An Old Mistake

The Easter section of the old Book of Common Prayer, from the 1662 Act of Uniformity to 1750, starts with an erroneous description of Julian Easter — "EAſter-day (...) is always the firſt Sunday after the firſt Full Moon which happens next after the One and twentieth day of March. ...". It should say on or after March 21st.

The Calendar Act of 1751, Section 3, repeats the words; but after And whereas according to the rule prefixed to the book of common prayer ...; so it makes a true statement about a false one. Note that the "enacted" part of the Section follows.
[www.merlyn.demon.co.uk]
-------------------------


John Wallis and the Date of Easter
On February 22nd The Times published three letters from readers who set out to explain why Easter this year, 2008, fell as early as March 23rd. All three readers concurred that Easter is the Sunday after the first full moon after the spring equinox; the first two put the equinox on the 21st and the full moon on the 22nd. In fact, however, the equinox fell on the 20th, as it does more often than not, and the full moon on the 21st. Nevertheless, the 21st for the equinox and the 22nd for the relevant moon are the dates adopted in the ecclesiastical calendar, the former because the church believes that it has been the official date since at least AD 325, the latter because it follows from the lunar calendar drawn up by the church long ago. Accordingly, the same two readers gave the 23rd as the earliest possible date for Easter. Why, then, has it sometimes fallen on March 22nd, most recently in 1818?

The Book of common prayer gives this formula: ‘Easter Day … is always the first Sunday after the full moon which happens upon, or next after, the twenty-first day of March; and if the full moon happens on a Sunday, Easter Day is the Sunday after.’ The formula allows March 22nd if the full moon happens on Saturday March 21st. The three readers trimmed ‘upon or next after’ to ‘next after’, and the two who gave the 23rd as the earliest possible date were consistent but wrong.

From 1662 to 1750, however, the Book of common prayer supported the three readers: it said just ‘next after’. Robert Watts, fellow of St John’s, Oxford, who in 1712 surveyed the history of fixing Easter, traced this earlier formula to John Cosin, who in 1627 had laid it down that ‘Easter Day … is alwaies the first Sunday after the full moone which beginneth next the equinoctiall of the spring in March.’1 A Latin version in the Sarum Missal goes back at least to 1504: ‘Post veris equinoctium quere plenilunium, et Dominica proxima sacrum celebra Pascha,’ which perhaps began life as a mnemonic in accentual verse: Post veris equinoctium/ require plenilunium,/ et proxima Dominica/ Paschale sacrum celebra (After the spring equinox look for the full moon, and on the next Sunday keep the rite of Easter).

When Parliament in 1751 (24 George II) passed the Act that moved the country over to the Gregorian calendar, it repeated without comment the formula adopted in 1662, but the change to ‘upon or next after’ was made in the accompanying tables, which have appeared in the Book of common prayer ever since. Before the change, if the tables conflicted with the formula, the actual dates of Easter show that the tables counted; but they rested ultimately on what should have been the same formula.

Though the formula had been discussed in the wider debate that led to the Act, a bitterer debate had taken place decades before. As early as 1664 the moon flagrantly ignored the ecclesiastical calendar, and nonconformists refused to be bound by a work that contained untruths about the heavens. Defenders of the work replied that just as March 21st was the ecclesiastical equinox, not the astronomical one, so the full moon was the ecclesiastical full moon, not the astronomical one. Nevertheless, the nonconformists’ protest may account for a change made to Cosin’s formula when the seventh edition of his book appeared in 1664: ‘Easter Day is alwaies the Sunday after the day which is called the Easter limit, which is found for any year by the help of its golden number in the table before.’2

The difference between the ecclesiastical and the astronomical moon no longer needed labouring when a distinguished mathematician stepped in: John Wallis of Emmanuel, who incorporated from Exeter when he took up the Savilian chair of Geometry at Oxford in 1649. In a letter of 1684 to John Fell that chiefly concerned whether St Matthias’s Day in leap years should fall on the 24th of February or the 25th, he mentioned that the formula for Easter in the Book of common prayer ruled out March 22nd—even though, in the Julian calendar, it had fallen on that day in 1668—unless ‘next after March 21st’ meant ‘upon or next after March 21st’, a sense at odds with the sense of ‘after’ in the second clause, ‘if the full moon happens on a Sunday, Easter Day is the Sunday after.’ The required sense, he suggested, would be conveyed by an alteration in that clause: replace ‘the Sunday after’ with ‘that Sunday, not the Sunday after.’

The letter went unpublished, however, until Watts published it in 17113, and by then Wallis had published an opinion requested by the Lord Chief Justice on the date of Easter in 16984.
There, as in the letter of 1684, he assumed that ‘full moon’ meant ‘15th day of the lunar month’, and he repeated in different words his formula of 1684:
The fundamental rule of the Nicene Council (which we pretend to follow) for the keeping of Easter is to this purpose: Easter Day is to be that Sunday which falls upon, or next after, the first full moon which happens next after the vernal equinox; which vernal equinox was then observed to fall on the one and twentieth of March, and (in the Paschal tables) is yet reputed so to fall.
Evidently it had struck him in the meantime that by substituting ‘upon or next after’ for the first ‘after’, not the second, he could do without his second clause. By 1712, several objections had already been levelled at Wallis’s formula, but his critics did not always see that he had merely adapted to the 15th moon the formula they were defending. For them the ecclesiastical full moon was the 14th moon, which the Pentateuch lays down as the date of Passover (Exodus 12:1-11, Leviticus 23:5, Numbers 9:1-5, 28.16); and in a leaflet recommended as proper to be bound with Common-prayer books. Watts had just glossed the formula of 1662 by first specifying the relevant new moon and then adding ‘the full moon meant in this rule is the 14th day after inclusive.’5 For this equivalence he found precedent in a Bible of 1576, which placed the Passover ‘upon the fourteenth day of the first month, to wit, at the full of the moon.’6 As lunar months have roughly 29½ days, conventionally treated as an alternation of 29 and 30 and by the church as 29 in the neighbourhood of March 21st, the full moon ought to be their 15th day; but no-one had ever doubted that by the 14th moon the Jews meant the full moon, presumably reckoning not from the astronomical new moon but from the first sight of it7. ‘The full moon in mid-month,’ it has been said ‘would provide maximum light for an evening ritual.’8
Pope Gregory XIII had taken a different view. In the bull of 1582 by which he reformed the calendar, Inter gravissimas9, the term ‘full moon’ (plenilunium) nowhere appears. Instead he fixes Easter by reference to the 14th moon. In his commentaries on the reform, Clavius said explicitly that the 14th moon was the day before the full moon10. In his tables, though, he placed his 14th moon in such a way that it would stay closer to the real full moon than his 15th moon would11. Nevertheless, critics pointed out that if Easter was to be kept on the Sunday after the 14th moon, and the 14th moon was not the full moon, the Sunday after it might well be; yet keeping Easter on the full moon, the day of the Passover, was surely what the early church had eventually agreed to ban12.

Wallis never mentioned the 14th moon, but he must have understood it as Clavius did. Perhaps he chose not to cite him for fear of seeming to align himself with Rome, or perhaps he just went by the arithmetic. In any event, the difference between the 14th and 15th moon yields different dates for Easter whenever the 14th moon falls on a Saturday, and so each needs its own formula if both are to fit the tables. He neatly gave one for the 15th by moving ‘upon or next after’ to the Sunday from the full moon, where the Book of common prayer ought to have put it but instead had put just ‘next
after’.
Not only did his critics object that he allowed Easter to be kept on the full moon, to which Clavius would have replied that what mattered was not keeping it on the 14th moon, but they bent over backwards to defend ‘after’. They urged that ‘after March 21st’ could indeed mean ‘next after the commencement of March 21st’, i.e. ‘upon or next after March 21st’.

According to one defence of this interpretation, inclusive reckoning went back to Antiquity, and Watts even turned Wallis’s own documentation of the practice against him; but not even taken inclusively does an expression like ‘the 40th day after’ mean ‘the 40th day on or after’, and Wallis had contrasted the inclusive reckoning of Antiquity and other countries with English practice. According to another defence, ‘the Sunday after the full moon’ already needed to mean ‘the Sunday on or after the full moon’ if the second clause in the formula of 1662 was not to be a mere restatement of the first13; this defence contained an important element of truth, that the second clause was redundant, but turned restatement into contradiction by interpreting ‘the Sunday after the full moon’ as ‘the Sunday on or after the full moon’ only for ‘on the full moon’ to be ruled out by the second clause. According to yet another defence, ‘March 21st’ meant the vernal equinox, and the full moon could well happen after it on the same day14; but though that may well explain why the relevant moon was originally allowed to fall either next after the equinox or on it, as an interpretation of ‘after March 21st’ it was more ingenious than plausible. Watts accepted the critics’ objections and declared that there was nothing amiss with the second clause, but not before admitting that the formula used ‘after’ in two senses and giving these words to a fictional Dissenter:
This is a very odd way of speaking, methinks. I fancy you will not find many that will believe the words will bear this sense you put upon them.

Even so, it might have seemed that Wallis had lost the argument by making a fundamental mistake about the full moon; but he had died in 1703 and could not answer back.

In 1737, however, one William Jackling, ‘philomath’, noticed that the tables put the 14th moon on or after March 21st, which I must confess seems to me to be contrary to the meaning of the rule in our Book of common prayer, for there it is expresly said, the first full moon AFTER, and not UPON, the 21st day of March15. He does not mention Wallis’s letter of 1698, but in 1747 it was reprinted without the objections of his critics16. In 1750 Lord Macclesfield, when he gave the President of the Royal Society a foretaste of the case destined to sway Parliament, spoke of ‘that fourteenth day of the moon inclusive, or that full moon, which falls upon or next after the 21st day of March’17; and at last, in the tables that accompanied the Act of 1751, Wallis carried his point. That at the same time he lost a bigger battle, against the adoption of the Gregorian calendar, is another story18.
Michael Reeve
Pembroke College, Cambridge


1 Anon. (he identifies himself elsewhere), The rule for finding Easter in the Book of common prayer explain’d and vindicated (London 1712) i-ii, from A collection of private devotions (London 1627), p. 35 in the edition of P. G. Stanwood & D. O’Connor (Oxford 1967).
2 Stanwood & O’Connor (n. 1) 303. Watts noticed the change: op. cit.(n. 1) i n. *.
3 Anon. (see n. 1), The true time of keeping St Matthias’s day in leap years (Oxford 1711) 11-43.
4 Philosophical transactions of the Royal Society 20 (1698) 185-9.
5 Anon. (see n. 1), The rule for finding Easter in the Book of common prayer explain’d (London 1711) 3.
6 Op. cit. (n. 1) iii-v. The Bible in question, where see f. iiii r, is no. 144 in A. S. Herbert, Historical catalogue of printed editions of the English Bible 1525-1961 (London 1968) 82.
7 So Watts, op. cit. (n. 1) 23, and H. Prideaux, The Old and New Testament connected II (London 1718) xiv-xv; see also W. H. C. Propp, Exodus 1-18 (New York 1998) 383, 390.
8 Carol Meyers, Exodus (Cambridge 2005) 96.
9 The phrase leads immediately on the internet to a text and translations.
10 Novi calendarii Romani apologia (Rome 1588) 23-7, 254-5; Romani calendarii a Gregorio XIII p. m. restituti explicatio (Rome 1603) 64-8, 380-81.
11 Apologia 140-41 and 256-7 = Explicatio 128-9 and 381-2, mocked by J. J. Scaliger, Hippolyti episcopi canon Paschalis (Leiden 1595) 53-5, and more amusingly by A. De Morgan in the Companion to the almanac for 1845, pp. 17-20; but De Morgan too, pp. 35-6, rejects an astronomical Easter, which might fall a week later at St Paul’s than at Westminster Abbey.
12 Scaliger, p. 52.
13 J. Jackman, Philosophical transactions of the Royal Society 24 (1704-5) 2126.
14 Anon. (W. Bunbury, identified in the copy that he presented to Sidney Sussex College Cambridge), The rule for finding Easter explain’d and vindicated (London 1709) 10-11.
15 The Paschal solemnity rectified (London 1737) 64-5. Prideaux (n. 7) xiii,235, 242, misreads the tables and tacitly passes from ‘after’ to ‘upon or next after’.
16 J. Hodgson, An introduction to chronology (London 1747) 80-85.
17 Philosophical transactions of the Royal Society 46 (1749-50) 428.
18 R. Poole, Time’s alteration: calendar reform in early modern England (London 1998) 92-101; for this reference I thank Jennifer Higham of Lambeth Palace Library. On the matters aired here I consulted with profit Leofranc Holford-Strevens, Eamon Duffy, John North, and John Stockton.