Misplaced Pages

Revision as of 21:18, 11 June 2010

This article has not yet been rated on Misplaced Pages's content assessment scale. It is of interest to the following WikiProjects:

Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details. Time High‑importance This article is within the scope of WikiProject Time, a collaborative effort to improve the coverage of Time on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.TimeWikipedia:WikiProject TimeTemplate:WikiProject TimeTime High This article has been rated as High-importance on the project's importance scale. Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details. Classical Greece and Rome Mid‑importance This article is within the scope of WikiProject Classical Greece and Rome, a group of contributors interested in Misplaced Pages's articles on classics. If you would like to join the WikiProject or learn how to contribute, please see our project page. If you need assistance from a classicist, please see our talk page.Classical Greece and RomeWikipedia:WikiProject Classical Greece and RomeTemplate:WikiProject Classical Greece and RomeClassical Greece and Rome Mid This article has been rated as Mid-importance on the project's importance scale.

This article is written in British English, which has its own spelling conventions (colour, travelled, centre, defence, artefact, analyse) and some terms that are used in it may be different or absent from other varieties of English. According to the relevant style guide, this should not be changed without broad consensus.

Archives

Archive 1 Archive 2

Triennial cycles debunked

There is an inscription which says that in Asia in an unspecified year the last day of the old lunar calendar was 14 Peritios = a.d. X Kal. Feb. (23 January). The following day, from which the calendar would remain aligned to the Roman calendar, was 1 Dystros, a.d. IX Kal. Feb. (24 January). Thereafter, the Asian month would begin on a.d. IX Kal. of the Roman month. The old calendar being lunar, the problem comes down to seeing when 23 January equates to a full moon.

My table shows two likely candidates - 8BC and 5BC.

There was an eclipse on March 23, 5BC (Julian date). There was thus also a full moon on January 24, 5BC (Julian date). I have not investigated the arrangement of intercalary years in the ancient Greek calendar.

The inscription mentions an intercalation. It is unlikely that the irregular Julian calendar was being introduced (in 8BC) because it was no longer intercalated at that time, and the purpose of the reform (as in Egypt and Rome) was to introduce the correct Julian calendar. The likely date is therefore 5BC, with the regular Julian intercalation coming a few weeks after adoption.

There is clear indication that, having moved to correct the calendar in Rome in 9BC, Augustus turned his attention to Asia. He would have been well aware of the situation in Egypt, and the fact that he felt no need to take any action there indicates that no action was needed. Professor Jones says:

The Egyptians must at some point have become aware that the Roman dates that they assigned to particular days differed by one or two days from the dates according to the pontifices, but we should not assume that they would have immediately changed the reckoning to conform with the official version of the calendar. The calendar equation Roman July 19 = Egyptian Epeiph 27 discussed by Hagedorn indicates that conformity was imposed by 2BC.

This date equation puts the wandering year out of the picture, but to conclude that that indicates that the Egyptians had been forced to abandon the fixed relationship with the Alexandrian calendar seems to me misguided. From 9BC the pressure was all the other way.

I used the Easter holiday to translate a paper made use of by Dr Bennett in his argument.

Dieter Hagedorn

On the Egyptian calendar under Augustus

from: Zeitschrift fuer Papyrologie und Epigraphik 100 (1994) 211 - 222.

Copyright: Dr. Rudolf Habelt GmbH, Bonn

ON THE EGYPTIAN CALENDAR UNDER AUGUSTUS

Theodore Cressy Skeat is the author of that basic description of the way the Ptolemaic calendars work , of which practical conversion tables we all make use, dealing with dates, which through the naming of the regnal years of one of the Ptolemies and of the day in an Egyptian month are fixed, to convert them to their Julian equivalent.

The problem there lies in this, that one used in Ptolemaic time the wandering Egyptian year with a constant 365 days (consisting of 12 months of 30 days and 5 additional days placed at the end of the year, epagomenai), while in the Julian calendar it is well known that every fourth year is a leap year with 366 days. Through that the difference between the Egyptian and Julian year increases one day every four years from 29 February.

Under Augustus the Egyptian calendar was reformed. The final result of this reform, which we can trace over centuries, is the "Alexandrian" calendar, in which regularly every fourth year is a leap year with 366 days. The Alexandrian and the Julian year agree thereafter in the long term, with an average 365 1/4 - day exact agreement, so that there is exact equivalence;always the leap day in the Alexandrian year was added not on the same day as in the Julian calendar, namely on 29 February , but at the end of the year as the sixth additional day. While the New Year's Day, 1 Thoth, normally falls on 29 August, it corresponds after an Alexandrian leap year with 30 August. This is regularly the case in the year which precedes the Julian leap year; up to 29 February of the following year there is always a difference compared to the normal relationship of one day.

With the completion of the above - mentioned standard work Skeat has now recently brought out a small volume with similar tables which for that will serve to illustrate the changeover from the Ptolemaic - Egyptian calendar to the Alexandrian calendar which - so it is said - was accomplished under Augustus.

In consideration of the high regard which Skeat as well as Fachmann enjoys on questions of the chronology, it is also to be expected that his interpretation becomes quickly and perhaps also uncritically generally accepted. To me there are after the reading of the little book however considerable doubts remaining on the rectitude of his approach. To me it seems that this interpretation is based on unproved (and by our present information also unprovable) theories. These are not one's own theories, which are based on other thoughts, so one comes to different conclusions. The purpose of this paper is to describe these conclusions and equally to offer them as an alternative to Skeat's solution to the question.

First the circumstances are set out in which the question how the reform of the Egyptian calendar under Augustus was accomplished became more complicated.When we speak of the "Julian" calendar, so by that we mean the calendar that theoretically was brought into effect through C. Julius Caesar's calendar reform in the year 45 in Rome. In reality however Caesar's ideas, which he had ordered, that quarto quoque anno should be intercalated, were falsified, because this expression, on account of the inclusive counting system customary in Rome, was misunderstood as "in every third year". One intercalated from there falsely in the years 42, 39, 36, 33, 30, 27, 24, 21, 18, 15, 12, and 9 BC. After that at first Augustus brought through his decree, future quinto quoque anno a leap year, however, to ensure again to save Caesar's original intentions for the system's validity through omission of the leap day in 5BC, 1BC and 4AD the already excess intercalated days. From 1 March AD4 onwards the actual misaligned and the ideal Julian calendar were in agreement, on 29 February AD8 it was for the first time intercalated in agreement with the ideal calendar. The modern scientific research uses the ideal Julian calendar exclusively, which it also applied during the time before 1 March AD4, reconciled with the fact that up to this time it was nowhere in use.

Skeat makes now in the succession to W. F. Snyder the following theories:

1. The reform of the Egyptian calendar brought in by Augustus had for its aim the alignment of this to the one found to be in use in Rome, i.e. perverted, to correct the Julian calendar.The second theory hangs tightly together with it: the reform must have taken place in a year in which 1 Thoth fell on 29 August (according to the Julian calendar then found to be in use). Skeat fixes that on the fact, that this equation was the usual one after the firm establishment of the Alexandrian calendar.

The postulate put forward under 2 is fulfilled in 30BC. The theory formulated under 1 forces the assumption, that in Egypt in the years 28, 25, 22, 19, 16, 13 and 10BC one had inserted a 6 epagomene, thereafter the intercalation was abandoned, with the correct intercalation to begin in a four - yearly cycle for the first time in the year AD 7. This is the essence of Snyder's solution to the problem, which Skeat follows. But are these theories acceptable? If the aim of the reform really was an alignment with the system practised in Rome, why has one then not plain and simple introduced the Roman calendar in Egypt? At the reorganisation of Egypt into a Roman province it had been lightly managed. On the other hand, i.e. at the reorganisation of the wandering new year and the 30 - day Egyptian months, a congruence with the months was never achieved, but a divergence in the numbers of the day - dates was inherent in the system. Already the diverging day - dates do not first of all stand after the reform in unchanging correlation, rather the correspondences were displaced, as above explained, after every Egyptian leap year for a half year by 1 day, more weight being given when one actually, as Skeat accepts, intercalated every three years.

One takes it as given, the reform of the Egyptian calendar by Augustus had another purpose, then the consequences are also quite different. Why should it not be the case, that Augustus in Egypt has wanted to achieve precisely this in the action, what Caesar also in the sense had had in Rome but had not been understood, and what already over 200 years before Ptolemy III. Euergetes I had planned,but likewise not had been able to carry through, namely an alignment of the calendars with the astronomical year, which it is well known comprises (roughly) 365 1/4 days. That the interval of three years fixed in the intercalation practised in Rome was absolute nonsense, the insights already soon allowed it to be grasped that this was right. That also Augustus listened to them, proves the fact, that he finally had achieved also in Rome the right understanding in the action.

If from now on the reform in that was effected, that with no consideration for the practice in Rome immediately the right manner of intercalation in a four - yearly interval was brought in, then it is a simple calculation to find out, when for the first time a 6 epagomene must have been intercalated, fixed, that through that the Alexandrian calendar so familiar to us at a later time was probably established. The solution already found for a long time is: at the end of Augustus' 8th regnal year, i.e. on 29th August 22BC (in the "ideal" Julian calendar); cf also the following tables.

For an irrefutable proof for the rightness of the traditional basis to take, which I share, I see myself not in the situation; because an absolutely certain verdict between the two solutions put forward might well be only then hit upon, if in the source a date equation on a 6 epagomene were found, and actually in one year, which according to only one of the two hypotheses was a leap year. I have not been able to find such evidence.

But while Snyder and Skeat, on every attempt to do without, to substantiate their theses through documentary evidence, I believe a contemporary source to be able to furnish a compelling argument for the rightness of the traditional basis.

It comes in the form of the well - known Latin letter P. Vindob. L1c = CPL 247.The text, which with certainty dates from the time of Augustus in an unknown year carries in 2.16 the date equation

XIIII K(alendas) August(as) Epeiph 27.

Also 19 July is equated with 27 Epeiph, in which the same thing is understood, that the Latin statement only can cover the calendar actually in use, not our ideal Julian calendar. One now takes Skeat's tables as the basis, then it quickly becomes apparent that the correspondence of a 27 Epeiph with 19 July of the "Current Julian Calendar" is never possible. In "Table B" on pp 8ff namely 1 Mesore falls on 25 July in every year of the "Current Julian Calendar", 27 Epeiph consequently on 21 July. This is unavoidably so, because according to Snyder/Skeat the Egyptian calendar under Augustus already employed all vagaries of the Julian calendar actually in use (cf footnote 6 above), so that in the relation of the Egyptian calendar to the Julian calendar actually in use already the equivalences must be valid, which are known later in the relation of the Alexandrian calendar to the ideal (and up to this time no more different from the actually used) Julian calendar. 27 Epeiph then likewise always corresponds with 21 July.

That however on the basis of the traditional opinion a correspondence of 19 July in the calendar actually in use with 27 Epeiph is definitely possible J. Kramer has already pointed out.It occurs namely in regnal years 6 - 11, 13 - 14, 17 and 25 - 28 ( = 24 - 19, 17 - 16, 13 and 5 - 2 BC.) In those years 27 Epeiph falls thereafter on 21 July of the ideal Julian calendar, but the Julian calendar actually in use lagged behind the ideal Julian calendar an exact two days here: to that just compare Skeat's "Table A". 27 Epeiph fell in them consequently on 19 July.

One will also not dismiss the date equation in P. Vindob. L1c as a copying error, then can Snyder and Skeat's portrayal not be true, while on this basis a plausible explanation can be found.

The following tables illustrate the equivalences between the Egyptian and the ideal Julian calendar, which on this basis are produced for the first 9 regnal years of Augustus.For the next following years the standard tables provided for the Alexandrian year mentioned in footnote 2 above can be used. By that is to observe, that Augustus' regnal years 12, 16, 20, 24, 28, 32 etc. in the Egyptian calendar were leap years, so that to the start of each next following year (i.e. Julian 18/17, 14/13, 10/9, 6/5, 2/1, 3/4, 7/8 etc. the earlier alternative equivalences are allowed for.

Notes

1 Th. C. Skeat, The Reigns of the Ptolemies (Muenchener Beitraege zur Papyrusforschung und antiken Rechtsgeschichte, No. 39), Muenchen 1 1954, 2 1969.

2 These correlations illustrate e.g. the tables by P. W. Pestman, Chronologie egyptienne d'apres les textes demotiques (Papyrologica Lugduno - Batavia, vol. 15), Leiden 1967, after p. 8 and R. S. Bagnall - K. A. Worp, The Chronological Systems of Byzantine Egypt (Studia Amstelodamensia ad epigraphicam, ius antiquum et papyrologicam pertinentia, vol. 7), Zutphen 1978, pp 96 - 102.

3 Th. C. Skeat, The Reign of Augustus in Egypt. Conversion Tables for the Egyptian and Julian Calendars, 30 B.C. - 14 A.D. (Muenchener Beitraege zur Papyrusforschung und antiken Rechtsgeschichte, No. 84), Muenchen 1993.

4 I follow here entirely Skeat's implementations (The Reign of Augustus, pp 2 - 3), the explanation at that time of P. V. Neugebauer, Der julianische Kalender und seine Entstehung, in: Astronomische Nachtrichten 257, 1935, No. 6149, Sp. 65 - 74 reviews. Cf also A. E. Samuel, Greek and Roman Chronology. Calendars and Years in Classical Antiquity (Handbuch der Altertumswissenschaft I 7,) Muenchen 1972, pp 155 - 158.

5 W. F. Snyder, When Was the Alexandrian Calendar Established? AJPh 64, 1943, 385 - 398. The verbose in superior mixture rather than argumentatively written paper attempts through the use of illustrative tables to give the impression of its exact scientific nature, but it contains not one single compelling argument to the immediate support for the aforementioned theories. To my evaluation Snyder has only created confusion after, when the old research already has recognised the right thing for a long time - compare only Wilcken, Grundzuege p. 55 et seq.

6 Cf p. 4: "Thereafter the reformed Egyptian Calendar must have shared in all the vagaries of the current Julian Calendar, since the intention of the reform was to stablish a constant relationship between the two calendars".

7 Cf pp 1 - 2: "Since, as a result of this reform, 1 Thoth always fell on 29 August ..., it is obvious that the reform must have taken place in a year when it did in fact fall on 29 August in the Roman Calendar currently in use". Also p. 3: "The problem, already stated, is to find a period when Thoth 1 in the unreformed Egyptian calendar fell on 29 August in the Roman Calendar currently in use."

8 He had likewise already wanted to intercalate every four years a 6 epagomene. The intention is in the Canopus decree, documented and well - founded; cf OGIS 56 = A. Bernand, Le Delta egyptien d'apres les textes grecs, 1 - les confins libyques, vol. III, Cairo 1970, pp 989 - 1036, lines 32 - 37.

9 I have searched with help the CD of the Duke Data Bank as well as also in the extremely helpful Lists of C. Balconi, Documenti greci e latini d'Egitto di eta augustea, Aegyptus 56, 1976, 208 - 286. It gives in the uncertain period in general no date equation for a 6 epagomene.

10 Cf on that J. Kramer, Die Verwendung des Apex und P. Vindob. L1c, ZPE 88, 1991, 141 - 150. One finds there pp 143 - 144, footnote 17 the necessary references to older literature. Kramer's paper could not be more consulted than the new edition by P. Cugusi, Corpus Epistularum Latinarum Papyris Tabulis Ostracis servatarum (Papyrologica Florentina, Vol. 23), Florence 1992, No. 8.

11 cf pp 144 - 145.

12 Kramer's paper, loc. cit., which purely arithmetically puts also the year 27 and 26 into consideration, needs correction. In the years 27 and 26 in my opinion 22 July of the ideal Julian calendar corresponds with 27 Epeiph; there the date in the calendar actually used is likewise retarded by two days, one must have in this year expected XIII K(alendas) August(as). There further the renaming according to tradition of the month Sextilis as Augustus was effected first in the year 8 BC (cf Samuel, loc. cit., p. 155, footnote 6), only the years 5 - 2BC for the writing of the papyrus fall into consideration.

13 In far scarcer detail already Martina Richter has achieved the same thing in ZPE 86, 1991, 252.

14 Those tables respectively parts of the tables by Skeat, which illustrate the equivalence between the initially practised and the ideal Julian calendar, are through this not relevant.

(Translator's Note)

There then follows a set of tables. The sixth epagomenal day makes a solitary appearance between 31 August 30 BC and 28 August 21BC on 29 August 22BC. The Julian leap day (taken as 29 February) appears in 29, 25 and 21 BC. The last day of the wandering year is taken as epagomene 5, 28 August 22BC.

Conclusions

Hagedorn's paper is not, as Dr Bennett claims, support for his proposed text. The letter is written in Latin, obviously by a Roman. The situation is similar to that of a Russian equating his old style Christmas (25 December) with the corresponding Meletian date (7 January). 212.85.12.187 (talk) 12:19, 20 April 2010 (UTC)

Eastern European calendar: naming proposal

On this glorious Easter Tuesday, united around the world, here is an update on the progress of the ballot.

Option 1 - Meletian calendar - 1 vote (recommended option)

Option 2 - New calendar (Eastern churches) - no votes (this option is not recommended)

Option 3 - No change - 2 votes (this option is not recommended)

Option 4 - "Revised" Julian calendar - no votes (this option is not recommended)

To vote by proxy, write QUICKVOTE and sign with four tildes. If you want your proxy to vote in a particular way, add the option number in brackets. Thus QUICKVOTE (1) means your vote will be cast in favour of option 1.

The tilde is the wavy line ~ sometimes placed above n (in Spanish) or a or o in Portuguese where, following the medieval Latin copyists, it marks the omission of a following letter n.

This is not the place to vote. Click on this link Talk:Revised Julian calendar#Proposal to change article name, read the manifestos and then add your vote underneath the others.

Uma Paschoa muito feliz a todos. O povo unido jamais sera vencido. 212.85.12.219 (talk) 18:41, 6 April 2010 (UTC)

Christmann theory of triennial intercalation

Chris Bennett's summary, introduced into "Leap year error" at the weekend, contains a number of errors. I was surprised to see it when I switched on the computer yesterday morning to contribute to this section as the text was remarkably similar to a text I had also prepared over the weekend and was intending to save. Most of the additional material relates to musings by medieval chronologers who didn't have much grasp of their subject. It is important that the relevant passage from Macrobius should be included as well so that the errors can be appreciated. Theories which involve thirteen triennial intercalations are non - starters, as are those which postulate the omission of four regular leap days.

This is made clear in my text which is reproduced below. Any registered user feeling up to wiping the existing text and transcluding this in its place? I have not mentioned either Soltau or Christmann as these are really oddball theories. What Christmann actually proposed was 43, 40, 37, 34, 31, 28, 25, 22, 19, 16, 13, 10BC, AD7, which is nonsense because one thing we know for certain is that AD7 was not a leap year, and Augustus would not have omitted four regular leap days (9, 5, 1BC, AD4).

Although the new calendar was much simpler than the pre - Julian calendar, the pontifices apparently misunderstood the algorithm for leap years. They added a leap day every three years, instead of every four years. According to Macrobius, the error was the result of counting inclusively, so that the four - year cycle was considered as including both the first and fourth years. This resulted in too many leap days. Macrobius says:

This error continued for thirty - six years by which time twelve intercalary days had been inserted instead of the number due, namely nine.

Augustus remedied this discrepancy after 36 years by restoring the correct frequency. He also skipped several leap days in order to realign the year. Once this reform was complete - after AD8 at the latest - the Roman calendar was the same as the Julian proleptic calendar..

The historic sequence of leap years in this period is not given explicitly by any ancient source, although the existence of the triennial leap year cycle is confirmed by Macrobius' account. The chronologer Joseph Scaliger established in 1583 that the Augustan reform was instituted in 8BC, and inferred that the sequence of leap years was 42, 39, 36, 33, 30, 27, 24, 21, 18, 15, 12, 9BC, AD8, 12 etc. This proposal is still the most widely accepted solution.

The sixteenth century continental chronologer Buenting suggested that there was an additional bissextile day in the first year of the Julian reform, i.e. that 45BC was also a leap year. This would give 45, 42, 39, 36, 33, 30, 27, 24, 21, 18, 15, 12BC, AD4, 8, 12 etc.

Of the two solutions, Buenting's would seem the more plausible. If there were no intercalation after the last year of confusion (46BC) the triennial cycle would suggest the first intercalation took place in 43BC, rather than 42BC. Macrobius continues:

But, when this error was at length recognised, it too was corrected, by an order of Augustus that twelve years should be allowed to pass without an intercalary day, since a sequence of twelve such years would account for the three days which, in the course of thirty - six years, had been introduced by the premature action of the priests.

The inference is that Augustus omitted three regular intercalations.

Other solutions have been proposed from time to time. In unpublished papers Thomas Harriot (1560 - 1621) considered 43, 40, 37, 34, 31, 28, 25, 22, 19, 16, 13, 10BC, AD4, 8, 12 etc. Kepler proposed in 1614, on the same material used by Scaliger, that the correct sequence of leap years was 43, 40, 37, 34, 31, 28, 25, 22, 19, 16, 13, 10BC, AD8, 12 etc. This has the disadvantage that it postulates the omission of four regular intercalations.

In 1883, the German chronologist Matzat proposed 44, 41, 38, 35, 32, 29, 26, 23, 20, 17, 14, 11BC, AD4, 8, 12 etc., based on a passage in Dio Cassius that mentions a leap day in 41BC that was said to be contrary to (Caesar's) rule. In the solutions of Buenting, Harriot and Matzat the Roman calendar was not finally aligned to the Julian calendar of later times until 25 February (a.d. VI Kal. Mart.) 1BC. On Kepler's and Scaliger's solutions, the two calendars were aligned on 25 February AD4.. Clavius, the mastermind of the Gregorian calendar, sided with Kepler and Scaliger.

In 1999, an Egyptian papyrus was published that gives an ephemeris table for 24BC with both Roman and Egyptian dates. The Roman dates are not aligned with any of these solutions - they are aligned with the Julian calendar as it would have been if it had been operated correctly. One suggested resolution of this problem is that the triennial cycle never found favour in Egypt (see Talk:Julian calendar#Triennial cycles debunked). Another is a sixth triennial sequence - 44, 41, 38, 35, 32, 29, 26, 23, 20, 17, 14, 11, 8BC, AD4, 8, 12 etc.. This is not supported by academics generally for the following reasons:

(1) It supposes that the ideal and actually used Julian calendars coincided in 24BC, although all sources are agreed that they coincided at inception (1 January, 45BC) and the difference increased until around 9BC, (at which point it had reached three days), when intercalation was suspended until either AD4 (assuming 45 not to have been a leap year) or AD8 (if it had been).

(2) It supposes thirteen triennial intercalations (Macrobius says there were twelve).

(3) It supposes the omission of two regular intercalations. Macrobius says quite specifically that three extra leap days had been included by mistake, and this was compensated for by twelve years without intercalation. Every period of twelve years includes three leap days, so this is the number of leap years which must have been omitted. Macrobius confirms this when he says that twelve leap days were inserted instead of the correct nine, and that the suspension of intercalation for twelve years was to remove the extra three days.

(4) It sets the epoch at the last day of the last year of confusion (46BC), which is the day before 1 January 45BC. Macrobius says unequivocally that the extra days added were compensated for by the omission of the same number of leap days, which means that both the ideal and the actually used Julian calendars must have begun on the same day.

(5) It argues from a supposed triennial cycle introduced in Asia Minor in 8BC. But since in 8BC the triennial cycle in Rome was abandoned as inaccurate, why would Augustus go to all the upheaval of introducing it afresh in Asia Minor? At that time Asia Minor was using an old Greek lunar calendar, so what we would be looking for is a conversion directly from the lunar calendar to the correct Julian calendar. This is exactly what we find (see Talk:Julian calendar#Triennial cycles debunked). 212.85.12.187 (talk) 11:39, 25 May 2010 (UTC)

The Oxford Companion to the Year

I was looking at this book over the bank holiday weekend (this, because of creeping secularisation, is the former Whit Monday now called the "Spring Bank Holiday"). We are the only country which celebrates bank closures instead of real events - the same mindset which prohibits naming streets after significant dates, grudgingly gave us New Year's Day off only in 1974 and led to Birmingham City Council renaming Christmas "Winterval".

I've not been aware of errors in this work but it seems to be full of them. It's edited by Blackburn and Holford - Strevens, published by Oxford University Press in 1999.

On p 670 it says

Every so often a board of priests known as pontifices (who were active politicians, and often behaved accordingly) would curtail February at the 23rd or 24th and insert an extra month (mensis intercalaris or intercalarius, also spelt with a k, 'called between') of 27 days to give a year of 377 or 378 days; statements in late sources that a month of 22 or 23 days was inserted within February are contradicted by better evidence.

This book usually gives the evidence in great detail - the lack of justification here suggests the editors are not very sure of their ground. They may be simply saying that because of the peculiarities of the Roman dating system "inserting a month within February" is the same as curtailing February on the day before the extra month begins and then tacking the last five days of February on to the end of the extra month. This is the explanation of Varro and Celsus. Either way, you come up with an intercalary month of 27 or 28 days. Celsus mentions the figure of 28 days - the "better evidence" is no more than a claim that Celsus wrote "xxvii" and a scribe mistakenly penned an extra "i". The last five days of February are counted as days before the beginning of March, as are all the days after the ides of the intercalary month which precede them, so the effect is that the intercalary month gains an extra five days.

No author says that February was ever curtailed at the 24th: rather there was sometimes an intercalary day inserted after the Terminalia (23 February) to prevent the nones and the beginning of the year falling on a market day. In Rome, there was a superstition that the first day of the year must not fall on a market day, because that would be unlucky: the pontifices, who regulated the calendar, therefore took steps to prevent it. It is easy to see how the superstition arose. Here is A Smaller Dictionary of Greek and Roman Antiquities, William Smith, (John Murray, London, 1868), p 114:

The time at which the old consuls laid down their office and the consules designati entered upon theirs, differed at different times. The first consuls are said to have entered upon their office in October, then we find mention of the 1st of August, of the ides of December, the 1st of July, and very frequently of the ides of March, until, in B.C. 153, it became an established rule for the consuls to enter upon their duties on the 1st of January; and this custom remained down to the end of the republic. On that day the senators, equites, and citizens of all classes conducted in a procession (deductio or processus consularis) the new magistrates from their residence to the capitol, where, if the auspices were favourable, the consuls offered up sacrifices, and were inaugurated. From there the procession went to the curia, where the senate assembled, and where the consuls returned thanks for their election. There they might also speak on any subject that was of importance to the republic, such as peace and war, the distribution of provinces, the general condition of the state, the feria Latinae, and the like.

These activities are incompatible with the holding of a market. The calendar was arranged so that the nones, the ides and the last day of a particular month always fell on the same day of the market week. February was a special case. There is one recorded instance (170BC) of the start of the mensis intercalaris being pushed away from the day following the Terminalia for this reason, but some people have reconstructed the entire 400 - year history of the Republican calendar on the theory that this happened every time the year was given 378 days. This is, of course, total nonsense, because once you introduce this fixed relationship you lose the ability to make the ad hoc adjustments required. Significantly, 170BC was one of the years which began on the ides of March (the Romans named their years according to the consuls who were in office).

This manipulation, recorded by Macrobius (Saturnalia, 1.13), is commonplace. The events described are similar to those of the Lord Mayor's Show, when the newly elected Lord Mayor of London (there is a contest each year) travels in procession from his residence (the Mansion House) to the Royal Courts of Justice. This always happens on a Saturday, which in the City of London is indistinguishable from a "bank" (public) holiday. There are more banks quartered in this one square mile than anywhere else on earth. When, however, an event is tied to both a particular date and a particular day of the week something has to give, and that something is the calendar. This is particularly marked in the Jewish calendar where, as a general rule, each date may only fall on one of four weekdays. A similar stratagem was employed in the Supputatio Romana, the Roman Easter table, where the dates of the full moons were carefully arranged so that Easter would not fall on impossible dates. This is still done, though less blatantly, in the current Easter table.

I noticed this gem in the archives:

The last day of the 23 - day truncated February was the Terminalia, so that the following days, in a March - based year, were considered as being after the end of the year. However, the following day was the festival of the Regifugium. In the Fasti Antiates, the Regifugium is marked twice, once in February and once in Intercalaris. Clearly, in an ordinary year the Regifugium was celebrated on the day after the Terminalia, and in a 377 -day year it was celebrated on 23 Intercalaris. In both cases the date was the same: ad VI Kal. Mart. What is unclear is when the Regifugium was celebrated in a 378 - day year ... Chris Bennett 15:30 21 June 2006 (UTC).

One thing I've noticed about Chris is he always looks for a complicated answer when the simple answer is staring him in the face. The Romans always celebrated the Regifugium on a.d. VI Kal. Mart. In a normal year, that was on the day after the Terminalia. In an intercalary year the day after the Terminalia was Kal. Interkal. or Prid. Interkal. if it was desired to add the dies intercalaris to shift the nones/ides of March from a market day. The Regifugium was postponed by 23 days (give or take a day) but was still observed on a.d. VI Kal. Mart.

The book may also be adverting to the "Fasti Antiates Maior", a mural calendar which depicts the mensis intercalaris with 27 days. But this is the default representation, as is the prayer book calendar of Edward VI, which depicts February with 28 days and no mention that it was ever intercalated. The allocation of the nundinal letters (Intercalaris 27 has A and March 1 has B) is also the default representation, just as the prayer book has E for February 23 and F for February 24 with no mention that an extra day would sometimes slip in between them.

On p 671 the book says

Since the tropical year was some 6 hours in excess of 365 days, Caesar ordained that 24 February, the sixth day before the Kalends of March, should in leap year be counted twice (see * Leap Year). This intercalation was to take place quarto quoque anno, by which he meant what we mean by 'every fourth year', however, such expressions usually being inclusive in Latin, after his murder in 44BC the pontifices understood him to mean every third year. An inscription of 9BC, reforming the local calendar of Asia Minor on Roman principles, explicitly prescribes a three year cycle, 'beginning with this year': but the emperor Augustus, who upon the death of the pontifex maximus Lepidus (Shakespeare's 'slight unmeritable man') in 12BC had succeeded to his office, was about to correct the error by suppressing the leap day due in 5BC, 1BC, and AD4. Intercalation was resumed in AD8, and took place every four years thereafter.

This passage is internally contradictory (see "Triennial cycles debunked" above). Our knowledge of the triennial cycle comes from Macrobius, not the inscription. Why do the editors not quote the actual text of the inscription, as they did with Caesar's edict? It is apparent that because of inclusive counting the interval specified in the decree would have to have been "two years" for this theory to have any chance of being correct. According to B A Buxton and R Hannah, in C Deroux (ed.) Studies in Latin Literature and Roman History XII 290, rather than stating that there is an intercalary Xandikos of 32 days in the year of the decree, the intended meaning is that Xandikos shall be 32 days in each intercalary year. Further, rather than specifying that intercalation shall take place every third year, the intended meaning is that intercalation shall commence on the day after 14 Peritios in the third year following promulgation of the decree (counting inclusively). In their view, Fabius was proconsul in 7/6, the decree was passsed in 7, and the first intercalation was intended to occur on the Julian cycle starting in 5BC. I am indebted to Chris Bennett for this analysis, which makes a lot more sense than supposing a new triennial cycle to have been instituted at the exact time when in Rome the self - same cycle had been abandoned.

I would add only that it is impossible for there to be an intercalary Xanthikos under this scheme for two reasons:

(1) a 32 - day month indicates a move to the solar calendar

(2) the decree was for the abolition of the lunar calendar, so that no more would the thirteenth (intercalary) month be required. Augustus' birthday was made the first day of the month, and all the months were made to begin on this same day of the Roman month.

On pp 678/9 the book says

(The suggestion that in counting back from the Kalends 'earlier' means 'later' and vice versa is disproved by the verses cited below on St Matthias' day). In 364 Valentinian I, summoned to become emperor at the death of Jovian on 17 February, refused to make an appearance on the bissextus, owing to its ill luck, but accepted acclamation on the next day, which chroniclers record as V Kal. Mart. in leap year, not the 25th (as the Greek church historian Socrates mechanically converts it) but the 26th; Ammianus Marcellinus, an eyewitness, states that the empire remained without a helmsman for ten days, namely the 17th to the 26th inclusively.

However, two non - legal scholars, Censorinus in AD238 and Macrobius some 200 years later, tell us the opposite: that the 24th was the leap day, as the logic of the backward count might suggest (VII, bis VI, VI): not by coincidence, both assert that the intercalary month of the pre - Julian calendar was added after the 23rd, though in fact it was sometimes added after the 24th.

Once again, there is internal contradiction. On p 672 the book says

The day after the Nones was always a.d. VIII Idus whether it was the 6th or the 8th of the month; that after the Ides was, in the Republic, a.d. XVII Kalendas , the Kalends being those of the next month, except for 14 February, which was a.d. XVI K. Mart. (the Republican abbreviation) or a.d. X Terminalia, a form necessary until one knew whether intercalation would be ordered, and more specific than a.d. XI or XII K. Interk. when it was. After the reform 14 February was still a.d. XVI Kalendas even in leap year.

To cut to the chase, has anybody ever found a document bearing the date a.d. XII Kal. Interkal? The fixed length mensis intercalaris theory is demonstrable nonsense because until a decision was made as to whether or not to intercalate there would have been no way to name the day following the Terminalia. The book says the naming before the calends of the following month was unusable, and the system actually used, naming before the Terminalia, does not permit the following day to be named. It is only that one nonsense day which is affected, because thereafter the real last five of February (or the mensis intercalaris) have their usual designation a.d. VI Kal. Mart. to Prid. Kal. Mart.

Also, the editors do not understand that the superstitious Valentinian would have avoided the ill - fated "twice sixth" day altogether and only assumed office on a.d. V Kal. Mart., if the Latin chroniclers have the story right. If the Greek version is right, Valentinian sat out the bissextum, a.d. bis VI Kal. Mart., and moved only on the following day, a.d. VI Kal. Mart. Either way. the evidence points to the bissextum falling in the usual place for adjustment, immediately after the Terminalia. Both Censorinus and Macrobius (who presumably had access to the actual text of Caesar's reform) say that it provided for the intercalary day to come between a.d. VII Kal. Mart. and a.d. VI Kal. Mart.

The passage relating to St Matthias on p 679 actually says the opposite of what is claimed for it.

Macrobius' account of the Roman calendar was extracted under the name of Disputatio Cori et Praetextati to become a stock text with the computists, making the order bis VI Kal. VI Kal. predominate, though the other was not unknown. This uncertainty persisted in the Western Church with regard to the feast of St Matthias on VI Kal. Mart. (see *24 Feb.), which in leap year was generally celebrated on the 25th; the rule was expressed in the verses:

Bissextum sextae Martis tenuere Kalendae,

Posteriore die celebrantur festa Mathiae.

To render doggerel by doggerel:

Mars his sixth Kalends have the Leap - day gript;

On the next Day, Matthias Feast is kept.

In other words, the day called in common years VI Kal. Mart. (=24 Feb.) occupies the bissextus position in leap year, St Matthias' feast being deferred to the day after (the 25th).

The book's stable companion, The Oxford Classical Dictionary, (3rd edition, ed. Simon Hornblower and Antony Spawforth), Oxford University Press (1996) agrees at p 274:

The Egyptian solar calendar was adapted to Roman use, by inserting enough days in the shorter months to bring the total up to 365 and arranging for the insertion of a day, not a month, between 23 and 24 February, in leap year (thus 23 February occurred twice; the non - existent date '29 February' is a modern absurdity).

92.27.84.129 (talk) 10:42, 2 June 2010 (UTC)

• All unassessed articles
• C-Class Time articles
• High-importance Time articles
• C-Class Classical Greece and Rome articles
• Mid-importance Classical Greece and Rome articles
• All WikiProject Classical Greece and Rome pages
• Misplaced Pages articles that use British English