The Peoples of Middle-earth (24 page)

(Entare) and 2; the first day of Viresse (April 1); the first day of Lotesse (May 1); and the last day of Lotesse (May 30). The second column of figures is the cumulative total of days in each season corresponding to those in the first column: thus 1 Lotesse is the 33rd day of the year (2 Tuilear, Viresse 30 days, Lotesse 1, = 33).

In Appendix D this elegantly balanced structure had been abandoned and replaced by a different system, somewhat obscurely recounted. But it seems extremely probable that it was here, in this original account of the calendar of the New Era, that the Quenya names of the months first entered. It will be seen that the names given in Appendix D are all present except that of February, Nenime, which is here Nendesse; while the opening day of the year is Entare (on the first form of the table written Entale and then changed) for later Yestare, and Midyear's Day is Arendien for later Loende. The changed names were written onto the table later, probably much later.

On the back of the second table, and clearly intended to be continuous with it, is the following.

Alternative names.(11)

Autumn: Endien (Midyear).

April: Ertuile October: Lasselanta (Noquelle) May: Notuile November: Errive

June: Ellaire December: Norrive

July: Nolaire January: Meterrive

August: Metelaire February: Ercoire

September: Erquelle March: Nocoire

The Noldorin month names and seasons corresponding were: Spring: Ethuil: April Gwirith. May Lothron.

Summer: Loer: June Norui. July Cerfeth. August Urui Autumn: Firith: September Ifonneth. October Narbeleth.

Winter: Rhiw: November Hithui. December Girithron.

January Nerwinien.

Stirring: Echuir: February Nenneth. March Gwaeron.

Occasional variants: Autumn: [Dant, Dantilais >] Dannas, Lasbelin. June: Ebloer. July: Cadloer. December: Ephriw.

January: Cathriw.

The old Year beginning corresponded to N[ew] E[ra] January 8, the 68th day of Winter, and 281st day of the year.

The old Midyear corresponded to N.E. July 6, the 36th day of Summer, and 98th day of the year.

The old Year's-end corresponded to N.E. January 7, the 67th day of Winter, and 280th day of the year.

The Shire Lithedays = N.E. July 5, 6, 7

The old Gondor Summerdays = N.E. July 4 - 8 inclusive.

Here the Noldorin (Sindarin) names first appear, and are the same as those given in Appendix D, with two exceptions: Nerwinien for Narwain (January) and Nenneth for Ninui (February). Narwain and Ninui were written in much later, together with the following changes: Metelaire, Meterrive > Mettelaire, Metterrive; Cerfeth > Cerveth; Ifonneth > Ivanneth; Loer > Laer, Ebloer > Eblaer, Cadloer > Cadlaer.

Together with many other tables of comparative reckoning, some beautifully made, this effectively completes all the calendar material that I can certainly identify as belonging to this primary phase of my father's work on the subject. It is remarkable that (amid abundant but very rough and difficult notes) there is no further text until the typescript (itself rough and a good deal emended) from which the text of Appendix D was printed.(12) It seems very unlikely that any intervening text should have been lost, and I am inclined to think that my father was still developing and refining his theory of the Calendars when the need to submit his text to the publishers became imperative and urgent. Of the final form I believe that what I said of Appendix F, that

'had circumstances been otherwise the form of that appendix would have been markedly different' (p. 82), can be repeated of Appendix D

with greater force.(13)

NOTES.

1. I have added the final 6, absent in the manuscript.

2. As noted on p. 119 this is a reference to the manuscript F 2 of the Appendix on Languages, pp. 38 - 9, $$23-4. The sentence at the end of $24, 'But most remarkable of all are the Hobbit month-names, concerning which see the note on Calendar and Dates', was an addition, but one made with care near to the time of writing of F 2 - no doubt when my father reached this point in D 1.

3. Rethe was not the original name. On the back of this page of the manuscript is a comparative table, struck through, of Hobbit dates and modern dates by month and day; and the third month is here Luyde, not Rethe. Luyde occurs also in one of the Prologue texts (see p. 9), but I have not found it anywhere else; nor have I found any other names preceding those in the list given here, which survived without change.

4. Winterfylleth was the Old English name of October. Its meaning was discussed by Bede (died 735), who explained the name by reference to the ancient English division of the year into two parts of six months each, Summer and Winter: Winterfylleth was so called because it was the first month of Winter, but fylleth, Bede supposed, referred to the full moon of October, marking the beginning of that period of the year. My father's interpretation of the name in D 1, 'the filling (completion) of the year', 'winter'

being used in the sense 'year', is at variance both with Bede and apparently with that in the published text (RK p. 388, footnote),

'the filling or completion of the year before Winter'. In either case it must be supposed that the 'true' words underlying translated

'filth' and 'Winterfilth' could make the same pun!

On the former beginning of the Hobbit year after Harvest see p. 121, footnote.

5. The Great Writ of Tuckborough (later the Yearbook of Tuckborough) is mentioned also in the Appendix on Languages, p. 40, $27.

6. The words '(in spoken form)' refer to the spelling Munday. In the published text the sentence reads: 'In the language of the time of the War of the Ring these had become Sterday, Sunday, Monday ...', instead of 'had become written', thus avoiding the question of Munday, Monday: the latter being a mere peculiarity of English spelling, as in many other words with the vowel of but, as monk, son, etc. Sunday was once often spelt Sonday. - So also Hiday of D 1 is given the modern spelling Highday in the published text. But in the list of archaic (Old English) forms it should not be Highdei but Hihdei, as in D 1.

7. Contrast Appendix D (RK p. 384): 'the merrymaking in that year

[1420] is said to have been the greatest in memory or record.'

8. The different computation in D 1 (p. 120), whereby the Eldarin calendar would have 36525 days in a century (leading to the removal of the last day of a century) was reached thus: the last four years of a century were computed as half of an eight-year cycle, that is 2922 (seven long years and one short year) divided by two, 1461. Added to the total (35064) of the days in 96 years this made 36525. In D 2 the last four years are not half of an eight-year cycle but three full years and one short year, that is 35064 + 1458 in a century, total 36522 (leading to the addition of the two extra days outside the structure of weeks and months, Quantdrie and Vinydrie).

9. I have found no further list of the Elvish names of the days of the week, nor any mention of individual names, before the third (and final) text, from which Appendix D was printed (see p. 136).

There, the Quenya names of the fourth and sixth days, Aldarya and Valarya, were still in that form (but Tarion had replaced Tarinar); my father emended them very clearly, on both copies of the typescript, to Aldauya and Valanya. On the proof of the first of these was printed Aldanya, and he emended this to Alduya, as it appears in Appendix D.

On the use of 'Noldorin' for 'Sindarin' see the Appendix on Languages, especially p. 36, $18, and commentary, pp. 65-6.

These month-names next reappear in the final typescript, already changed to Orgilion, Oranor, Orithil, Orgaladhad, Ormenel, Orebelain (or Rodyn); and similarly with the name of the

'Sea-day' added by the Numenoreans, changed from Aroeren to Oraearon.

10. In the Eldarin calendar (p. 126), in the long (normal) years, January had 31 days and February 30; thus March 25th was the 86th day of the year. In the Gondorian King's Reckoning (p. 131) the same was true. In the Shire Reckoning Yuleday preceded January 1, but both January and February had 30 days, so that March 25 was again the 86th day of the year.

In the Gondorian Steward's Reckoning (p. 132), on the other hand, the count is simply 30 days in January and 30 days in February, so that March 25 is the 85th day of the year; while in our calendar 31 + 28 + 25 makes March 25 the 84th day.

11 Some of these alternative names are included in the first form of the table, with the difference that Errive and Norive (so spelt) are alternatives respectively for Ringare (December) and Narvinye (January).

12. It is a curious point that this typescript begins with the printed

'Shire Calendar for use in all years' exactly as it appears in Appendix D (RK p. 384): my father's typescript begins below it ('Every year began on the first day of the week ...'), and the same is true of the carbon copy. Presumably this calendar was printed first and separately and copies were sent to my father, who used them in this way.

The manuscript calendar from which this was printed is extant, and it is interesting to see that on the left-hand side there is a column headed 'Weekday' with the names of the days of the week set out against each of the three transverse groups of months, thus for example in the second month Solmath: Stirday - 5 12 19 26

Sunday - 6 13 20 27

Munday - 7 14 21 28

Trewsday 1 8 15 22 29

and so on. This column of the days of the week would have made the calendar easier to understand; but on the manuscript it is struck through, by whom is not clear. I can see no reason for this but that of space on the page, which one would think could have been quite easily accommodated. - This manuscript table undoubtedly goes back to the original phase of my father's work on the calendars, described in this chapter.

13. Among various alterations made by my father on the proof, it may be noted that the text of Appendix D as first printed ended thus: 'Some said that it was old Sam Gardner's birthday, some that it was the day on which the Golden Tree first flowered in 1420, and some that it was the Elves' New Year. The last was (more or less) true, so all may have been.'

V.

THE HISTORY OF THE AKALLABETH.

The development of Appendix B, The Tale of Years, was naturally associated with and dependent on that of Appendix A, which as published bears the title Annals of the Kings and Rulers. But more unex-pectedly, the Tale of Years of the Second Age was closely associated with the evolution of the history of Numenor and of the Akallabeth.

In the presentation of the early forms of these Appendices I have found after trial and error that the best course is to divide the Tale of Years into two parts, the Second and the Third Ages, and to treat them separately; and also, to introduce at this point an account of the Akallabeth, followed by the Tale of Years of the Second Age in Chapter VI.

In the History of Middle-earth I have given no indication of how this work, a primary narrative of the Second Age, developed to the form given in the published Silmarillion, or when it first came into being. The early history of the legend, closely related to the abandoned story The Lost Road, was studied in Volume V, where the two original narratives of The Fall of Numenor, which I called FN I and FN II, were printed (V.13 ff.). In Sauron Defeated (IX.331 ff.) I gave a third version, FN III, which I have ascribed to a fairly early stage in the writing of The Lord of the Rings. The massive development of the legend in the work called The Drowning of Anadune, closely associated with The Notion Club Papers and the emergence of the Adunaic language, was studied in Sauron Defeated, where I ascribed it to the first half of 1946 (IX.147, 389-90): this dating was subsequently con-firmed by the observation of John Rateliff that W. H. Lewis recorded in his diary that my father read the work to the Inklings in August 1946 (Foreword to Morgoth's King, X.x).

In my commentary on The Drowning of Anadune I indicated and discussed at many points its relationship to the Akallabeth, but for the text of the latter I made use only of the final form, as printed in The Silmarillion, pp. 259 ff. Since the writing of the Akallabeth evidently post-dated the writing of The Lord of the Rings I postponed discussion of its history, but I found no room for it in the very long books Morgoth's Ring and The War of the Jewels.

I did, however, in Sauron Defeated make an extraordinary mis-statement on the subject of the Akallabeth, which must be repaired.

When discussing (IX.406) my father's late note on The Drowning of Anadune, in which he referred it to 'Mannish tradition', I said: The handwriting and the use of a ball-point pen suggest a relatively late date, and were there no other evidence I would guess it to be some time in the 1960s. But it is certain that what appears to have been the final phase of my father's work on Numenor (A Description of Numenor, Aldarion and Erendis) dates from the mid-1960s (Unfinished Tales pp. 7 - 8);(1) and it may be that the Akallabeth derives from that period also.

This last remark is patent nonsense. The great extension of the line of the Numenorean kings, which entered in the course of the development of the Akallabeth, was present in Appendix A (and a mere glance through the texts of the work is sufficient to show, simply from their appearance, that they could not conceivably date from so late a time).