http://abob.libs.uga.edu/bobk/cccmenu.html 2001.03.28 http://abob.libs.uga.edu/bobk/ccc/cc032801.html ============================ * LETTERS TO THE MODERATOR * ============================ (6) THE DATE OF GERVASE'S "WONDERFUL SIGN" IN 1178 From Graeme Waddington Benny, Having been duly shamed to put finger to keyboard by Duncan Steel the following note concerns the apparently errant date used in relation to Hartung's postulated lunar impact. The note has been extracted from a series of notes that I circulated to interested parties about ten years ago and from a Usenet posting of 12 Dec 1991 (in the thread "1991 VG and Previous Impacts") wherein I pointed out the significance of the phrase "luna prima" and Hathorn's mistranslation of it in Hartung's 1976 paper. The note itself concerns only the dating of the report as given in the text, I make no comment on what the "wonderful sign" referred to may have been. To cut to the chase - the outcome is that Gervase's record explicitly gives 19 June 1178 [Julian] (*) as the date of the event that was reported to him. The commonly supposed dating of 18 June has arisen due to there being two calendrical markers in the Gervasian text - not just the [erroneous] day of the week - which apparently conflict. As a result of the translation given in Hartung's paper the second calendrical marker has been completely disregarded and, following Bishop Stubbs, the date has been universally held to be Sunday 18 June despite the fact that the record concerns an observation that could not possibly have been made on that date but which was eminently possible on the date identified by the mislaid date specifier. A single day's error in the ubiquitously assumed day of the week is all that is needed to reconcile the two date markers. So there we have it. Two markers, one referring to a possible date and one to an impossible date that can, at a pinch, be reconciled with the possible one. In spite of this, all discussion of this text has centred on the impossible date and ignored the evidence of the text itself (which is reproduced in Hartung's original paper). Graeme (*) 26 June in the proleptic Gregorian calendar ---------------------- THE DATE OF GERVASE'S EVENT OF JUNE 1178 Under the year 1178 Gervase records an appearance in the sky occurring, apparently, on the Sunday before the Nativity feast of St. John the Baptist -- that is, on June 18 in the Julian calendar (as noted by Stubbs in his marginal notes on page 276 of his edition of Gervase's Chronicle). The "event" noted concerns a distortion of the thin crescent moon after sunset and has been taken to refer to an actual impact on the moon by Hartung. Without wishing to comment on the merits or otherwise of Hartung's hypothesis, the given dating presents problems as the moon was too near conjunction at this time and would not have become visible in the evening sky until the following day, monday 19 June. Gervase's account commences thus, "Hoc anno, die Dominica ante Nativitatem Sancti Johannis Baptistae, post solis occasum, luna prima, signum apparuit mirabile, qunique vel eo amplius viris ex adverso sedentibus. Nam nova luna lucida erat, novitatis suae more cornua protendens ad orientem ; ....." The English translation of which, due to R. Y. Hathorn and as given in Hartung's 1976 paper, is "In this year, on the Sunday before the Feast of St. John the Baptist, after sunset when the moon had first become visible a marvelous phenomenon was witnessed by some five or more men who were sitting there facing the moon. Now there was a bright new moon, and as usual in that phase its horns were tilted toward the east ; ....." Here we note that Hathorn has followed Stubbs in the usual assumption that "die Dominica" refers to a Sunday, whereas in mediaeval (not medieval!) monastic tradition the phrase should more correctly be rendered as the Lord's day and as such may refer either specifically to a Sunday or, generically, to any ecclesiastical feast day (which included all sundays) in a monastry's liturgical calendar. The normal way of specifying a given weekday was through the sequence feria prima, feria ii, etc. (as per the immediately following record in Gervase's Chronicle). For what it is worth, the equating of dies Dominica with feria prima (as per Stubbs and Hathorn) is not, in itself, objectionable. Indeed, it is entirely reasonable and accords with practice elsewhere. Where the problem arises is in the fact that the following day (monday) is specified explicitly in the text via the phrase luna prima - that is, the first day of the ecclesiastical lunar month (following the 19-year cycle). Using the precepts laid down by Bede it may be readily shown that luna prima was monday 19 June in 1178. It should be stressed that there is a world of difference between classical Latin and mediaeval (not medieval) monastic Latin as used in Britain and so Hathorn (a classicist) can be forgiven for effectively changing "the first day of the moon" to "the moon had first become visible" - although it should also be noted that the phrase appears in the text isolated by commas as was usual for this kind of dating (as per the immediately following record in Gervase's Chronicle) and that alone should have set the alarm bells ringing. On this basis, the beginning of the Gervasian record of the June event should, perhaps, be rendered as "In this year, on the Lord's day before the Nativity of St. John the Baptist, after the sun had set, first day of the moon, a wonderful sign was seen by five or more men who were sitting facing it [the sunset/moon]. For there was a bright new moon and, as is its custom when new, its horns stretched towards the east ; ....." Thus, we have two date markers in the same sentence of the text which seem to refer to two separate days for the same event. Since the moon could not have been seen on the sunday (for a standard atmosphere, I calculate a maximum sighting probability of only 0.1% for this date, so there is no doubt that it could not have been seen) but would have been visible much as described later in the text (writhings etc. notwithstanding) on the luna prima monday, the weight of evidence is in favour of the monday being the date intended. (As an aside we may note that the tuesday may be excluded since the text has "luna prima" not "luna i" which could have been subject to a scribal error for "luna ii", thus making tuesday a possibility.) So, can we reconcile dies Dominica with monday 19 June ? Given that the Lord's day can, perhaps, refer to any feast day, not just sundays, it is open to us to see if the two date markers can be reconciled on this basis. From Gervase's own establishment (Canterbury, Christ Church) we see that in the century before he wrote his Chronicle the following entries were marked in the liturgical calendar for the week before the Nativity Feast of St.John the Baptist (which was itself a prime calendrical marker, being one of the quarter days), 18 June xiv kal Iulii Sacctorum Marci et Marcelliani 19 xiii Sanctorum Geruasii Et Prothasii 20 xii Solstitium 21 xi 22 x Sancti Albani martiris 23 ix Sancte AEpeldrypae uiginis 24 viii Natiuitas Sancti Iohannis Baptiste Thus, treating dies dominica as a truly generic feast marker would imply that we are dealing with the 23 June (the feast of Etheldreda of Ely), in which case why not simply refer to the date as the vigil of St.John rather than the feast before St. John? In any case, the 23rd cannot be reconciled either with luna prima or the statement that the new moon was shining brightly (as per the second sentence of the text). But what about the feast of the protomartyrs of Milan, one of whom was Gervase's namesake? Since this date accords with the mistranslated date-specific marker (luna prima) in the text, could Gervase have been having an "in-joke" here? Moreover, in the Christ Church calendar it is also the day before the assumed solstice so any possible confusion between the solstice and the quarter day could allow us to reconcile this "feast" marker with that from the ecclesiastical lunar cycle (given that Gervase is repeating details given to him by others such an error could well be understandable, if a little strange for a Chronicler of Gervase's standing). Thus, it is possible to reconcile the impossible date specification with the possible one, although, as with most things in annals of this period, it remains a somewhat less than satisfactory state of affairs. Even so, I offer the thought for what it may be worth. W.G.Waddington MRC Biochemical & Clinical Magnetic Resonance Unit Magnetic Resonance Spectroscopy John Radcliffe Hospital Oxford =============