Tuesday, November 08, 2005

When do we start saying ותן טל ומטר לברכה?

In EY we started saying it tonight on ז' חשון. Outside of EY we start 60 days after the tekufa, which today is assumed to be December 4th or 5th. This calculation has always bothered me greatly for the following reason. If you look at the Beis Yosef (and all the other poskim from that time)you will see that he writes that 60 days after the tekufa was November 22. However, the later acharonim all say, that was based on the Julian calendar, but now, since the world changed to the Gregorian calendar which meant that when they changed they skipped 12 days, we start on December 4th. I don't know much about these issues, but this seems very difficult to me. Those 12 days were a 1 time adjustment because the Julian calendar was not accurate enough, if so why should the 60 days from the tekufa change (it was a 1 time thing to fix the calendar drift)? Isn't the tekufa still around September 21st? In short, the Gregorian calendar just fixed the drift of the Julian calendar by skipping 12 days (once) and changing the leap year rules, why should that change when 60 days from the tekufa is?

I would really appreciate an answer or a pointer to someone who discusses this issue, this has bothered me for a long time.


AbbaFive said...

I don't have an answer - but it is patently obvious (although the reasoning is suspect, at best) that all post-16th century poskim uniformily concluded that the "tekufa" referred to by Shmuel(Eruvin 56a) was not the actual factual celestial event (the obvious choice) but rather "September 23rd (or 22nd) on the secular calendar".

Go know.

The more fundamental, less technical question, of course, is what are we doing praying for rain based on the needs of the Euphrates Valley (see Ta'anis 10a -shitas Chananya). The road not taken, in this case the "common sense" road - is expressed with great clarity by the Rosh "al atar" and in his teshuvos.

On the other hand, at least one person who had the nerve to raise this most-obvious problem with Rav Moshe lived to regret it: See Igros Moshe, Orach Chaim 4:17 and 5:7 and his strident defense of "minhag". (While his fundamental point about minhag is pretty much on the mark - and nothing new for us "Briskers" - his reference to the Avudraham is less than compelling).

chardal said...

The Julian calendar went off almost an entire day every 100 years. Thus, on the 4th of December, the earth is in the same position as it was on the 22nd of November 16 centuries earlier.

AbbaFive said...

Actually, that's not correct on several levels, and your presumption that this all just really "works out" once you have all the information is, lamentably, untrue.

First, without correction (by substracting an added leap year day) the calendar will move off the solstace/equinox days (3/21, 6/22, 9/22 and 12/22) one day every 128 years.

Second, December 4th of this year is in fact the same as November 22, 1582 - but not the same as November 22, 352.

And neither 12/4/2005 nor 11/22/1582 constituted 60 days after the vernal equinox (i.e., the tekufa) that Shmuel refers to in Eruvin.

The vernal equinox did occur, more or less, on 11/22/352 and then "slid" slowly over a bit more than a millenium.

Here's the math:

Shmuel (in Eruvin 56a) states that each "tekufa" (i.e. quarter-year/season) is 91 days, 7.5 hours - thus a year is 365 days, 6 hours.

Given Shmuel's access to scientific tools, this calculation turns out to be uncannily accurate (tell this next time someone says that Chazal used kiddush hachodesh because they couldn't figure it out at that time in history - utter nonsense!).

In fact, a year is 365 days, 5 hours, 48 minutes and 46 seconds.

Shmuel was off by 11 minutes 14 seconds. Not bad, not bad at all.

But, take 11 minutes 14 seconds and multiply it by 128 days, and you get a day (actually you get 1,437.9 minutes, not 1440, but close enough).

Multiply that by the gap between the year 325 (the year that March 21st was established as the vernal equinox) and when good old Pope Gregory XIII looked at this 1582 (a gap of 1,257 years) - and lo and behold, the vernal equinox has moved by 14,120.3 minutes - or 9.8 days.

Getting the vernal equinox back to March 21 from March 11th was Pope Gregory's objective (his interest is focused on something to do with an annual remembrance for a guy who gets killed right after a Passover seder), and he did it by adding 10 days to the calendar on October 4, 1582 (thus causing havoc on current contracts - a total bonanza for lawyers ... in 1582).


He questioned why all of this matters. Shmuel said 60 days after the Tekufa, not 60 days after "September 22nd on the Pope's calendar".

In answer to THAT question, I responded:
1. I don't know; good question.

2. The better question is the Rosh's question why we in the U.S. (let alone the absurdity of Australia - it's summer there!) pray for rain based on agrarian needs in Iraq.

3. With respect to "halacha lemaase", Rav Moshe's response that Minhag Avoseinu has taken over this issue.

The Observer said...

Actually the length of the year abbafive quotes is not Shmuel's, but Rav Adda's. The tekufah in question is not calculated using this value for the length of the year but with Shmuel's slightly longer rounding off of, 365-and-a-quarter days. This is the same year length underlying the Julian calendar, so the date of the tekufah suffers from the same drift as does that calendar. As it happens 23 Sept 2005 (Julian) was Tzom Gedaliah this year. 60 day after that takes you to 5 Dec 2005 (Gregorian), which is when we indeed start.

The details of how to do the calculation is given in Chapter 9 of the Rambam's Hilchot Kiddush HaChodesh.
Here an article dealing with the issue in much more detail.

AbbaFive said...

Excellent article - thanks for the link.

As to the accurate citing of the time period, as well as the amorah responsible for this shita, "Observer" is incorrect. The Gemora in Eruvin states, expressly:

אמר שמואל:.... ואין בין תקופה לתקופה אלא תשעים ואחד יום ושבע שעות ומחצה

As to Rav Adda (or Ada), Rav Moshe writes:

שנוהגין כל ישראל בדבר טל ומטר בחו"ל שהוא מיום ס' לתקופה לפי חשבון של שמואל (עירובין נ"ו ע"א) ולא לפי חשבון של ר' אדא שהוא שגגה

As to the math, Observer misses the point - entirely - which is pretty disappointing given that he only need read the article that he cited.

First, great for Tzom Gedaliah falling out on the 9/23 this year, but what does that have to do with the Tekufah? December 5th is also 60 days after my late, lamented dog's (Keli, obm) birthday (but only using a dog-year calendar), but that is not particularly relevant to understanding the halahcha of visain tal in chutz la'aretz.

Second, the "drift" is solved by the elimination of 3 leap years in each 400 year period (that Pope Gregory was one smart pope - or maybe he was just smart enough to listen to really smart advisers - we'll never know).

THAT adjustment - as well as the 6 hour shift discussed in Eruvin (usually fixed each 4 years by the addition of a 2/29) - will result in December 4/5 shifting to December 5/6 in the next century, again as the article points out.

Sorry for being harsh, but (i) this is a blog, and (ii) if you go out of your way to correct someone, I believe you assume the responsibility to be "correct".

bluke said...

Observer thanks for the link.

In other words, for the tekufa with regards to this issue we follow Shmuel's opinion which is the same as the Julian claendar and therefore the drift occurs.

The Observer said...

Sorry abbafive, I misread your post. Shmuel's annual period is indeed 4*91 days 7.5 hours, i.e. 365.25 days, as we've both said. And, as it goes on to say, this is the period that is used for tal umater. I.e. the same as the Julian year. Accordingly the tekufah is 9/23 Julian, which happens to have been on 3 Tishrei 5766. To get the date this year you need to convert from Julian to Gregorian. The whole point is that when the tekufah calculation was set up, the rabbanim deliberately used the known, inaccurate, Julian value. If the West had stayed on that calendar, there would be no drift in the calendar, but the 9/23 would fall increasingly distant from the true, astronomical, tekufah. As does the halachic tekufah.

In other words, to anwer bluke's question, we start 60 days after the halachic tekufah, not the astronomical one. Why? Perhaps because it is easier to calculate.

AbbaFive said...

Apology accepted Observer, but I think I disagree with your explanation to bluke.

You state: "The whole point is that when the tekufah calculation was set up, the rabbanim deliberately used the known, inaccurate, Julian value."

Shmuel lived very close in time to when the Julian calendar was "set" to March 21 - vernal equinox (the year 325)!

So, for Shmuel, his use of the Julian calendar, while deliberate, could not possibly be what you say - a deliberate reliance on an acknowledged innacurate system. In fact, we're certain that Shmuel fully believed it to be absolutely accurate!

It took centuries to recognize the impact of 11 minute gap - centuries after Shmuel lived.

Finally, although absolutely true that it is the underpinning of our current pesak - there is simply no way (that I am aware of) to discern that Shmuel was referring to the Gregorian calendar date of 9/22 with the word "tekufa" - the more obvious presumption (although admittedly not taken) is that he was speaking of "tekufa" as the celestial event.

So, although I concur that your view is plausible, it is not demonstrable nor is it the most logical. I disagree with you on two levels:

1. First, I disagree that Shmuel's statement is a clear reference to the then prevailing Christian calendar (although I acknowledge that others must have concluded as such - that conclusion is difficult to understand - hence Worker's initial question).

2. Second, I disagree (strongly) that Shmuel knew of - and nonetheless relied upon - an 11 minute annual innacuracy in the calendar that would only play out centuries later.


And to get to the date this year AND EVERY YEAR, you need to take the beginning of the Tekufa - a celestial event (9/23) then add 60 days then add 10 more days that were "subtracted" in 1583 (i.e., the difference in the two calendars).

The Observer said...

I never said Shmuel used or even knew about the Julian calendar. That calendar is just a convenient way to explain what is going on. All you need to have to get to the current discrepancy is Shmuel's then accurate tekufa plus his slightly too long estimate for the length of the year. Every year the date of the tekufah based on this drifts slightly later compared to the true tekufah. The chachomim knew that this was a different length to the solar year compared to the one used in the rest of the fixed calendar, i.e. Rav Adda's. And, apparently, for the purpose of tal umater chut l'aretz, this value was good enough.

AbbaFive said...

Oberserver states: "I never said Shmuel used or even knew about the Julian calendar."

But Observer previously stated:
"The whole point is that when the tekufah calculation was set up, the rabbanim deliberately used the known, inaccurate, Julian value."

I now understand from the post that Observer did not intend the word "innacurate" to refer to "the deliberate actions of the rabbanim" in the immediately prior clause of that sentence.

Ah, the indefinite, dangling modifier. Sloppy, sloppy.

And, for your "apparently ...this value was good enough" conclusion, I sincerely wish I agreed.

It just seems obvious to me that this is an innocuous mistake that played over centuries and centuries, the impact of which was not fully understood, and over time became Minhag Yisrael - v'ain lishanos.

I can live with that (I have a lot bigger fish to fry, such as the two flood and the two meraglim stories, basar kafui or those mean spirited nashim tzidkaniyos who were machmir a full WEEK (why not just a day?)).

Personally I prefer to avoid the apologetics that it either (i) is actually correct (choke, choke), or (ii) was understood and knowingly accepted.