contained 13 months. I'm still deciding how intercalation worked - whether they intercalated weeks or months. A solar year is roughly 365.25 days long (that's the Julian not-quite-correct approximation, but it will do for this purpose). A year of 13 28-day months is 364 days long - that's 1.25 days short. That would mean that an intercalary week would be needed every 5 or 6 years:

5 years * 1.25 days short = 6.25 days short in 5 years
6 years * 1.25 days short = 7.5 days short in 6 years

An intercalary month would be needed every 22 or 23 years:

22 years * 1.25 days short = 27.5 days short in 22 years
23 years * 1.25 days short = 28.75 days short in 23 years.

I suspect initially, their ecologically-based year start determination isn't very precise, which means realizing that an extra month is needed is more likely than realizing that an extra week is needed.

Days are not normally consecutively numbered within months; rather they're identified by which day of the week and which week of the month they are, and since weeks and months are aligned, this is a consistent designator. Weeks in a month are numbered; I suspect days of the week are generally referred to by name, not number.

The system of ecological timekeeping without months may still survive to some degree - either formally for ritual purposes, since ritual purposes tend to be conservative, or informally with people still making use of ecological period-names because they represent something in their experience that isn't captured by the system of months.

Eventually, after the period I'm focusing on, they develop more sophisticated measurement and observation technology, including methods for observing the sun with greater accuracy than just feeling on your body which direction its warmth is coming from, and they likely ultimately switch to a solar-based year start rather than an ecologically-based one, and may determine a calculated scheme of intercalation rather than an observational one.