Talk:Date of Easter

Whether the pattern repeats
Tom, what do you think? I added the following paragraph:

One may wonder whether the pattern of dates of Easter repeats. As can be verified in a table of the dates, the dates from 1948 to 2047 are the same as for those 152 years later (2100 to 2199). This period of 152 years is one day short of two Calippic cycles, and is a whole number of weeks. Also the dates for 1800 to 1899 are the same as those for 152 years later (1952 to 2051) in all but 11 cases. The epacts differ by 1, as do the dates of the ecclesiastical 14th of Nisan, but usually the next Sunday falls on the same date. It is possible to have the same dates for the 14th of Nisan and on the same day of the week for other intervals, but since the golden number is not the same, this situation lasts for less than 19 years at a time. For example, the 100-year period 1980-2079 is almost repeated 220 years later (2200-2299), but with six cases that don't agree. An interval of 18,000 years achieves this in some cases

Footnote (which I can't seem to put here in the talk page): It is easy to verify that this occurs if the interval contains 58 lunar corrections and the golden number of the first of the two years in question is less than 13, or if there are 57 lunar corrections and the first golden number is greater than 12. In either case, these cancel out the 135 solar corrections.

and is also a whole number of 400-year Gregorian cycles, but again the golden number will be different, so the pattern does not repeat exactly.

Five minutes later Jc3s5h reverted it, saying that it's a "wall of text" that doesn't answer the question I asked at the beginning -- whether the pattern repeats, and he says it repeats in 5.7 million years. Well of course, we all know that. But does that justify reverting my edit? The question I mean is whether the pattern repeats for a while, like a century, and the answer is yes. I think it's quite interesting, and should be in the article. Eric Kvaalen (talk) 19:25, 10 September 2023 (UTC)


 * I was going to revert it and Jc3s5h beat me to it by minutes. It reads far too much like a blog post, beginning with "One may wonder whether the pattern of dates of Easter repeats" which is exceedingly unecyclopedic. It was also too far too long for the thoughts expressed. Arguably it works better as a talk page post here to open a discussion on whether/how to express your thoughts. DeCausa (talk) 19:33, 10 September 2023 (UTC)


 * So why don't you modify the wording instead of deleting the whole thing?? I wish you people would stop complaining every time someone puts something interesting into an article. Nobody is forcing you to read it, after all! Other people might want to read it. Eric Kvaalen (talk) 20:34, 10 September 2023 (UTC)
 * I didn't delete it. But if I had I would reject your point because it's not my job to correct your failings. Why should I put effort into dragging out some sort of value in your edit when I can't see any? It's not interesting and it shouldn't be in the article. DeCausa (talk) 20:37, 10 September 2023 (UTC)


 * You and Jc3s5h act like Caesar watching the gladiators, and giving a thumbs up or thumbs down. Who made you the judges? I think it's very interesting that the pattern repeats exactly or almost exactly over periods of 100 years, at certain intervals. Another case I didn't put is that 1700-1747 is exactly the same as 1852-1899. One would think that people interested in Easter would show a bit of Christian charity. Eric Kvaalen (talk) 09:01, 11 September 2023 (UTC)
 * In that case, we may as well have a triumvirate. I'd have reverted it too, per WP:TRIVIA. It is a mathematical artefact of no significance. A form of pareidolia. --𝕁𝕄𝔽 (talk) 09:22, 11 September 2023 (UTC) revised 09:27, 11 September 2023 (UTC)
 * I love a thread where pareidolia is an issue. :) DeCausa (talk) 10:00, 11 September 2023 (UTC)
 * I think the issue is interesting enough that if a good source (other than the dates themselves) can be found, and if it can be stated briefly and clearly, it may be relevant for the article -- but it is up to user:Eric Kvaalen or others who want to include it to demonstrate that. In the Gregorian calendar there is a period-4 stemming from leap years and a period-7 stemming from days of the week, combining into a period-28 (not related to churches, Easter or holidays) so that e.g. from 1901 to 2099, the calendar for any two years separated by a multiple of 28 years is the same (all Mondays fall on the same dates, etc.) - but the "missing" leap years in 1700, 1800, 1900, 2100, 2200 ect. spoil the fun. Incidentally, the true period-400 in the leap years coincide with a whole number of weeks (400*365 + 100 - 3 = 146097 days = 20871 weeks), so there is a true, perpetual period-400 in the Gregorian calendar. -- Is that relevant here? No, but if similar neat patterns could be descirbed for dates of Easter - and sourced - including it would be fine. Nø (talk) 12:37, 11 September 2023 (UTC)


 * Well, as explained in our article, the only true period for the Easter dates is 5,700,000 years. I disagree that what I wrote is just a "mathematical artefact of no significance". The reason what I wrote works is that 152 years is a day short of two Calippic cycles, which means that the phase of the moon is more or less the same, and the fact that a Calippic cycle has four days more than a whole number of weeks means that two Calippic cycles minus 1 day is a whole number of weeks. So whenever the 152 years spans one year like 1800 where the Gregorian calendar drops a leap year, and there's also a lunar correction, then you get periods that follow exactly the same pattern, for 48 or 100 years. If the 152 years does not span a year like 1800 (for instance going from 1900 to 2052), then we have exactly two Calippic cycles, so the day of the week is off by 1. That gives periods where the pattern is almost the same. All this is of interest to mathematically minded people like me. I have no idea where to find a reference for all this, but why do we need a reference for something that can be easily checked? We don't need a reference in order to say that it's an interesting question, after all. I suggest that those of you who don't find it interesting stop reading it and go find something else to do in life rather than deleting the work of other people. Eric Kvaalen (talk) 16:20, 11 September 2023 (UTC)
 * Eric Kvaalen asked "...but why do we need a reference for something that can be easily checked?" If a numerical curiosity is mentioned in several well-regarded sources, that hints that it's worth including in our article. If no reliable sources felt it was worth mentioning, it probably isn't worth mentioning. Jc3s5h (talk) 16:31, 11 September 2023 (UTC)
 * I suggest that those of you who don't find it interesting stop reading it and go find something else to do That could be addressed to our readership. This article isn't a personal blog existing for your self-indulgence. DeCausa (talk) 19:27, 11 September 2023 (UTC)

I hope it's clear from what I wrote yesterday that this is not a case of pareidolia. I have a question for Caligula, Claudius, and Nero: You say what I wrote is not interesting, and I suppose you are concerned for the disk space of Wikimedia. What if I find three μαρτῠ́ρους who will attest to finding that it is interesting? Is it still you who get to decide? Eric Kvaalen (talk) 07:09, 12 September 2023 (UTC)
 * You need to read WP:CONSENSUS (although "find" sounds like you also need to read WP:CANVASS). And don't you mean Caesar, Pompey and Crassus? DeCausa (talk) 07:17, 12 September 2023 (UTC)
 * And you have to show a reliable source that considers this factoid to have any significance, specifically in respect of calculation of the date of Easter. --𝕁𝕄𝔽 (talk) 15:04, 12 September 2023 (UTC)


 * It's Catch 22. There are three Caesars, or shall I call them watchful dragons, guarding this page to prevent any changes they don't agree with. If they revert you, you're stuck. You can't revert back or you get warnings of "edit wars". You can't ask someone else, or you get accused of canvasing. I have three friends who I think would agree with me that what I wrote was interesting, but I'm told that it doesn't matter. What the Caesars want is what happens. I'm told that I have to find a "reliable source" to say that it's interesting! Just another way of preserving the statum quo ante. Now I'll probably get a warning for arguing. Eric Kvaalen (talk) 06:54, 20 September 2023 (UTC)
 * No warning, but I am surprised that after 18 years and 10k edits on Wikipedia (I just checked) these very basic policy principles seem to be news to you. Do you really not know about WP:CONSENSUS, WP:CANVASS, WP:V and WP:DUE? DeCausa (talk) 07:00, 20 September 2023 (UTC)


 * Yes, I know the cards are stacked against me. Eric Kvaalen (talk) 10:02, 20 September 2023 (UTC)
 * Eric, no, the cards are not stacked against you, or to be more precise, they are stacked against any editor who tries to add original research or trivia. Let me repeat: all you need to do is to find "a reliable source that considers this factoid to have any significance, specifically in respect of calculation of the date of Easter". Consensus (or lack thereof) does not trump reliable sources. --𝕁𝕄𝔽 (talk) 10:18, 20 September 2023 (UTC)
 * Well, I don't actually agree with that - WP:ONUS says specifically otherwise. DeCausa (talk) 11:28, 20 September 2023 (UTC)
 * True but as it is a (very) hypothetical question, I didn't consider it necessary to qualify my response with details of the more advanced aspects of source quality and preponderance of expert opinion. We are nowhere near that stage; right now, we have no sources whatever that attach any significance to Eric's mathematical coincidence. --𝕁𝕄𝔽 (talk) 12:08, 20 September 2023 (UTC)
 * But to be unambiguously clear: if Eric succeeds in producing citations that meet my test, I will change my !vote from 'oppose' to 'support'. "When the facts change, I change my mind. What do you do, sir?"—Paul Samuelson. --𝕁𝕄𝔽 (talk) 15:50, 20 September 2023 (UTC)
 * There were three issues with this edit, only one of which was lack of sourcing. The second is whether it's worth making the point. I could be swayed on that by sourcing but it would take more than one rogue source to do that. But the third is the meandering blog-style WP:WALLOFTEXT would have to be converted into encyclopedic language which was considerably shorter (and which certainly doesn't begin with "One may wonder..."). DeCausa (talk) 20:34, 20 September 2023 (UTC)
 * If it hasn't materialised by now, it isn't going to. My point was about the general principle. --𝕁𝕄𝔽 (talk) 22:17, 20 September 2023 (UTC)

Better to provide citation or delete table?
Astrolynx added a citation needed template to the caption for an image of a table. I have a hard time reading the table and can't tell if it supports what is stated in the caption. If the table says what the caption says it does, the table serves as it's own source. But I question the merit of having an illegible table from 1907, and I think the caption mostly consists of trivia that doesn't belong in the article. I'm inclined to just delete the table. Jc3s5h (talk) 17:58, 22 November 2023 (UTC)

Every century has an April 18th Ecclesiastical Full Moon
Since April 19 is always replaced with April 18 for Epact 24, in order to keep Easter before April 26, so the ecclesiastical full moon is moved from Sunday, April 19 to Saturday, April 18, so April 26 is always replaced with April 19, and for the same reason with Epact 25 when the golden number is 12 or greater where April 18 is replaced with April 17 for the Ecclesiastical Full Moon, and the date of Easter (April 25) is substituted with April 18, for Epact 26 the ecclesiastical full moon (April 17) is never replaced with April 16, and the date of Easter (April 24) is never substituted with April 17 because the golden number would have to be higher than 22 and golden numbers cannot go above 19. So as a result every century has an April 18 Ecclesiastical Full Moon, with most of them having April 17, the centuries that don’t have an April 17 Ecclesiastical Full Moon is when Epact 25 is golden numbers 9, 10, and 11. 2603:6011:E00:4C41:0:0:0:1004 (talk) 23:37, 27 December 2023 (UTC)

If Astronomical Full Moon was used for Easter, instead of the Ecclesiastical Full Moon
If we use the Astronomical Full Moon to determine when Easter is, the Astronomical Full Moon dates would range from March 21 to April 19, as March 20 is often the date of the Equinox, the additional day (April 19) would need to be included as the lunar month on average is 29 days, 12 hours, 44 minutes, and 3 seconds long that extra 12 hours, 44 minutes and 3 seconds would need to be added onto April 19, so Easter would occasionally fall as late as April 26 under this rule, and if April 19 is on a Monday through Saturday, the date of Easter wouldn't change if the full moon were to be one day earlier (April 18).

Although the true Astronomical Equinox is March 19, March 20, and March 21 depending on leap years and time zones (but the equinox is most often on March 20), so the Astronomical Easter Full Moon would range from March 21 to April 19. 2603:6011:E00:4C41:0:0:0:100D (talk) 20:51, 30 January 2024 (UTC)

Incorrect date for Jewish Passover
On the Hebrew calendar, passover starts on 15 Nisan, not 14 Nisan. Jonmlevinson (talk) 13:14, 25 March 2024 (UTC)


 * It would help if you would provide a citation to a reliable source for this information, that makes it clear (1) what version of the Hebrew calendar the claim applies to and (2) includes the time of day. It's confusing because I understand that in some situations Hebrew days start in the evening while modern secular days start at midnight. Also, please quote the exact sentence you object to so we can find it in the article. Jc3s5h (talk) 17:30, 25 March 2024 (UTC)
 * I assume that that Jonmlevinson is referring to the second sentence of #Background: In the Hebrew lunisolar calendar, Passover begins on the 14th day of Nisan.
 * I wonder if the intro to Passover Seder might give a clue? It is conducted throughout the world on the eve of the 15th day of Nisan in the Hebrew calendar (i.e., at the start of the 15th; a Hebrew day begins at sunset). The "eve of the 15th day" is the 14th day, is it not? (That would imply that the Seder article is theologically wrong because formally it is conducted at the start of 15 Nisan, which to Gentiles is still the 14th for another six hours or so. To say "the eve of the 15th" seems rather inconsistent?) Either way, User:Jc3s5h's request #2 is critical. --𝕁𝕄𝔽 (talk) 17:47, 25 March 2024 (UTC)
 * I see we are not the first to encounter this conundrum (Why Is Passover on Nissan 15, Not Nissan 14?) so we need at least to add an explanatory note. Right now, we don't even have a citation. --𝕁𝕄𝔽 (talk) 18:38, 25 March 2024 (UTC)
 * This source is clearer. It explains that Pesach starts on 14 Nisan but Passover proper starts on the 15th. Clarification note definitely needed. --𝕁𝕄𝔽 (talk) 18:43, 25 March 2024 (UTC)
 * I'm not sure how reliable that source is? Does anyone else make a distinction between Pesach and Passover? --𝕁𝕄𝔽 (talk) 17:24, 26 March 2024 (UTC)
 * I have revised that sentence so that it reads now
 * "In the Hebrew lunisolar calendar, Passover begins at twilight on the 14th day of Nisan."


 * I assume that this resolves the question. --𝕁𝕄𝔽 (talk) 17:24, 26 March 2024 (UTC)

Time Zone?
The very first paragraph summarises the "bottom line" of the lengthy article that follows as: Easter is celebrated on the first Sunday after [...] the first full moon on or after 21 March. There is no reference to time zone. However, the "date of the first full moon" depends on the time zone; the last Pascal Full Moon, e.g., was on 25 March 2024 UTC but on March 24, 2024 Pacific Standard Time. Alternatively, if one would define Easter not via the "fixed approximation" of 21 March but instead by the actual Vernal Equinox, then the phrase "first Sunday after" would need to be augmented by a time zone specification, as is done by the reform proposed in 1997, which references Jerusalem astronomical time. I don't know what the actual current definition is, but it should be included in the article. MsohniCam (talk) 12:00, 1 April 2024 (UTC)


 * Perhaps it should be clarified, but there's no actual astronomy - or time zones - in the easter calculation; rather, there is a mathematical approximation that only deals with dates, not time of day, be it e.g. UTC or local, and not the exact time of the equinox. The text mentinos that 21 March is an approximation to the equinox, but fails to mention that the date of the full moon is also an approximation. Perhaps the first full moon on or after 21 March should be a date for the first full moon on or after the spring equinox on the Northern hemisphere, calculated using certain approximations? Nø (talk) 14:02, 1 April 2024 (UTC)
 * None of the above. It is the Paschal Moon (within a day or so of the astronomic full moon) and 21 March (usually one day after the equinox in the Gregorian calendar, 14 days after using the Julian calendar). Bede got it wrong 1500 years ago and people still repeat it. There is no satisfactory short description, afaics. 𝕁𝕄𝔽 (talk) 15:32, 1 April 2024 (UTC)
 * It is a cartain mathematical approximation to the first Sunday after the first full moon after the equinox -- but it is not a good approximation.
 * Equinox is appoximated as March 21 (let's call that approximation a1).
 * The full moons are approximated too (let's call that a2) -- and may be a day (or two?) off because of a2, or about four weeks off because of a1.
 * Finally, the Sunday -- well, there is not a third approximation here; we do know the days of the week -- but if the full moon is a day off because of a2, or just a few hours, the Sunday may be a week off, and/or it may be four weeks of because of a1.
 * However, to say whether it in a given year and location is correct or it is one, three, four or five weeks too early or too late (I suppose all those are possible - perhaps two or six weeks early or late as well), we'd need to know the exact "first Sunday after the first full moon after the equinox", and there are several ambiguities in that phrase that would have to be resolved before we could say that; the question about time zones could one of them.
 * Despite all the problems with the approximations, I suppose the computus does in fact most years, most locations, and with most sensible resolutions to the ambiguities of the phrase, give us the exact "first Sunday after the first full moon after the equinox". Nø (talk) 16:21, 1 April 2024 (UTC)