The Latium parapegma and the nundinal days

In the Chronography of 354 A.D., which may be found online here, part 6 consists of a calendar.  The days of the month are listed. I give an extract from January here:

I don’t know what the first column is.  The second column, in Roman numerals, are the days of the week, 1-7.  Each week of 7 days corresponds to one of the 4 phases of the moon, which results in the “lunar month” of 28 days.

But what about the third column?  This shows a “week” of 8 days, numbered 1-8 in Roman numerals?

These are the nundinae, the “nundinal days”.

How do we know this?  Well, we might look at the “Latium parapegma”.  This is a slab of rock with inscriptions on it, with holes against the words.  Here’s a photograph, followed by a proposed restoration.[1]

Note the “nundinae” column on the right hand side.  Note the peg hole, and the list of 8 names.  Most are the names of towns in Latium, except for “in vico”, i.e. in the village, i.e. “here”. 

This is, I am told, all about market days.  Once every eight days, there would be a market and the farmers could buy and sell there.  So there was a cycle of eight days.  It is hypothesised that each town held a market on a different day, and therefore the names above indicate which town was holding a market on that day. 

The word “nundina” is supposed by modern scholars to be derived from novem and dies, i.e. nine and day.  We count 8 days from market day until the next market day; but the Romans counted both market days in that span, making a total of 9, or so I am told.  It would be most interesting to see the data on which all this is based.

A word about peg-calendars (parapegma) is perhaps in order.  I learn from Lehoux that the peg calendar is a farmers’ tool.  It was necessary because the secular calendars did not keep in sync with the seasons. 

We all know how the Julian calendar came into being; because the Roman calendar had drifted so far away from the real months that winter was in summer and so on.  Likewise politicians would muck around with the calendar for political advantage, adding days and so on.

But this caused a real problem for the farmers, who needed to put their crops in the ground and gather the harvest at set times in the year, when the weather was right.

Their solution was to follow the fixed stars, which rise and set regardless of politics.  And they could then keep track of days using a bunch of lists, and move a peg along the list, each day.  If they had a slab with several lists on it, as in the Latium parapegma, this would synchronise all the various markers; days of winter, nundinal days, ordinary week days, and so on.  In this way the farmer could know what time of year it was and when to go to market.

It makes you grateful for modern calendars!

It also raises a question.  If illiteracy was so widespread in antiquity as some assert, why do parapegmas exist?  The answer is perhaps that middle-class illiteracy was by no means as rare as some might think, and that farm managers would need to be both literate and numerate.

  1. [1] Both of these in much reduced form from Daryn Lehoux, Astronomy, Weather and Calendars in the ancient world.  Google books preview here.

From my diary

I’ve continued working on the PHP scripts for the new Mithras site.  It’s slow, because I don’t do much development work in PHP.  The reason for doing this is so that I can work on the site from anywhere, work or home; and so that it will support things such as footnotes, not found in standard HTML.

I was struck today by the conviction that HTML is travelling in the wrong direction.  I remember the first HTML.  It was simple, and anyone could master it.  Today I learned that all of the attributes on the horizontal rule element, the plain old <hr> tag, are to be unsupported by HTML 5.  If you wanted a single line, all you had to do was <hr size=1>.  Now, to achieve the same effect … well, I did a google search, and had to experiment to find a CSS syntax that would work.

There is a disease that affects software products.  It happens when the developers forget that 99% of the time, the user is doing a few simple things; and start concentrating on the 1%.  In this case the HTML developers are so busy trying to separate presentation from content — a mantra of much software development, and not a bad thing — that they have forgotten that the first, most important thing is that creating a web page should be SIMPLE!!!  Idiots.

I’m still under the weather, but I also opened Daryn Lehoux’s book on ancient weather and calendars, and made a start.  I was deeply impressed by the opening pages, which gave a remarkably clear reason why such calendars were necessary, and nicely anchored it in farming in modern society.  Someone give this man a professorship: he has managed to produce a seminal piece of work on a very difficult, highly technical subject, and has done it in such a way that any reasonably educated man may get up to speed.  Marvellous!

From my diary

I’ve been doing some more work on the Mithras Project pages.  This has been entirely PHP and perl coding, tho.

Daryn Lehoux kindly sent me a copy of the paperback of his book, Astronomy, Weather, and Calendars in the Ancient World.  CUP are now selling this on Amazon at USD$40.  It’s very excellent, and I shall have to spend some time with it, once my pile of books from last weekend diminishes!

Daryn Lehoux’s thesis on ancient astronomical calendars is online

I’m still thinking about parapegma, the ancient peg-calendars to predict star- and weather-movements.  A google search has revealed that the opening parts of Daryn Lehoux’s 2007 book are a version of his 2000 PhD thesis, which is online at ScribD here.   It’s a Canadian thesis, which leads me to wonder whether Canadian theses are online for free, and if so where?

The thesis doesn’t include the texts and translations.  But it doesinclude the key question, “what is a parapegma”?  The general quality of writing, tho, is considerably inferior to the final version, and it is not nearly so readable.   Those with a library room in their country estate, and $155 to spare, will be better advised to buy the book.  We peasants, however, will still gain value from the thesis.

in this work I address the broader question of how one mode of prediction, cyclical astrological prediction, functioned in the ancient world. I examine a diverse set of texts and instruments collectively known as parapegmata. These were used for predicting and tracking such things as astronomical events, day-to-day weather changes, lunar phenomena, and certain types of astrological influences. …

The word parapegma (pl.: parapegmata] refers to an ancient instrument which was used to keep track of astronomical, astrological or astrometeorological cycles using a moveable peg or pegs. By extension, the word also refer to a group of texts which were derived from these instruments, and which tracked the astrometeorological cycle typically by linking it to a calendar.

 …  they provided some means for locating the current day in the context of the larger temporal scheme, either by indexing the cycle to a calendar, or by indicating the current day with a peg. I call this process tracking a lunar or astrometeorological cycle. In inscriptional parapegmata, each entry would have a hole drilled beside it to receive a moveable peg. The peg would be shifted on each consecutive day, and thus the inscription beside the peg would contain the information pertaining to the current day.

There were also non-inscriptional, literary parapegmata in both Greek and Latin. A typical example of these would list the dates of a coming year in, for example, the Roman or Egyptian calendar, and, for particular dates, offer astronomical and weather predictions for that year. In this respect, they are rather like a pared-down version of a more modern Farmer’s Almanac. These calendars were used in Greece from at least the fifth century B.C., and there are Western European and Byzantine examples dating well into the Middle Ages and beyond.

The Romans translated Greek parapegmata into Latin, and they were developing their own versions by the first century B.C., with some interesting modifications. In particular, their inscriptional parapegmata were often used to keep track of lunar days, hebdomadal days and nundinal days. There was also a corresponding Egyptian tradition dating from at least the fourth century B.C., which may or may not be independent from the Greek.

The astronomical phenomena frequently recorded in the parapegmata are the solstices and equinoxes for a given year, and what are called the ‘phases’ of the more important fixed stars.  …

Until the early twentieth century, the only known parapegmata were found in the astronomical or divinatory manuscripts of, for example, Ptolemy, Geminus, and Johannes Lydus. A typical entry from one of these looks like this:  …

(Month of] [Thoth,] [day] 1: [at the latitude where the day is 14 1/2 hours [long], the (star) on the tail of Leo sets.  According to Hipparchus , the Etesian winds stop. According to Eudoxus, rain, thunder, the Etesian winds stop.

Looking at entries such as this, it was unclear why this sort of text was called a parapegma in Greek, which derives from the verb parapegnumi meaning ‘to fix [something) beside something else.’ The sense of this derivation remained obscure untii the discovery of the Miletus parapegmata at the beginning of the twentieth century. These newly discovered parapegmata differed from the literary ones in three respects: They were carved in stone rather than being written down in manuscripts, they contained no calendrical information [i.e., no dates in any civil or religious calendar], and they had holes bored into them beside or near the weather or astronomical entries. These holes corresponded to the number of days between, for example, two phases of a star.  …

It seems therefore probable that there was only one peg which was moved each day frorn hole to hole, thus indicating only the current date, and the current astronomical or astrometeorological situation. The empty peg holes would aliow one to count the number of days between now and the next significant event.

All this is extremely interesting, although an edition of the texts and translations, and ideally a corpus of inscriptions, is required to substantiate this.  It seems that Lehoux realised the same, when preparing the book version.

All this reminded me strongly of the calendar in the Chronography of 354, which displays the nundinal days.  I never found a good account of how these worked, but the thesis contains an explanation of these too.

Lehoux’s thesis is good.  But it remains just a thesis.  The final product, the book, is infinitely better.  I just wish I could afford to buy a copy!

Books like this one are a real blessing.  This is a book that takes a field in which there is only scholarship in German — and English-speaking scholars do not, on the whole, display much in the way of German language skills — and makes the whole subject accessible to English speakers, places it in context, and deals with questions of the kind that I have struggled with, while working on Antiochus.  It’s like a travel handbook to a new and totally unfamiliar country.  But once he has this, the interested reader will be able to locate the major cities, find a hotel, and know which sites to visit!

Manuscripts of the calendar of Antiochus of Athens

I thought it might be useful to signal how the calendar of Antiochus got to us.  We have Boll’s nice printed edition, and Daryn Lehoux’s even nicer text, translation and explanatory notes.  But … how did these get to us?  What is the text based on?  Boll lists the copies available to him, which were the following:

Ms. Vaticanus Graecus 1056 (V) is a paper manuscript of the 14th century.  The main content is three books of a five-book collection of Greek and Arabic-Persian astrological materials, apparently compiled in the Byzantine period.  The manuscript is described in detail in the Catalogus codicum astrologicorum graecorum (=CCAG), vol. 5, part 3, p.7-64, and also by Boll in his Sphaera, p.34.  In the first book, on folio 29, is the calendar or parapegma of Antiochus.  Unlike the other manuscripts, the calendar here begins in March.

Ms. Munich Graecus 287 (M) is a manuscript of the late 14th or early 15th century.  There is a detailed description in CCAG vol. 7, pp.8-24.  The calendar appears on fol. 127v-132v.

Closely related to the Munich ms. is Ms. Modena Graecus 85, belonging to the Biblioteca Estense (III C 6).  This is also known as the Mutinensis.  It was written at the end of the 15th century by a scribe named Michael Suliardus, and was later in the possession of George Valla.  See also CCAG 4, 28-33, and Boll’s Sphaera p.53 f.  The calendar is on folios 69-74v.  Franz Boll, publishing the text, gives the opinion that it is either a copy of the Munich ms., or else derived from a common exemplar, in which case, he feels, it can be disregarded.  In the latter case, of course, it could well contain some truth not found in its brother  ms. so Boll is in error here.  He prints a facsimile of two pages from it, tho, because the drawings are clearer, and the Modena ms. is also less well known.

The next manuscript is in Oxford, one of the Selden mss. (number 16) in the Bodleian library, and is described in Coxe, Catalogi codd. mss. bibliothecae Bodlianae vol. 1, p.593 f..  Boll gave it the siglum O.  It dates from the 15th century.  The calendar is on folios 147-149.  It is quite unrelated to any of the preceding manuscripts.  On f.145 are extracts which are labelled here and in other manuscripts as from the Thesaurus (=Treasury) of Antiochus of Athens.  Following the calendar are further extracts again labelled here and elsewhere as by Antiochus.

Ms. Cambridge, Trinity College, O 7, 39, pars III, is a copy of the Oxford manuscript made in the 17th century, probably by Edward Bernard for Thomas Gale (see M.R.James, The western mss in Trinity College, Cambridge, III 375).  The only deviations from O are obvious typos.

Vienna ms. philos. gr. 179, was brought from Constantinople.  It is 14-15th century, and from folio 41 onwards contains excerpts from Antiochus.  It is described in Kroll, Beschriebung Catal. VI. 28f., and also by Boll in his Sphaera p.52f.  Unfortunately there are pages missing from this section of the manuscript, but a title is given in the list of chapters, indicating that it was present.

Boll therefore edited the text from the three independent mss, M, V and O.  He found the text was better preserved in V and O, than in M.

In V and M, and the related Modena ms., the text is anonymous.  Only in the Oxford ms. does the name of Antiochus appear.   But the calender is just one chapter in O, sandwiched between other chapters under the name of Antiochus.  Some of those chapters are also in M, but again no author name appears.But we know that O is correct, for there are other astrological manuscripts which also contain these non-calendar chapters, and identify the author as Antiochus.  So we have to accept that the loss of the author’s name in M is merely an accident of transmission.  The same is true in the Vatican ms.  Boll concludes that we can reasonably suppose Antiochus to be the author, that an ancient tradition attaches his name to the text, but that when V and M were compiled, the name of Antiochus was omitted for some now unknown reason.

No biographical information has reached us about Antiochus himself, but he is one of the better known ancient astrologers, because of those who refer to him.  Porphyry, in his Isagoge of Ptolemy’s Tetrabiblos, quotes him; Hephaistio of Thebes, in book 2, chapter 1, tells us he was from Athens, quoting some of the same material word for word.  Firmicus Maternus (II, 29) mentions him.  The Arab Massala (Catal. I 82) tells us that he wrote in seven books, together with the titles of two works, περι γενεθλίων and περι ερωτήσεων.  He is also listed in the anonymous writer of the year 379 AD, in the consulate of Olybrius and Ausonius, in a list of older writers, between Valens (2nd century) and Antigonus of Nicaea (2-3rd century).  Antiochus makes use of the data of Ptolemy (early 2nd century).  Boll therefore dates him to around 200 AD, “unless some unexamined Paris or Vatican astrological manuscript” should contain more information about him.

All this may seem rather dry.  But it is worth remembering that, for many or most writers of antiquity, this sifting of slight facts is how we gain knowledge.  We are so accustomed to the Pliny’s and Vergil’s and Tacitus’ — even to the point at which people get revisionist about this work or that — that it is salutary to be reminded by what little threads we receive much of what we know.

The calendar of Antiochus of Athens as a ‘parapegma’, and an existing translation (!)

In a very useful comment on a recent post, Alexander Jones drew my attention to the term “parapegma”, and to Daryn Lehoux, “Astronomy, weather, and calendars in the ancient world: parapegmata and related texts in classical and Near Eastern societies”, CUP, 2007.

The link is to the Google Books preview.  At a price of $155,[1] this is the only way most of us will ever see any of this book.  The table of contents is online at the start, and from it I learn that Dr Lehoux is a man with a sense of humour, as well as a detailed knowledge of this recondite subject.  From the preview, the book looks very well written and referenced.  It looks like a fine piece of work, indeed.  The preview is a generous one, for which we may all be grateful.  I notice that Dr Lehoux has wisely kept the copyright in his own hands.  When it falls out of print, I hope that he will make the book available online.

Now I had never heard the term parapegmata, but this is what the calendar of Antiochus is.  The early examples were engraved on marble with holes for pegs, which could be advanced each day, as a way to determine astronomical and weather information.

Lehoux catalogues these sort of texts, and describes each, and then — I nearly missed this — gives the text and translates them.  The calendar of Antiochus is described on p.162, and is item A.x in Lehoux’s classification.  Here is what he says.

A.x. The Antiochus parapegma [27] is a short Greek parapegma that correlates stellar phases with changes in the weather and occasionally with causal statements such as ‘July 14: The whole of Orion rises at the same time as the sun; it causes (poiei=) rain and wind.’ All dates are in what I call the modified Julian calendar (i.e., dates are given as 1 July, 2 July, etc. rather than by the traditional method ofcounting down to the Kalends, Nones and Ides), which system seems to have begun to be used in the fourth century ad, rather than the sixth, as Mommsen thought.[28] Unique features of this parapegma are its mention of the ‘u9ywma of the sun’ on 10 April, and the duration of a change in the weather ‘for seven days’ on 23 May and 5 November, ‘nine days’ on 5 October, and ‘fifteen days’ on 6 November. It mentions a religious festival to celebrate the Nile flood on 22 October. Only one stellar phase is attributive (19 July: ‘Rising of Sirius, according to the Egyptians’), and it also has ‘birth of the sun, light increases’ on 25 December.


27.  Extant in six manuscripts, of which the earliest is fourteenth-century, and the latest is seventeenth.  Edition: Boll, 1910a.
28.  For this argument see Ferrua, 1985.



I wish I’d known about this book, as it would have saved me a couple of days work making my own translation.  Lehoux’s translation of Antiochus is on pp.338-343.

  1. [1] UPDATE (April 2012): Now available in paperback at $40, and on Amazon here!  Now that’s much more possible!