When Sanhedrin existed, each month had to be sanctified by Sanhedrin after testimony of 2 witnesses (or automatically by Sanhedrin if no one appeared by the end of the 30th day.) Refer to Maimonedes Hilchot Kiddush Hachodesh for details.

Since each month was individually sanctified, I assume that it was possible to have several months each with 30 days or several with 29 days, unlike the alternating lengths that we have for most of the months in our current fixed calendar.

Was there any limit to the number of consecutive months having the same length? For example, in a leap year having 13 months, would it be possible to have a year that was 390 days long if all the months had 30 days? If not, what was the maximum number of months that had the same length?

A long run of one type or the other would eventually become obvious to anybody who witnessed the new moon, but if nobody witnesses and comes to Jerusalem to testify, the Sanhedrin could conceivably have a long run of default-30-day months. Is there a halachic limit?

  • 1
    8 of one type per year. Erchin 9b or so.
    – Double AA
    Commented Sep 11, 2015 at 2:58
  • @DoubleAA I'll try to view this later on. If it makes sense, please post as an answer. Would you know how many are allowed to be consecutive of the same type? Shana Tovah.
    – DanF
    Commented Sep 11, 2015 at 3:01
  • Logically, multiple 30 day months would force 29 day months because the new moon would be seen "early" and the minimum time is 29 days Commented Sep 11, 2015 at 11:38
  • @sabbahillel, not if no one happened to see it, get to Jerusalem in time, and be able to testify coherently. (But see judaism.stackexchange.com/q/18202 .) However, the opposite is true: successive short months would force a long one.
    – msh210
    Commented Sep 11, 2015 at 13:03
  • @msh210 Sooner or later there would be enough "extra" days, that someone would be sure to have seen the moon and be able to testify. Even if the whole rainy season is too cloudy, the dry season would cause the moon to be seen to go back to 39 day months. Consider that people knew when it was expected and would be looking for it. Commented Sep 11, 2015 at 13:30

1 Answer 1


The Rambam (Hilchos Kiddush Hachodesh ch. 18) deals with this whole issue, and he discusses this exact scenario, where you have some months where it's cloudy, others where no witnesses see it because it was small, and others where it just wasn't yet visible.

In a nutshell, as DoubleAA mentioned in a comment, you can never have a year that has more than 8 (or, one time, 9) of one type. But more to the point, the Rambam explains that the Sanhedrin aimed to make it that in any given month the moon wouldn't be visible earlier than the night after the 29th, so that alone puts some kind of limit on the number of possible consecutive full or defective months (although he doesn't give a specific number).

Tosefta, Erchin 1:4, says:

מעולם לא נראו ששה חדשים המעוברים בשנה זו אחר זו

It was never seen (or: considered appropriate to have) six consecutive full months

-- implying that sometimes they had five.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .