•YULE AND CHRISTMAS
THEIR PLACE IN THE GERMANIC YEAR
BY
ALEXANDER XILLE, Ph.D.
LECTURER IN GERMAN LANGUAGE AND LITERATURE IN THE UNIVERSITY OF GLASGOW
DAVID NUTT, 270-271 STRAND, LONDON 1899
^^HBHAL
I 7S77f
Only two hundred copies of this book are for sale
GLASGOW ! PRINTED AT THE UNIVERSITY PRESS BY ROBERT MACLEHOSE AND CO,
THIS BOOK IS DEDICATED TO MY DEAR FRIEND,
GEORGE NEILSON,
AUTHOR OF "trial BY COMBAT," " PEEL ; ITS MEANING AND DERIVATION," " CAUDATUS ANGLICUS," ETC., ETC
TO WHOSE PERSONAL ASSISTANCE, LEARNING, AND LIBRARY
IT OWES MORE THAN TO ANY SOURCE REFERRED
TO IN THE TEXT OR NOTES.
175779
PREFACE
This book treats of the problems connected with the Germanic year — the three-score-day tide of Yule, the Germanic adoption of the Roman calendar, and the introduction of the festival of Christ's Nativity into a part of the German year, which till then had apparently been without a festivity. It traces the revolution brought about by these events, in custom, belief, and legend up to the fourteenth century. By that time, the Author believes, most of the fundamental features which go towards the making of modern Christmas had already come to have their centre in the 25th day of December.
Five chapters of the present book — but somewhat shortened — appear simultaneously in the Proceedings of the Glasgow Archmological Society.
ALEXANDER TILLE.
2 Strathmore Gardens, Hillhead, Glasgow, March, 1899.
CONTENTS
CHAP.
I. The Germanic Year,
II. The Beginning of the Anglo-German Year, ,
III. The Feast of Martinmas,
IV. Martinmas, and the Tri-Partition of the Year, V. Martinmas, and the Dual Division of the Year
VI. Martinmas and Michaelmas,
VII. Solstices and Equinoxes, .
VIII. The Calends of January,
IX. Tabula Fortunae, ....
X. The Nativity of Christ,
XI. Beda, De Mensibus Anglorum, .
XII. Nativity, Christes M^ss, and Christmas,
XIII. The Scandinavian Year,
XIV. Scandinavian Offering Tides, . ... XV. Scandinavian Yule,
XVI. Results,
I
24
34
49
57
71
81
107
119
138
158
177
189
200
214
YULE AND CHRISTMAS:
THEIR PLACE IN THE GERMANIC YEAR.
CHAPTER I.
THE GERMANIC YEAR.
The oldest descriptive remark on the mode in which the Germanics divided their year is exactly eighteen hundred years old. It is found in the Germania of Tacitus, which, in all probability, was written a.d. 98, and runs thus : " They do not divide the year into so many seasons as we do. Only winter, spring, and summer have a name and a meaning among them ; the name of autumn they know as little as its gifts." ^ It plainly means that the Germans of the first century of our era divided their year into three seasons, the names of which cannot, of course, have exactly corre- sponded to the Latin terms, htems, ver, and aestas, each covering a quarter of a year. This statement has been assailed from various sides, and for various reasons, even Jacob Grimm expressing his belief that it was based on some misconception by Tacitus.^ He understood Tacitus to refer solely to the meaning of the words, and remarked that the Romans did not use
^ Germania, chap, xxvi., " Unde annum quoque ipsum non in totidem digerunt species : hiems et ver et aestas intellectum et vocabula habent ; autumni perinde nomen et bona ignorantur."
^Deutsche Mythologie, p. 717.
A
3 YULE AND CHRISTMAS
the name of atitumnus for the harvest of grain, but for the gathering of fruit, vintage, and after-math, things which were at that time unknown to the Germans. But such a view is scarcely tenable. For Tacitus speaks decidedly of the seasons as such, and in the case of autumnus, at the non- existence of which the Romans might wonder, he makes an explanatory and rather melancholy observation. In course of time, on a closer study of the questions connected with the Germanic partition of the year, extensive material has been discovered which undoubtedly goes to support Tacitus. Grimm himself lived to collect part of it, and to admit that he had been wrong.^
Another scholar has told us that he knows better than Tacitus, and that the ancient Germans had the word herbst, with the meaning " time of fruits." But that word seems to have meant originally, just like English harvest, the act of reaping the ripe grain and fruits, and not the time of their ripeness, though it was later used to denote the period of bringing in the harvest. Considerations of that kind can as little influence our judgment V on Tacitus' report as can the fact that we are unable to say exactly
which German word he meant to correspond with Latin ver, spring ; for spring, lent (German Lenz), and Friihling are, as is generally admitted, of later growth.
The tri-partition of the Germanic year is an unshakable fact. It has been preserved for a very long time on legal ground. The three seasons answer to the three not-ordered law courts, i.e., the three annual legal meetings which were fixed by tradition and not called by special royal ordinance. This fact is even admitted by Professor Weinhold of Berlin in his book on the German division of the year, who, on the whole, takes the view that the Germanics, just like the Romans, quartered their year according to solstices and equinoxes.^ Professor Weinhold, however, there concedes so much as to acknowledge that those law courts were originally held at the beginning of winter, in spring, and about midsummer : whilst later the beginning of winter, midwinter, spring; midwinter, Easter, mid-
^ Geschichte der deutschen Sprache, Leipzig, 1848, Vol. L, p. 74. ^ Uber die deutsche Jahrteilung, Kiel, 1862, p. 8.
THE GERMANIC YEAR 3
summer ; and February, May, autumn, took the place of those terms- Professor Weinhold gives ^ proofs for the several cases. Others agree with him in this proposition. So the greatest German authority on chronology, Grotefend, says : ^ *' The tri-partition of the year has been preserved almost exclusively in juridical relations, and there finds its principal appli- cation in the so-called dreidinge, echteitdinge, echtendage, or eiting, the not-ordered law court of the country, which was held at three terms in the year. The terms vary, though with a general prevalence of midwinter (or beginning of winter), Easter, and midsummer (also the Twelve-nights, Easter, and Pentecost, or St. John's day), the basis of the tri-partition being a division of the year into winter, spring, and summer."
The capitulary of Louis the Pious, of 817, ordains "/« anno tria solummodo generalia plactda"^ which, of course, can only be taken as a codification of existing law, and not as a creation of a new jurisdiction. This usage lived on till at least the fifteenth century.^ The fact of the early existence of three German annual law courts is so generally admitted that it is an exception for any authority to disagree. And even those who disagree have to account for a number of important indisputable facts. So Pfannenschmid,^ believing that there were four Germanic law assemblies annually, finds it extremely strange that far more frequently only three such assemblies are enumerated, and that the examples of four assemblies are both rarer and later than those of three.
In Anglo-Saxon times the tri-partition of the year was preserved in the mode of paying the wages of female servants, who received a sheep for the feast at the beginning of winter, a measure of beans for the mid- lent dinner (Sunday Jnvocavit), and whey ' on siimera ' (corresponding
^ C/l>er die deutsche Jahrteilung, Kiel, 1862, pp. 18, 19.
^ Zeitrechnung des deutschen Mittelalters, 1891, p. 90, Jahreszeitefi.
^ Sohm, Fr'dnkische Keichs- und G eric htsverf as sung, p. 398.
* " 1407 in unsen geheygeden gerichten to Luneborch drie des jares to den eddagen " (Centralarchiv zu Oldenburg), Grotefend, Zeiirechnung, II., 2, 194, Hannover und Leipzig, 1898.
* Germanische Erntefeste, 1878, p. 338. .
^
4 YULE AND CHRISTMAS
to Old Icelandic ^ at sumri^ i.e., June 9), which is about July lo.^ It not only appears from the value of the gifts mentioned that the gift for the winter feast was the largest, but besides the enumeration of the three terms begins with that term, as the old Germanic, and so late as in the eleventh century the economic, year began with it.^
In the thirteenth century three terms existed in some districts of Eng-
^ Thorpe's Ancient Laws, I., 436, 7, "Rectitude Ancillae : Uni ancillae Vlll. pondia annonae ad victum. i. ovis vel III. denarios, ad hiemale companagium, i. sester fabae ad quadragesimalem convictum. In estate suum hweig vel i. denaiium ; Be Wifmonna Metsunge. Dheowan wifmen viii. pund comes to mete, i. sceap odhdhe 11 1. peningas to winter-sufle, l. syster beana to Isengten-sufle. hwseig on sumera odhdhe l. pening."
2 Male servants also received three such gifts a year [Ibid., I., 436, 7: "Omnibus ehtemannis jure competit Natalis firma, et Paschalis sulhsecer, id est, carruce acra, et manipulus Augusti in augmentum jure debiti recti ; Eallum gehte-mannum gebyredh Mid- winter feorm. and Eastor-feorm sulh-aecer. and hserfest-handful. to-eacan heora nyd-rihte "), though two of the terms for these had, in the eleventh century, shifted to the two Christian festivals, Christmas and Easter, while the third had, in the same direction, moved onwards to August. The payment of shepherds' wages is regulated not so much by an old tri-partition of the year as by the development of sheep during the year [Ibid., I., 438, 9 : " Pastoris ovium rectum est, ut habeat dingiam xii. noctium in Natali Domini, et i. agnum de juventute hornotina, et l. belflis, id est, timpani vellus, et lac gregis sui, Vli. noctibus ante equinoctium, et blede, id est, cuppam plenam mesgui de siringia, tota estate ; Sceap- hyrdes riht is that he hsebbe twelf nihta dhingan to Middan-wintra. and i. lamb of geares geogedhe. and I. bel-flys. and his heorde meolc Vll. niht sefter emnihtes dsege. and blede fulle hweges odhdhe syringe ealne sumor"), just as the payment of goatherds is [Ibid., I., 438, 9 : " Caprarius convenit lac gregis sui post festum Sancti Martini, et an tea pars sua mesgui, et capricum anniculum, si bene custodiat gregem suum ; Be Gat-hyrde. Gat-hyrde gebyredh his heorde meolc ofer Martinus msesse dseig. and ser dham his dsel hwasges. and i. ticcen of geares geogodhe. gif he his heorde wel begymedh "). The dinners given to the farm servants varied considerably about 1030, being held partly at the two Christian festivals, Christmas and Easter, partly at other times [Ibid., I., p. 440, i : "In quibusdam locis datur firma Natalis Domini, et firma Paschalis, et firma precum ad congregandas segetes, et gutfirma ad arandum, et firma pratorum fenandorum et hreaccroppum, id est, macoli summitas, et firma ad macolum faciendum. In terra nemorosa, lignum plaustri ; in terra uberi, caput macholi : et alia plurima fuerint a pluribus, quorum hoc viaticum sit, et quod supra diximus ; on sumere [in some !] dheode gebyredh winter-feorm. Easter-feorm. ben-form for ripe, gyt- feorm for yrdhe msedh-med hreac-mete. aet wudu-lade wsen-treow. set corn-lade hreac-copp. and fela dhinga de ic getellan ne mgeig. Dhis is dheah myngung manna biwiste and eal that ic ser beforan ymberehte ").
THE GERMANIC YEAR 5
land.^ On German ground it is the very same. To give at least one instance, which covers both the law courts and the terms of payment from the twelfth century: on Jan. 9, 1106, Archbishop Frederick I. of Cologne fixed at 14 solidi the fee to be paid to the provost of Gerresheim on each of the three annual law-days. ^
However well established these facts are, etymology cannot be adduced in favour of an ancient tri-partition of the Germanic year ; ancient names of three ancient seasons cannot be given ; nay, etymology decidedly points to a dual division.^ We have, therefore, to accept this as a fact, as well warranted as the tri-partition of the economic year itself. Whilst no other Aryan language possesses the same terms denoting a period of about a hundred and eighty days, all Germanic languages have in common the two words winter and summer, whilst there is no third season-name to join them. Nay, even more : the word winter appears in no Aryan language except the Germanic, all other Aryan languages using for the denomination of the coldest season of the year a word from a root ghitn ighiem) which means snow or storm (Greek x^V*"')) so that we have Latin hiems, Greek x^'-H-^^, Old Bulgarian and Zend zima, Sanskrit hhnanta. We know of no root from which winter might be derived, the derivation from wind being excluded on philological grounds. With the word summer it is not much different. It appears in all Germanic languages as the name of the warmer half of the year, but exists in no other Aryan language, notwithstanding that words from the same root, though formed by means of other suffixes and having a similar or the same significance, are found in several of them, such as Sanskrit samd, year ; Zend hama, summer ; Armenian amarn, summer ; Cymric ham, haf, summer.
^Nasse, Uber mittelalterliche Feldgemeinschaft in England, Bonn, 1869, p. 51, Ur- barium of the Monastery of Worcester of the thirteenth century, fol. 103'' : "In hoc manerio sunt 8 virgatae servilis conditionis, quarum quaelibet, si censat, dabit ad quemlibet trium terminorum \2^ pro omni servitio, ut dicunt."
^Kessel, Der selige Gerrich, Stifter der Abtei Gerresheim, Diisseldorf, 1877, p. 187.
^On the dual division of the original Aryan year, compare O. Schrader, Die aelteste Zeitteilung des indogermanischen Volkes, Berlin, Habel, 1878, pp. II ss., where the etymological parallels of hiems and ver are given.
6 - YULE AND CHRISTMAS
When Hildebrand, in the Old-High-German Hildebrandslied, describes his thirty years' wanderings, he says :
"Ih wall6ta sumaro enti wintro sehstic ur lante;"^
Anglo-Saxon legal language having the same phrase. So the laws of Ine provide that the wife of a ceorl who died, if she has a child, should be given, in addition to vi. shillings, "a cow in summer and an ox in winter";^ also that a ceorl's close ought to be fenced winter and summer.^ These two names do not stand alone as the supports of a dual division ; there is a number of other phrases which show that the dual division of the year was extremely familiar to the Germanic mind. To denote the whole course of a year, especially in legal language, the terms were used : iin rise und im love,^ im ruwen und im bloten,^ and bi stro and bi grase.^
Etymology shows that the dual division of the year was of Aryan home growth ; and the very fact that etymology fails as to the tri-partition goes a long way to prove that the tri-partition is of foreign extraction. It certainly is so, and, as far as we can see, it is of Egyptian origin, although it was taken over by the Aryans very early, perhaps even at the time before they divided into self-dependent tribes which evolved idioms of their own. Ewald sums up his investigations as to the division of the Oriental year as follows:^ "People in those countries of Asia and Africa, according to all evidence, had at first three equal seasons. These were fixed in the most ancient Egyptian almanac, and according to that fact in the hieroglyphic writings, the four months of each of these seasons
^Braune, Althochdeutsches Lesebuch, Halle, 1881, p. 77. A St. Gall document of A.D. 858 mentions two brothers, Wintar and Suniar (O. Schrader, Die aelteste Zeitteihmg des itidogermanischen Volkes, Berlin, Habel, 1878, p. 18). On the combats between Winter and Summer, compare Uhland's Volkslieder. Prof. Max Mliller's attempts to show in Greek legends a great number of similar traits seem to me to be rather bold.
"^Ancient Laws and Institutes of England (ed. by Thorpe), London, 1840, I., p. 126, xxxviii. : " cu on sumera. oxan on wintra. "
^ Ibid., p. 126, XV.: "wintres ond sumeres."
^ Grimm, Deutsche RechtsaltertUmer , III., 256, 258. ^ Ibid., III., 249.
'^ Ibid., III., 31, 62, 130, 190, 223; Grotefend, Zeitrechnung des deutschen Mittel- alters, I., 77.
"^ Die Altertiimer des Volkes Israel, 3rd. ed., pp, 455, 456.
The GERMANIC VEAR
7
being very simply counted as the first, the second, etc."^ About the earliest Indian year Grimm remarks t^ "In earliest antiquity the year seems to have been divided into three parts only, the Indians distinguishing either vasanta, spring ; grisckma, summer ; and sarad, raining time ; or, according to the oldest commentator of the Veda: grischma, summer; varscha, raining time ; hema?ita, winter ; and elsewhere six seasons. The Greeks had ca/a, spring; Oepos, summer; xei[jLu>v, winter." The early Aryans, like the Orientals in general, subdivided their three large seasons into six smaller, of the duration of about three-score days each. Ewald, after the passage just quoted, goes on to say:^ "A further step was to divide each of the three seasons into halves and so count six seasons. This habit became law in ancient India, as is shown by Kalidasa's Ritusanhara, but it must also have once been prevalent in Syriac and Arabic countries. The proof of this is the fact that, in the Syriac as well as in the Arabic almanac, frequently two subsequent months are distinguished as the first and second of the same 'tide,' and that tide after which they are called is evidently a season. The distinction between a first and second month according to such seasons has, it is true, only effect when the months, at least in principle, are at the same time calculated by means of the solar year. But, as we know, that was the case pretty early."^
These very same three-score-day tides are found among the Germanics, Eastern and Western. But the strange fact that no satisfactory Germanic or even Aryan etymology can be given for the oldest names of Germanic three-score-day tides, Jiuleis (Gothic), Lida, Hlyda (Anglo-Saxon), and per- haps Rheda, Hreda (Anglo-Saxon), and Hornung, Horowunc (German), seems
^ Lepsius, Chronologic der Aegypter, I., p. 134.
^ Geschichte der deutschen Sprache, 1848, I., p. 72. ^ Ibid., p. 456.
■•O. Schrader is the first to avoid the presupposition that the early Aryans based their partition of the year on a knowledge of the stars. He did so with full consciousness (O. Schrader, Die aelteste Zeitteilung des indogermanischen Volkes, Berlin, Ilabel, 1878, pp. 24), 32, and expressly says that the three roots used for denoting sun in the Aryan languages contain no element referring in any way to time or partition of time, whilst as regards the moon he attributes to her merely a secondary rank in that respect, and remarks that the origin of the months dates from no earlier period than the time when the Aryan tribe had split into several peoples.
8 YULE AND CHRISTMAS
to point to the probability that these names, like the institutions they denote, have their origin beyond the world of the Aryan family of languages and nations, and were borrowed from Egyptian and Syriac, or some other Oriental language, together with the six three-score-day tides which formed the course of a year. This probability is enhanced by the fact that Gothic Jiuleis in the forms lAaios, tovAaios, lovXirjos, and IovXlo<s is found to denote the time from Dec. 22 to Jan. 23 in old Cyprus,^ which can scarcely be ascribed to chance. It is Jacob Grimm's merit to have gathered a number of important facts which show the same habit to have prevailed among several Aryan tribes, including the Germanics.
"Stress is to be laid," says he,^ "on the connection of two (or even three) subsequent months through the same name, which connection seems to be a relic of an original partition of the whole of the year into six (or four) 2 parts. Thus, among the Anglo-Saxons, there appeared a double /td/ia for the pair June-July, which elsewhere also appears bound together as brachot-houwot, or the two resatlle-mois, and a double geola. Thus, in Middle-High-German, there appears a double oiigest, a double wintermonat, (a threefold herbstmonat). January and February are even much later singled out as the large and the small horn; nay, here and there we find the second of two months presented as the wife of the first, and a sporkel followed by a sporkelsin and an ougest by an ogstin. Likewise we find among the Slavs a small and a large traven, a small and a large serpan, where the small precedes the large one, whilst our small horning succeeds the large horn. (The Liineburg Wenden also made a first wintermonat, September, precede the other, which was December.) According to Slavic order, however, the small cerwen preceded the large lerwenec. Something similar is found in the Celtic midu and ?nichrundu for November and December; ephan, summer, and gorephan, main summer, for June and July.
^ K. Fr. Hermann, Uber Griechische Monatskunde, Gottingen, 1844, p. 64.
^ Geschichte der deutschen Sprache, 1848, I., iios.
" I put in round brackets what seems to me to be wrong in Grimm's argument. When the meaning of the common heading of two subsequent months was forgotten, and Roman quarters of years had become popular, a third month was sometimes added under the same name, a usage to which the intercalary month may have led.
THE GERMANIC YEAR 9
Even the gipsies, whose month-names are given by Pott (I. 116), style June and July by the cognate names nutibe and nunutibe; and in Albanese yoo-Ti and yoo-To^teo-re for August and November we have the same thing. This coupling is in my eyes a testimony of great age. (The Attic calendar in the leap year added another IlocreiSewv after the first, as the Jews did after their adar, a veadar or other adar.) The Arabic lunar year still shows its months connected in six regular pairs : rebi el avvel and rebi el accher, dscheviadi el avvel and dschemadi el accher, dsulkade and dsulhedsche. The Syriac year shows a theschrin I. and II. and a khamm I. and II.; whilst in the Persian and Jewish calendar this coupling has been lost. But it is quite apparent in the division of the Indian year into six parts, each of which embraces two months, most of which have cognate names, viz. : vasanta, spring, contains the months madhu, mead or honey, and mddhava, honey sweet; grischma, summer, contains the months shukra, the light one, and s/iukhi, the shining one ; varscha, the raining tide, contains the months nabhas, cloud (Latin nubes, Slavic nebo, cloudy sky), and nabhasja, the cloudy one; sarad, sultry tide, contains the months ischa and iirg/ia, the nourishing one; hemanta, winter, contains the months sahas, strength, and sahasja, the strong one ; stsira, dew tide, contains the months iapas, warmth, and tapasja, the warm one. The relation of the names taj>as and tapasja, nabhas and nabhasia, sahas and sahasja, Jtiadhu and mddhava is analogous to sporkel and sporkelsin, ongest and ougsiin ; gosti and gostobieste, cerwen and cerwenec, and the Sanskrit names given here seem to be more popular than the learned ones, which were fixed for the aditjas; and through the division of the Indian year into six seasons, the division of the Germanic year into three seasons, which immediately proceeds from it, is justified in a way that must be welcome to us." Further on ^ he says, "A connection between our month-names and the six Indian seasons, and the coupling always of two subsequent months, which proceeds from them, must be acknowledged. "2
^ Geschichte der deutschen Sprache, p. 113.
^ In Mahabharata the six Indian seasons vasanta, grtsma, varsa, farad, hemania, and fi^ira are represented as six men who play at golden and silver dice (O. Schrader, Die aelleste Zeiiteilung des indogerntanischen Volkes, Berlin, Habel, 1878, p. 22).
lo YULE AND CHRISTMAS
Grimm indicates in a general way the facts that point to an early six- partition of the Germanic year. But since his day so much material bearing upon the point has accumulated, that it is necessary to enumerate the most important items of it. On the Nether-Rhine the division of the year into six tides or periods of sixty days each Avas known till so late as the fourteenth century, although then it was thought to be antiquated as compared with the new Romano-Christian way of deter- mining seasons. As the starting-point people then, as of old, took one of the three ends of the seasons, July 12, counting from then to September 17, November 11, January 13, March 17, and May 12, by ancient Ger- manic three-scoreday tides or half seasons. Neither eight weeks nor nine weeks exactly covering these tides, eight weeks and nine weeks were alternately taken. On November n, therefore, winter began, on March 17 early summer, and on July 12 later summer.^
It is rather difficult to say what were the names of these GermarRic three-score-day tides, although in German legal and literary documents there occur quite numerous denominations which clearly cover a longer time than a month, and yet neither amount to three nor to four months.
Such are, e.g., in der brache, in der zwibrache, in der herbstsat^ in der erne, im houwet, im hanffluchet^^ ze afterhahne und houwe,'^ in der bonenar/ie.° Others are im brdchet, im wimmot, in der sat, in dem sniie, laubbrost, and laubrtse,^ haberschnitt, and habererndte, covering August and September, and
^ " Urbarium of the Monastery of St. Victor, Xanten," in the State- Archive, Diisseldorf, under "Stift Xanten," R, No. 8=*, leaf 8=^. The passage was communicated to me, like all the unprinted material referring to the Rhine-country and Tirol, by Dr. Armin Tille of Bonn. It runs: "Item notandum, quod secundum antiquum modum computandi servicium potest poni per certos terminos infra dictos, scilicet a festo Margarete (ubi annus incipit) usque Lamberti sunt 9 ebdomade, item a Lamberti usque Martini sunt octo ebdomade, item a Martini usque ad festum baculi, quod est octava epiphanie, et sunt 9 ebdomade. Item a festo baculi usque Gertrudis sunt 9 ebdomade, item a festo Gertrudis usque ad festum Pancracii sunt 8 ebdomade, item a festo Pancracii usque Margarete sunt novem ebdomade et faciunt simul unum annum, scilicet 52 ebdomadas."
^ Grimm, Deutsche RechtsalterlUmer, III., 546.
^Ibid., I., 419. '^Ibid., I., 673, 679.
'Neocorus, II., 75, 426; Weinhold, Deutsche Jahrteilung, 1862, p. 13.
'Weinhold, Die deutschen Monatnamen, Halle, 1869, p. 2.
THE GERMANIC YEAR It
sometimes even including the time from July 25 to August i.^ To these lenz (lent, A.S. lengten) and herbst (harvest) in their, medieval senses are to be added, the latter being on German ground frequently replaced by augst (August), which, however, covers a longer time than August, so that July 25, St. James's day, can be z2iS\tdi Jacob stag im augstP- It was out of such words that the Germanic month-names were formed in the second half-millennium of our era, after the Roman calendar had become popular among the Germanics. So Professor Weinhold^ seems to be of opinion that in der erne is older than the month-name erntmanot; that im bracket and im houwet dite older than brdchmonat ; and that hcumat and im ivimmot are older than windumemanot : and he expressly states that, according to his belief, in der sat and in dem snite have given origin to sdtmdn and schnitmonat, and that laubbrost and laubrtse under our eyes become, by being taken in a narrower sense, something like month-names.* Beside these words English expres- sions like fall, backend, '•'• hotil d winter'^ are to be placed, and perhaps, also, two other words which later were used to correspond to Latin ver and autumnus. For ver the Western Germanics took a root, lang, perhaps connected with long, forming out of it a term for the time when the days grow longer (Old-High-German langiz, lenzo, lenzin ; Middle-High- German lenze; Dutch lente; Anglo-Saxon lengten, lencten; English lent), which, however, cannot be traced beyond the Western Germanic, not to mention the common Germanic.^ This makes it rather likely that it was not the name of an old three-score-day tide, but was formed new. The term adopted to correspond to autumnus is also confined to Western Ger- manic, although its root is common Aryan property. It is Old-High-
^Grotefend, Die Zeitrechnung des deutschen Mittelalters, 1891, I., 79.
'^Grotefend, Zeitrechnung, I., 87.
^ Die deutschen Monainainen, Halle, 1869, pp. I, 2.
^If Neocorus, II., 315, explains in howman edder in der howame (Weinhold, Ibid., p. 13), it follows that he regarded the term hmvarne as still more popular than the newly formed word howman.
^ Since the fifteenth century it has been in German supplanted by the term Friihling (from friih, Gothic frd, * early ; Greek Trput), and in English by spring (cp. I Sam. ix. 26, "about the spring of the day"; Shakespeare, 2 Henry IV. iv. 4. 35, "since the middle summer's spring").
li VULE AND CHRISTMAS
German herbist, Middle-High-German herbest, Dutch herfst, Anglo-Saxon hcerfest, English harvest, and belongs to Latin carpei'e, to pluck, and Greek Kapiro^, fruit. ^
Yet more important than these rather vague terms are several others which can be proved to have exactly covered a Germanic tide of three- score days. They are the more striking since, in two cases, it is simply Roman month-names which are used for denoting a tide of two months, so that two subsequent Roman months among the Germanics are fre- quently called by the same name, the first being called the former, and the second the latter month of that designation ; whilst some Germanic names of the same kind are of very great age.
The Scandinavian summer of six months is divided into three tides called Vaarmoaner, Sumarmoaner, and Haustmoaner j^ herbst as denoting the two months September and October is very common in the Middle Ages, so that the former month is called der erste herbst, and the latter der andere herbst?
In a Gothic calendariiim of the sixth century* November, or Naubaimbair,
■* Scandinavian haust or host is probably to be derived from August, which in the Middle Ages appeared as aust. In English the Germanic word hai-vest in the later sense of a season was completely superseded by Romance autumn.
'^Weinhold, Altnordisches Leben, Berlin, 1856, pp. 371-383.
"* Grotefend, Zeitrechnung des deutschen Mittelalters, I., 84; Weinhold, Deutsche Monat- namen, pp. 15, 42. There appears even a dritt herbst for November. The naming of three subsequent months as first, second, and third augst or hej-bst admits of several explana- tions. The Roman quarter of a year having taken the place of the old third, it was but natural that all the three months forming it should have received a common name ; there is no doubt that in this way herbst, which simply meant harvest, advanced to the meaning of the season of autumn. This is the more likely, since the tripling of the month-name is found just in that season, which, according to Tacitus, among the Germanics had no name. Beda also has such a tripling in the case of the Lida month, though for another purpose, viz., for forming an intercalary month for the leap year. His June, July, and third Lida covering to a large extent the same ground as the July, August, and Septem- ber, called the three herbste or augste, it may seem probable that the leap year had also to do with the origination of the series of three months bearing the same name. That it was August which was doubled may be inferred from Northic tvtmdnadhr, double-month, which is the name for August, and has not been understood by Professor Weinhold.
*Moritz Heyne, Ulfilas, Paderborn und MUnster, 1885, p, 226.
THE GERMANIC YEAR
13
is called fruma Jiuleis, which presupposes that December was called *aftuma liuleis ; in Beda's list of Germanic month-names ^ it is stated that the Anglo-Saxons called December and January together Gtu/i, for which, later, the terms cerra Geola and ceftera Geola were used ; in the same place Beda states that the Anglo-Saxons called June and July oerra Lida and ceftera Lida respectively; in Middle Germany up to this day January is called der grosse Horn, February der kleine Horn. Both names together occur in Christian Wolf's Mathematical Dictionary, Leipzig, 1742, which borrowed all kinds of things from the dialects spoken in Saxony ; but Hornung {i.e., small Horn or son of Horn) is found among the list ot German month-names composed by Charlemagne, and brought down to us by his biographer Eginhart.^ It being the only name in the list which is not a compound of mdnoth, it is bound to be of ancient German origin. The forms occurring elsewhere are Hornung, Horning, homer and horn.^
As a rule, der erst herbst is September,* and der under herbst October.^ But November also appears as der under herbst,^ though more frequently as der drit herbst^ September also being called Uberherbst.^ It shows a state of things a little further advanced when, as is frequently the case, der erste and der undere herbst are replaced by der erste and der andere herbstmonat f likewise there are numerous examples for November
^ De Tempormn Ratione, chap. xv. ^ Vita Caroli Magni, chap. xxix.
^Grotefend, Zeitrechnung des deutschen Miltelalters und der Neuzeit, 1891, I., 86. I should be incHned to see in Horn the name of an old German three-score-day tide, just as in Yule and Lida. Compare on Horn, however, Weinhold, Deutsche Monatnanien, p. 45.
* Diefenbach, iVovujti Glossariwn, 32; Codex Germanicus Monacensis, 93, 398, 700, 730; Grazer Kalender ; " der erst heribst," Codex Germanicus Monacensis, 349.
^ Klingenberger Kronik, 343; Diefenbach, Novum Glossaritim, 32; Codex Germ. Monac, 93> 398, 480, 700, 730, 771 ; Grazer Kalender; Huber, "der ander herbst," Giess. MS., 978 ; Weinhold, Die deutschen A/o}iatnat?ien, pp. 41, 42.
" Codex Germ. Monac, 32.
^Diefenbach, Novum Glossariwn, 32; Codex Germ. Monac, 349, 730.
^ Tegernseer Kalender ; Weinhold, Ibid. , p. 42.
* Grimm, Geschichte der deutschen Sprache, I., p. 85; Zellweger, No. 191, a. 1407; Weinhold, Deutsche Monatnamen, p. 15, note, and pp. 42, 43, where a long list of cases is given. November is sometimes called der dritt herbistmanot {Ibid., p. 43), and December der vierd herbistmonad ox letst herbistmotieth (Grotefend, Zeitrechnung, I., 84), a fact which leads us beyond Roman quarters of years into Germanic thirds of years. Unserfrauen
14
YULE AND CHRISTMAS
being called der erst winter, and December being called der ander winter (and even January being named manot des hindrosten winters) ;^ so November is also called der erst winter nianeid,^ and December d&r ander winter nianeid^ or der lest wintermond.'^ On Bavarian ground, a Tegernsee calendar^ calls March and April das erst ackermonat and das ander ackermonat ;^ to these cases two others are to be added, in which Roman month-names are used for the same purposes. There is quite an abundance of instances in which May is called der erst may, and June der ander may •,'^ the same holds good for der erst augst, meaning August, and der ander augst, meaning September, so that even the term occurs : in den tzweyen augsten.^ In the Diocese
dulttag in dem efsten herbsltnanode, Sept. 8, A.D. 1 290, Pilgram ; Grotefend, Zeitrechnung, I., 68.
^ See Weinhold, Detitsche Monatnaftien, p. 61. "^ Ibid., p. 62. ^ Ibid., p. 62.
* Grotefend, Zeitrechnung, I., 208, where also an instance is given of February being called der letzte wintermonat , a.d. 1536, Ulm, which again points to four winter months or Germanic thirds of years.
* Pfeiffer, Germania, IX, , 192 f ; Weinhold, Deutsche Monatnamen, 14. ^Grotefend, Zeitrechnung, I., i.
''Weinhold, Deutsche Monatnamen, pp. 13, 50, where a whole group of old Bavarian almanacs is mentioned which has this peculiarity ; Grimm, Geschichte der deutschen Sprache, 1848, I., 84, according to a Cassel manuscript of A.D. 1445. About some Alemannic and Swabian almanacs, comp. Weinhold, Ibid., p. 15.
^ Muglen bei Kovachich, p. 4 ; Grimm, Geschichte der deutschen Sprache, 1848, I., p. 85. In some Bavarian almanacs August and September are called der erste and der andere augst (Diefenbach, Novum Glossarium Latino-germanicutn, Francofurti, 1857, 34, and Tegernseer Fischbuchlein) ; so it is on Alemannic and Swabian ground (Weinhold, Deutsche Monatnamen, p. 15 ; Codex Germanicus Monacensis, 32 ; Diefenbach, Novum Glossarium, 34 ; Mone, Anzeiger, VIII., 496) ; der ander ougst is September (Grimm, Geschichte der deutschen Sprache, p. 85). Among the German communities of Valsugan and on the hills between Brenta and Drau, August is called erster Aux, and September dnderts Aux, a form in which it also appears in some Roman documents of Rhaetia (Hormayr, Geschichte der gefiirsteten Grafschaft Tirol, Tubingen, 1806, Part I., Section i, p. I41). Der erste august means August in Tegernseer Fischbuch ; der erst awgst, Giessen MS. , 978 ; der erst awst, Codex Germanicus Monacensis, 32 ; whilst der ander augst may be August (Codex Germanicus Monacensis, 93, 398, 7CX3, 848, 3384; Giessen MS. 978; Gmund's Kalender; Grazer Kalender; Ruber's /Calender; der andere auste, Diefenbach, Novum Glossarium, 4), or September (Megenberg, Diefenbach, Novum Glossarium, 34 ; Tegernseer Fischbuch {der ander august)). In the xiii. comuni there are even three Agester, meaning August, September, October (Cimbr. Worterbuch, 107; Weinhold, Die deutschen Monatnamen, p. 32).
THE GERMANIC YEAR
15
of Constanz August is called ougst, and September Haberougst} or August is called erster aux, and September under aux,'^ or August is called augest, and September Augstin, oegsten, auwestin, i.e., the small augest.^ Sometimes July is called der erste augst, and August der under uugst.^ Though Augstine and Aygsien appear a few times as meaning August,^ on the whole Ougstine, i.e., small August, means September f nay, there even appears for September the compound Herbistouwistinne'^ and the word Huberougst.^ So Konrad von Dankrotsheim in his Numenbuch ^ names August and September ougst and oegstin. On the other hand, augstmdnd, uuistmuent, uustmuent, owest- mun mean August exclusively.^*'
There is nothing whatever in the Roman calendar which can be said to have been suggestive of that strange custom, so that we have good reason for claiming it as a relic of a pre-Roman Germanic usage. If it was able to influence the Roman calendar so far as to force upon it the three-score-day tide, it must needs have been most deeply rooted and firmly estabHshed among the Germanic tribes in East and West, North and South. Not only is the six-fold division of the Germanic year a most important fact in itself, but it also furnishes us with the means of reconciling the seeming contradiction, according to which the Germanics at the same time had a dual division and a tri-partition of the year. The units of which
^ Ehinger Spitalbtuh, Germanic Museum, Niirnberg, No. 7008.
^ Sette communi ; Schtneller-Frommann, Bayrisches Wdrterbuch, 54 ; Grotefend, Zeitrechnung, I., 14.
^ Schmeller-Frommann, Bayrisches Worterbtich, 54 (1453, Baselland) ; Grotefend, Zeitrechnung, I., 14. Grimm explained the term wrongly as the wife of August.
* Grotefend, Zeitrechnung, I., 14. ^ Codex Germanicus Monacensis, 771.
^ Codex Germanicus Monacetisis, 558 (Schmeller, I^, 54), Ougstin, Dankrotsheim ; Augstin, Dasypodius 488d (1537); Ougsten, Diefenbach, Novum Glossarium, 40; Oegstin, Dankrotsheim; Ouwestin, Kbditz, Leben des heiligen Ludwig, Leipzig, 1851, 40, 61 ; Owestin, Hermann von Fritslar (Myst. I., 195) ; Weinhold, Deutsche Monatnamen, p. 32 ; Grotefend, Zeitrechnung, I., 14,
^ Koditz, Leben des heiligen Ludwig, Leipzig, 187 1, 66.
** Ehinger Spitalbuch ; Weinhold, Deutsche Monatnamen, p. 39.
* Weinhold, Ibid., p. 16; Strobel, Beitrdge zur deutschen Literattir und Literatur- geschichte, Paris, Strassburg, 1827, p. 109 ss.
^"Grotefend, Zeitrechnung des Mittelalters, Hannover, 1891, L, 14.
1 6 YULE AND CHRISTMAS'
their year consisted were sixths, and it is apparent that of these tides either two could each time be grouped together to form thirds, or thrSe could be grouped together each time to form halves. At the same time the simple fact of sixths being the units constituent of the Germanic year excludes any quartering of the year, since a quarter would consist of one- sixth and a half, and would thus most seriously interfere with the unity of the three-score-day tide. If Professor Weinhold^ says the dual division of the Germanic year was dislodged by a tri-partition, he is entirely in error; for so little can be said of a dislodgment of one mode by the other that for a long prehistoric period both existed peacefully alongside each other, and appear thus at the dawn of history. The Oriental tri-partition of the year would probably not have so deeply rooted itself in the Germanic mind had it not been supported by the economic and climatic conditions of the country they emigrated to. There the most decided season, the winter, fills, on an average, a period of exactly four months, which naturally leads to a division of the rest of the year into two equal parts of four months each. And the economic year was no less naturally divided into three parts — the rest of the plough, the cultivation and reaping of the grass, and the harvest.^ There is no reason to take refuge in speculations about symbols and the religious opinions of the early Germanics to explain their division of the year. Economic conditions have at all times weighed much heavier than fancies. The centre of animal activity, as well as of the strivings, hopes, and dreams of men, was in pre-Roman Germanic times, as it is now, to win food and to get on in life — in early times by hunting and keeping cattle, later on by cultivation of meadows, and finally by agri- culture in addition. Then as now the endless generation of human beings and the endless competition for the means of subsistence among them, which two factors have at all times determined the fates of families, tribes, nations, and races, pressed upon individuals, and compelled them to work by leaving them, if unwilling to do so, the alternative of perishing.
^ Deutsche Jahrteilung, 1862, p. 7. ^Weinhold, Deutsche [ahrteihmg, 1862, p. 7.
CHAPTER II.
THE BEGINNING OF THE ANGLO-GERMAN YEAR.
Both Caesar and Tacitus tell us that the Germans did not count by days as the Romans did, but by nights, reckoning the vigilia or eve as part of the following day.^ This habit lived on unbroken through the Middle Ages, and is still living among us, so that we count by fortnights instead of by fourteen days, and speak of the Twelve-nights of Christ- mas-time. It was taken over by the early Church, as we know by so good an authority as the Venerable Beda, as regards festivals at least.^ In the same way the Germanics reckoned by winters instead of by summers, counting the winter and the summer which followed it as- one year, a custom which, however, is not exclusively Germanic.^ The Saxon Chronicle
^ Caesar, Belluvi Gallicum, Book VI., chap, xviii. : " Spatia omnis temporis non numero dierum, sed noctium finiunt ; dies natales et mensium et annorum initia observant, ut noctem dies subsequatur. " Tacitus, Germania, chap. xi. : " Coeunt nisi quid fortuitum et subitum incidit, certis diebus, cum aut inchoatur luna aut impletur ; nam agendis rebus hoc auspicatissimum initium credunt. Nee dierum numerum, ut nos, sed noctium computant ; sic constituunt, sic condicunt ; nox ducere diem videtur. " Later instances are given from law literature by Weinhold, Deutsche Jahrieilung, notes 2, 3, II, 12, and 13; from documents by Grotefend, Zeilrechnung des deutschen Mittelaliers, Hannover, 1 89 1, I., p. 131, and 1898, II., 202, under Nacht.
^ De Temporum Ratione, chap, v.: " Merito autem quaeritur, quare populus Israel, qui diei ordinem iuxta Moysi traditionem a mane semper usque ad mane servabat, festa tamen omnia sua, sicut et nos hodie facimus, vespere incipiens, vespere consummarit dicente legislatore : A vespera usque ad vesperam celebrabitis sabbata vestra."
^Manilius, Astronotnicon, "per quinquaginta brumas"; and Martialis, "ante brumas triginta." On the parallels between night and winter and between day and summer, see O. Schx&dex, Die aelteste Zeitteilungdes indogermaniscken Volkes,'Be:x\m, Habel, 1878, pp. 12, 44SS.
B
l8 YULE AND CHRISTMAS
abounds with examples of that usage.^ Though it is generally the term winter which is used thus, it is by no means exclusively so, other terms, like louprise in the Alemannic dialects, being employed in the same sense.^ Thus there can be no doubt that the Germanic year began with the begin- ning of winter, and not in the middle of it, as did the Roman year with its dogmatic and unpractical way of dividing time. But when exactly was the Germanic New Year? On an average, in Germany actual winter sets in about the middle of November, when it ceases to be possible to leave cattle, swine, sheep, and horses on the pasture grounds to seek their own food ; when it begins to freeze ; and when snowfalls become very frequent. There is no doubt that for purely nomadic cattle-keeping tribes, such as at the dawn of history the Germanics certainly were, this is the term which compels them to change all their summer habits, and therefore marks the beginning of a new season in the most incisive way. It may, therefore, be regarded as certain that the Germanic winter did not begin later than at mid-November. But in order to determine whether it did not, perhaps, begin earlier, we need other means than mere economic speculation. We saw above that the Germanics in prehistoric times took over the Oriental year, which was divided into six three-score-day tides. Now the oldest such tide we know of among Germanics was the liuleis tide among the Goths of the sixth century.^ It exactly covered the Roman months, November and December, November being called fruma liuleis. It is more than unlikely that the beginning of the year should have interfered with any
^The Parker MS. of the Saxon Chronicle (ed. by Earle in " Two of the Saxon Chronicles Parallel" Oxford, 1865, p. 2) begins: "thy geare the wses agan fram Cristes acennesse cccc. wintra and xciiii. uuintra," and in the second paragraph gear and winter are used as synonyms ("and he hsefde thset rice xvi. gear . . . and heold xvii. winter . . . heold vii. gear . . . ricsode xvii. gear . . . riscode xxxi. wintra . . . heold xxxi. wintra," Ibid., p. 2). The oldest part of the version belongs to the second half of the ninth century (R. Wlilker, Grundriss zur Geschichte der angelsdchsischen Literatur, Leipzig, 1885, § 509).
^Weinhold, Deutsche Jahrteilung, pp. 12 and 19; just as the Bavarians counted after autumns {Lex Baiuvariorurn, VIII., 19, 4; Weinhold, Ibid.).
^The periods of three scores of days I call tides, the unities of two tides I call Germanic seasons, and the unities of three tides I call half-years.
THE BEGINNING OF THE ANGLO-GERMAN YEAR
19
such three-score-day tide ; therefore the conclusion will be allowed that it rather began with the beginning of one. Thus we should have to assume either November i or September i to have been the beginning of the Germanic year. But September bearing in Germany entirely the character of a summer month, — nay, towards the end of August, the heat frequently being the greatest in the whole course of the year, — the beginning of the Germanic year at the beginning of September is practically out of the question, so that only November i remains as a possible beginning. Among the Goths of the sixth century fruma liuleis and November were apparently absolutely identical, as appears from St. Andrew's day — which is marked down in that calendar as Frtima lUdeis 31 — and from some other saints' days. All we are allowed to conclude from the fact, how- ever, is that the Goths of the sixth century had taken over the Roman calendar, naming the Roman months by the home-made names of those Germanic tides which approximately covered them. It by no means follows from this fact that each of the Germanic pre-Roman three-score-day tides exactly covered two Roman months. It would be an astounding incident indeed if that had been the case, and only an extremely rare chance could account for it. Knowing that the Indian months began shortly after the middle of the Roman months,^ we have every reason to assume something similar for the Germanic three-score-day tides. This assumption is sup- ported by a great number of singular facts. Had each of the six ancient Germanic tides not exactly, but fairly covered two subsequent Roman months, we should have to expect that the same would be the case with the German month-names which sprung up for the denomination of the Roman months. But the contrary is the case. To quote a witness be- yond suspicion, I cite Professor Weinhold,^ who says : " We find a wavering of the names between several months : ackermonat wavers between March and April ; hartmonat between November, December, and January ; lasemdnt means December and January ; hornung means January and February ; hundeman is found to be applied to June, July, and August j rosenmant to
^ Grimm, Geschichte der deutschen Sprache, 1848, I., p. 75. "^ Die deutschen Monainamen, Halle, 1869, p. 2.
20 YULE AND CHRISTMAS
June and July; scttmant to September and October; slachtmcin to October, November, and December ; and sommermonat to June and July : the names containing the constituent vol (ful) occur for September, November, Decem- ber, January, and February, and wolfmonat appears denoting November, December, and January" — a list which could easily be increased. It is now generally admitted that the Germanic month-names are a very late product, and that they were merely formed for the purpose of replacing the Latin names. If, then, the ancient Germanic three-score-day tides extended from about the middle of one Roman month to the middle of the second next, it was bound to happen that, at the time when the Roman month-names were taken over, they were applied to the interval between the middles of two consecutive Roman months, which means neither more nor less than that each month-name could be used for two months ; so during the period of transition it could scarcely be avoided that e.g. the term November was at some places used for the time from October 15 to November 15; whilst on others it was, with equal right, made to cover the time from November 15 to December 1 5. When, later, the Latin name was replaced by a German word, the characteristic held good. In consequence it could not fail to happen, even in neighbouring places, that, of two consecutive Roman months, some- times the first and sometimes the second was called by the one Latin name. Finding, as we do, that the Goths called liuleis the time from November i to December 31, and the Anglo-Saxons called Geola the time from December i to January 31, we can scarcely help assuming that liuleis originally covered a period from about November 15 to January 15, and that, at the taking over of the Roman calendar, among the Goths that name was shifted a fortnight back, and among the Anglo-Saxons a fortnight forward, so as to create an incongruence of a whole Roman month. This argument must needs lead us to the conclusion that the Germanic liuleis tide extended originally from about mid-November to mid-January ; for had it extended from mid-October to mid-December, we should have to expect a wavering of the Yule tide between October-November and November-December, and not between November-December and December-January. All this points to the con- clusion that a Germanic three-score-day tide began originally about the middle of November, and that the beginning of it was at the same time
THE BEGINNING OF THE ANGLO-GERMAN YEAR 21
the beginning of the Germanic year. This result is supported by the Rhenish Urbary of the fourteenth century from the Monastery of St. Victor, Xanten.^ in which the terms of the six three-score-day tides are July 12, Sept. 17, Nov. II, Jan. 13, March 17, and May 12, Martinmas being marked out as a term as close to the middle of November as possibly can be expected.
The idea of the Germanic year beginning about Martinmas is not new. Fin Magnusen^ remarked, about a century ago, that the Germanics began their year about the Advent tide, which for a long time began with the Sunday after Martinmas. Even Weinhold admits that Martinmas coincides with the actual beginning ol winter,^ in which character it is clearly marked by the popular rimes :
"Sanct Martin
Feuer im Kamin " * and
" Sanct Marten Miss
Is der Winter wiss";^
whilst, on the other hand, he maintains that the Germans began their year about the same time as the Romans began a new quarter of the year, i.e., on September 29, St. Michael's Day. It is true he does not even attempt to prove that assertion historically, but with a few vague remarks, which can scarcely be taken seriously, jumps over the whole point which ought to have been the centre of his investigation. It never occurs to him that the Goths regarded November and December as their liuleis tide, and that, if their year was not begun at the ist of November, it was bound to commence on the beginning of September, when another three-score-day tide took its inception. Only a man who has never in his life left his study for fresh air can maintain that winter began at the close of
^ Staatsarchiv, Diisseldorf, under " Stift Xanten," R, No. 8*\ leaf 8^.
^Specimen Caletidarii gentilis, p. 1018, according to Pfannenschmid, Germanische Erniefeste, p. 512 : " Suspicor vulgarem inter veteres Germanos anni adventum posterius inter christianos certo modo mutatum fuisse in adventum domini sive initium anni ecclesi- astic!."
^ Uber die deutsche Jahrteilung, Kiel, 1862, p. 5.
^Graesse, Des deutschen Lattdmanns Practica, Dresden, 1858, p. 178.
'^Ibid.y p. 178.
2a YULE AND CHRISTMAS
September ! i Did Professor Weinhold realize that spring began then at the end of January, when Germany as a rule is ice-bound for another month and a half — two to three months before cattle are able to pasture on the meadows? He had not even the courage to follow out the conse- quences ; but makes spring begin " in March," and summer at the summer solstice ! To do him no injustice in any respect, I shall assume that the phrase " in March " (which implies thirty-one days to select from) is meant to mean the middle of March, Then we have a winter extending from the end of September to the middle of March; a spring extending from the middle of March to June 24; and a summer extending from June 24 to the end of September (which seems to mean September 29), i.e., a winter of more than five months and a half! a spring of three months and nine days ! and a summer of three months and five days ! — a calculation which certainly does all honour to the arithmetical attainments of our Germanic ancestors and their distinct sense of the equality of three thirds ! Would one take the phrase "in March" as "in the end of March," the time when storks and swallows return in flocks and the grass begins to grow green again, we should have a winter of full six months, a spring of not quite three months, and a summer of a little more than three months.
But perhaps one must not draw the consequences from these ill-considered assumptions. The truth is, that a beginning of the Germanic winter about the end of September is absolutely untenable, that it really took place about the middle of November, while the end of September did not become of importance as a dividing-point of the year until the introduc- tion of the Roman quartering of the year, under the reign of which a new quarter began on October i, and was fixed by the Church on September 29, i.e., on Michaelmas.
Having, at last, arrived at the starting-point, it will be necessary to cast a glance over the whole of the Germanic year, and to draw once and for all the theoretical conclusions from that result. Counting by original Germanic half-years, summers and winters, we have to fix the other junction-point of the two at mid-May, a term dear to all who have the
^Graesse, Des deutschen Landmanns Practica, Dresden, 1858, p. 718.
THE BEGINNING OF THE ANGLO-GERMAN YEAR 23
happiness of receiving house rents in Scotland. Counting with Oriental- made thirds of years, or Germanic seasons, we have to close the winter and begin the early summer at mid-March, and to close the early summer and begin the late summer at mid-July. If this really be the old Germanic division of the year, it is bound to be preserved in all kinds of recollections and institutions : above all, in legal institutions, in popular tradition, in folk- belief and rustic custom, in festivals and bonfires, and, last but not least, in ecclesiastical habits which, as far as they were created after the fifth century of our era, reflect an enormous amount of Germanic tradition and thought.
CHAPTER III.
THE FEAST OF MARTINMAS.
The term at which Roman legions in Gaul and Germany withdrew into winter quarters varied a little, although not considerably. As a rule it seems to have been before the middle of October, when the rainy season begins in those countries. It was the frequent rain which prevented any continuation of warfare, not the cold. No Roman general seems to have been bold enough to try to extend warlike operations till frost set in. When Caesar, for once, tried to keep his legions engaged in war beyond the usual term, he was compelled to retreat, as his soldiers could no longer sleep in the open field because of the rain.^ On the other hand, he did not like to retire into winter quarters too early; and when he had to do so, because no more work was to be done, he expressly mentioned it.2 Now, A.D. 14, Germanicus was fighting some German tribes. The autumn came, and he withdrew into winter quarters. The winter was imminent, but had not yet set in.^ The fifth and twenty-first legions were in winter quarters at the sixtieth stone, which place was called Castra
^Caesar, Bellum Gallicum, Lib. III., chap, xxix.: " Incredibili celeritate magno spatio paucis diebus confecto, cum iam pecus atque extrema impedimenta ab nostris tenerentur, ipsi densiores silvas peterent, eiusmodi sunt tempestates consecutae, uti opus necessario intermitteretur et continuatione imbrium diutius sub pellibus milites contineri non possent."
^Bellum Gallicutn, Lib. I., chap. liv. : "Caesar, una aestate duobus maximis bellis confectis, maturius paullo, quam tempus anni postulabat, in hiberna in Sequanos exercitum deduxit."
^Tacitus, Annales, Lib. I., chap. xliv. : "Ob imminentem . . . hiemem."
THE FEAST OF MARTINMAS
25
Vetera,^ when they mutinied. In order to turn their minds to something else, Germanicus undertook — autumn being far advanced, but apparently no snow having fallen yet and there being no frost — another small military expedition. He crossed the frontier, and invaded the enemy's territory. There were two ways to take. He chose the longer one, and hurried on; for scouts had informed him that a certain night was a festival night for the Germans, and gave occasion for gay banqueting. In a beautiful clear night he reached the village of the Marsi, and surprised them completely during their feasting.^ This can only refer to a festival held at the begin- ning of winter, before snow and frost had set in, which, in the German climate, can hardly have been at any other time than in the first half of November. In the second half Germanicus would have had to encounter the most serious difficulties as to the weather, whilst to assume that the festival had been in October would not leave sufficient time for the with drawing of the legions into winter quarters, the mutiny, and the warlike expedition after it. So we have a right to say that a German festival was held in the first half of November as far back as a.d. 14, while the date of the report of it is certainly to be set down before a.d. 117, in which year Tacitus died. It is the oldest Germanic festival on historical record ; and although half a millenium elapsed before it was mentioned again, we have no reason to doubt its existence. And it is mentioned again before the Christian Church had got proper hold of all the Western Germanic tribes, towards the end of the sixth century, when St. Martin had become a great saint of the Church, and November 11, the date of his death, had been made his day of commemoration.
■ 1 Tacitus, Annales, Lib. L, chap, xlv.: "Quintae et unaetvicesimae legionum sexagesimum apud lapidem (loco Vetera nomen est) hibernantium."
^Ibid., chap. 1. : " Delecta longiore via cetera accelerantur : etenim attulerant explora- tores festam earn Germanis noctem ac solemnibus epulis ludicram. Caecina cum expeditis cohortibus praeire et obstantia silvarum amoliri iubetur : legiones modico intervallo sequuntur. luvit nox sideribus illustris, ventumque ad vicos Marsorum, et circumdatae stationes, stratis etiam turn per cubilia propterque mensas, nullo metu, non antepositis vigiliis. Adeo cuncta incuria disiecta erant, neque belli timor ; ac ne pax quidem nisi languida et soluta inter temulentos," etc.
26 YULE AND CHRISTMAS
St. Martin was born in a.d. 336, and died in 401. About the middle of the sixth century there was definitely given as his own day November 11, which had before been celebrated in Gaul in his commemoration. Whilst France knows nothing of a popular celebration of it, Martinmas ^ in early times was held as the highest festival of the year wherever Western Germanics lived, the banqueting lasting, in the sixth century, all night long till morning broke. We know this from the terms in which the Synod of Auxerre in 578 forbade its celebration. ^ Beda, in De Temporum Ratione, testifies to a Germanic festival in November, saying that in that month, which they called Blot-monath or Offering-month, the heathen Germanics devoted to their gods their cattle, which they intended to kill.* Martinmas, probably, was in chief view when in 589 the Council of Toledo interdicted the same nightly feasting for all saints' days,^ and the Concilium Cabilonense of 650 A.D. repeated the prohibition.^
^A very brilliant sketch of his life and activity is given by Heino Pfannenschmid, Germanische Erntefeste, Hannover, 1878, pp. 193 ss., with numerous notes, pp. 464 ss.
^Pfannenschmid, Ibid., p. 466, note 10.
^"Omnino et inter supradictas conditiones, pervigilias, quas in honore domni Martini observant, omnimodo prohibete," Acta Conciliorum, Parisiis, 1 7 14, Vol. HI., col. 444, Synodus Autissiodorensis, A.D. 578, v. The supradictae conditiones can only be the contents of Canons iii. and iv., which run as follows: "iii. Non licet compensos in domibus propriis, nee pervigilias in festivitatibus sanctorum facere ; nee inter sentes, aut ad arbores sacrivos, vel ad fontes vota exsolvere : sed quicumque votum habuerit, in ecclesia vigilet, et matriculae ipsum votum, aut pauperibus reddat : nee seulptilia aut pede, aut homine lineo fieri penitus praesumat. iv. Non licet ad sortilegos, vel ad auguria respicere, non ad caragios, nee ad sortes, quas sanctorum vocant, vel quas de ligno, aut de pane faciunt, aspicere : sed quaecumque homo facere vult ; omnia in nomine Domini faeiat." It is important to notice that the only two feasts which are mentioned by their name by that Synod of Auxerre are the Calends of January and Martinmas, and from that the conclusion may be drawn that the heathen customs were more prominent at these two tides than at any other.
^ De Temporum Ratione, chap, xv., De Mensibus Anglorum: ^^ Blotmanoth, mensis immolationum, quia in ea pecora quae occisuri erant, Diis suis vovebant."
^ Acta Conciliorum, Parisiis, 1714, Vol. III., col. 483, Concilium Toletanutn, III., a.d. 589, xxiii. : " Exterminanda omnino est irreligiosa consuetude, quam vulgus per sanctorum solemnitates agere consuevit ; ut populi, qui debent offieia divina attendere, saltationibus et turpibus invigilent canticis."
^Acta Conciliorum, Parisiis, 1714, Vol. III., col. 950, Concilium Cabilonense, A.D. 650, xix,: " Malta quidem eveniunt, quae dum levia minime corriguntur, saepius majora eonsurgunt.
THE FEAST OF MARTINMAS 27
The veneration of St. Martin spread very rapidly in the old Church. After God, the holy cross, and the Virgin Mary, the blessed confessors Hilarius and Martinus were held in the highest reputation in the Gaul of the sixth century,^ and in Great Britain things were not much different,^
Valde enim omnibus noscitur esse indecorum, quod per dedicationes basilicarum, aut festivitates martyrum, ad ipsa solemnia confluentes chorus femineus turpia quidem et obscena cantica decantare videntur, dum aut orare debent, aut clericos psallentes audire. Unde convenit, ut sacerdotes loci talia a septis basilicarum, vel porticibus ipsarum, ac etiam ab ipsius atriis vetare debeant et arcere. Et si voluntarie noluerint emendare, aut excom- municari debeant, aut disciplinae aculeum sustinere." It was then the habit to sing in these days worldly love songs in the church, to dance in accompaniment ot them, and to banquet in the same sacred place. '* Non licet in ecclesia choros secularium, vel puellarum cantica exercere, nee convivia in ecclesia praeparare : quia scriptum est : Domus mea, domus orationis vocabittir" Acta Conciliorum, Parisiis, 1714, Vol. III., col. 445, Synodus Autissiodorensis, A.D. 578, ix. Acta Conciliorum, Parisiis, 1714, Vol. III., col. 1920-I, Concilium Germanicum, A.D. 742, v.: '* Decrevimus quoque, ut secundum canones unusquisque episcopus in sua parochia solicitudinem gerat, adjuvante gravione, qui defensor ecclesiae ejus est, ut populus Dei paganias non faciat, sed omnes spurcitias abjiciat et respuat ; sive profana sacrificia mortuorum, sive sortilegos, vel divinos, sive phylacteria et auguria, sive incantationes, sive hostias immolatitias, quas stulti homines juxta ecclesias ritu pagano faciunt, sub nomine sanctorum martyrum vel confessorum, Deum et sanctos suos ad ira- cundiam provocantes : sive illos sacrilegos ignes, quos Niedfyr vocant ; sive omnes, quae- cumque sunt, paganorum observationes diligenter prohibeant."
'*' Acta Conciliorum, Parisiis, 1 7 14, Vol, III., Epistola Sanctae Radegundis ad Episcopos, A.D. 567, col. 370: "Dei et sanctae crucis, et beatae Mariae incurrat judicium: et beatos confessores Hilarium et Martinum, quibus post Deum sorores meas tradidi defendendas, ipsos habeat contradictores et persecutores. " The Concilium Turonense, Ibid., col. 371-72, replied to this letter mentioning Martin in almost equal terms, saying ot God : " Beatum Martinum peregrina de stirpe ad inluminationem patriae dignatus est dirigere misericordia consulente. Qui licet apostolorum tempore non fuerit, tamen apostolicam gratiam non effugit. Nam quod defuit in ordine, suppletum est in mercede," etc.
'^When the Roman missionaries under Augustine, sent by Pope Gregory to Great Britain, settled down in Canterbury, they found there a St. Martin's Church, as Bede states, of Roman origin — a church which, after some medieval reconstructions, still exists, and is not the least interesting of the antiquities of Canterbury. Beda, Historia Eccle- siastica gentis Anglorum, I., chap, xxvi., ed. Plummer, Oxford 1896, p. 47: " Erat autem prope ipsam civitatem ad orientem ecclesia in honorem sancti Martini antiquitus facta, dum adhuc Romani Brittanniam incolerent, in qua regina, quam Christianam fuisse praediximus, orare consuerat." (Her name was Bercta, and she was a Frankish princess.) St. Martin, besides, had at Canterbury a potticus, in which King Aedilberct was buried (Beda, Hist. Eccl. II., chap. v. : " Defunctus veto est rex Aedilberct die XXiiii. mensis
28 YULE AND CHRISTMAS
whilst in Germany the cult of the same saint spread simultaneously with Christianity.^ The popularity of St. Martin was bound to increase from the fact that his day was placed at the greatest ancient Germanic festive tide; and however scarce is our information about medieval popular festivals, we know of no other so much as we do about Martinmas as a time of feasting and banqueting. If Martin is called the drunken saint, there is no doubt about the significance of that expression, nor is there any about the so-called Martin-geese. The oldest St. Martin's goose of which we know is a silver one, and belongs to the year 1171, although the testimony by which it is warranted is not contemporaneous. A monk of Corvei,
Februarii post XX. et unum annos acceptae fidei, atque in porticu sancti Martini intro ecclesiam beatonim apostolorum Petri et Pauli sepultus ubi et Bercta(e) regina condita est.") St. Ninian had a bishop seat, which was later on celebrated through the name and the church of St. Martin (Beda, Hist. Eccl,, III., chap. iv. : "Cuius sedem episcopatus, sancti Martini episcopi nomine et ecclesia insignem . . . iam nunc Anglorum gens obtinet "). Ninian probably died earlier than St. Martin {Lives of St. Ninian and St. Kentigern, ed. Forbes, 1874, xxvii., xxxviii., 256, 266, 271-273). There was also a monastery called after that saint (Beda, Hist. Eccl., IV., xvi. : " Et abbas monasterii beati Martini, ..." and "corpusque eius ab amicis propter amorem St. Martini, cuius monasterio praeerat, Turonis delatum atque honorifice sepultum est"; and Beda, Historia Abbatuin, § 6, ed. Plummer, p. 369 : " Ab Agathone papa archicantore ecclesiae beati apostoli Petri et abbate monasterii beati Martini Johanne . . ." and Historia Abbatum auctore Anonymo, ed. Plummer, in Venerabilis Baedae Opera Historica, Oxford 1296, Vol. I., p. 391, § 10: " Abbatemque monasterii beati Martyni"). Mr. Plummer, in his edition of Bede's His- torical Writitigs (Vol. II., p. 43) says: "To the popularity of the cultus of St. Martin (who died between 397 and 401 ) in Britain, Venantius Fortunatus (born about 530 at Ceneta and having died as bishop of Poitiers after 600) bears striking testimony, saying of him : " Quem Hispanus, Maurus, Persa, Britannus amat." Cf. Haddan and Stubbs, Councils and Ecclesiastical Documents relating to Great Britain and Ireland^ I., 13, where see note for references illustrating the connection of St. Martin with the British Isles ; Venantius Fortunatus, Vita S. Martini, Lib. IV., vv. 621 ss. {Monumenta Historica Germaniae, 4to series).
^ Boniface also speaks of " fundamenta cuiusdam destructae a paganis ecclesiolae, quam Willibrordus ... in castello Traiecto repperit, et eam proprio labore a funda- mento construxit et in honore S. Martini consecravit {Monuvienta Moguntina, pp. 259- 260, ed. Jaffe). Ducange, Glossariict?t, under Festum S. Martini: " Recensetur inter festa quae celebrari debent, in Lib. VI. Capitul., c. 189; in Capitulari Aquisgran., A.D. 817, c. 46; in Capitulis IValterii Aureliani, c. 18; in Concilio Lugdunensi sub Inn. III., etc.; in Capitulari Ahytonis Episcopi, Basiliensis, c. viii. ; Beletus, c. 163."
THE FEAST OF MARTINMAS
29
who, in the middle of the fifteenth century, compiled the Annals of that monastery,! ^q[\^ th^t in 11 71 Othelricus of Svalenburg, on the feast of St Martin, because he was a member of the fraternity of Corvei, gave the monks as a present a silver goose. This gift was probably in com- pensation for a payment of some annual duty on Martinmas.^
In the thirteenth century there was a song on St. Martin known through large parts of Gaul and Germany, which, like almost all songs in use at Germanic offering tides, was repudiated by the clergy as indecent. It was then ascribed to some evil spirit, and a story was popular according to which, in 12 16, a demon had boasted that he and a friend of his composed that song, and promulgated it over a large territory.^ Public bonfires on Martinmas can be proved to have existed as early as 1448, when Martinmas for that reason was called Funkentag;'^ and are again mentioned about the end of the sixteenth century in Fischart's (+1590) Gargantua, where baskets are stated to have been burnt in the St. Martin's fire. Perhaps it has also to do with St. Martin's fires that, when in 1557, at Augsburg, a house was burnt
^ Leibnitz, Scriptores, II., 308 ; Pfannenschmid, Germanische Emtefeste, p. 229.
^ Annales Corbejenses in Leibnitz's Scriptores, II., 308: "Othelricus de Svalenberg aigenteum anserem in festo S. Martini pro fratemitate (obtulit). " Pfannenschmid, Gernianische Emtefeste, 1878, p. 505. Old Leibnitz, Scriptores, II., /ntrodmtio, p. 28 : " Anserem assatum in festo S. Martini per omnes fere domos, mensis inferunt Gennani. . . . In\ntat anni tempus : turn enim anseres pingues habentur." Pfannenschmid, Ibid., p. 505, beats all speculators about the connection between St. Martin and geese, by the simple declaration that St. Martin's day is just the time of the year when geese are fat. This was of even greater moment in fonner centuries, when the accumulation of food was attended with considerable difficulties, and domestic animals were difficult to feed during winter time. Compare also D. Georg Joachim Marks, Geschichte vovt Martini- Abend und Martins-Mann, Hamburg und Giistrow, 1772, p. 20.
3 The story is told by Thomas Cantipratensis, who in 1263 wrote his book on the bee state. It is contained in his treatise Bonum universale de apibus: "Quod autem obscoena carmina finguntur a daemonibus et perditorum mentibus immittuntur, quidam daemon nequissimus, qui in Nivella urbe Brabantiae puellam nobilem anno domini 12 16 prosequebatur, manifeste populis audientibus dixit : cantum hunc celebrem de Martino ego cum collega meo composui et per diversas terras Galliae et Theutoniae promulgavi. Erat autem cantus ille turpissimus et plenus luxuriosis plausibus."
■•In a document of Count Friedrich zu Moers (A. J. Wallraf, Altdeutsches Worterbuch^ Koln, p. 23). Comp. Grotefend, Zeitrechnung, I., p. 71.
30
YULE AND CHRISTMAS
down, a report expressly remarks that the young fellows when feasting and holding Martinmas had neglected it.^
According to a seventeenth century source,^ in Holland the boys, on the eve of Martinmas, lit fires, singing :
" Stoockt vyer, mackt vyer: Sinte Marten konit hier Met syne bloote arnien ; Hy sonde hem geerne warmen."
About 1230 an Austrian poet represents peasants drinking to the praise and memory of St. Martin.^ A joke connected with the Martinmas of the thirteenth century is preserved to us in the poem St. MarttnsnachtA A rich farmer and jhis people have got drunk in honour of St. Martin. A thief breaks into the farmer's stable, and, when surprised by the owner, shakes off his clothes, pretending to be St. Martin. The farmer, believing him, goes on with his banquet, with the result that in the morning he finds his stable empty. That the festivity was equally familiar to monasteries is apparent from some documents. The monastery of Eilenrostorf received every year, from 1353, a quantity of wine — half for the mass and half for the convent who were to drink it in vigilia sancti Martini.^ What it received before that date is not known.
A.D. 1369 we have a description of a celebration of Martinmas. A knight, von Schwichelt, possessor of Liebenburg, asked Duke Otto of Gottingen to spend Martinmas with him. Duke Otto had gone to Harzburg, which he had taken from the Count of Wernigerode, at the same time compelling, by surprise of the town of Alfeld, the Bishop of Hildesheim
^Birlinger, Aus Schwaben, II., p. 132. This happened on October 10. Nevertheless that feasting was called Martinsnacht , apparently because in olden times it had been held on November 11.
'^ Gisbertus Voetius, Selectae Disputationes Theologicae, Utrecht, 1659, p. 448.
^Der Strieker, Kleine Gedichte, V., 167, Grimm's Worierbuch, IV. I, 1263: "Dem guoten sant Martine ze lobe und zu minnen."
^Hagen, Gesammtaienteuer, Stuttgart und Tubingen, 1850, No. 50.
' Reimann, Deutsche Volksfeste, 284 ; Pfannenschmid, Germanische Emtefeste, p. 222 ; Marks, Geschichtevom Martini- Abend und Marlinsmann, Hamburg, 1772, p. 20.
THE FEAST OF MARTINMAS
31
to provide the castle of Harzburg with food. On Martinmas eve he arrived with his army before Liebenburg, and was invited for St. Martin's banquet. He accepted the invitation, and on the following day presented his host with the castle of Harzburg.^ Oswald von Wolkenstein (1367-1445) sings: Trinckh martein wein, und genss I'ss Ott^'^ and bones of Martin's geese were used for prophecy about the middle of the fifteenth century.^ Sebastian Franck (1500-1545) says in his Weltbuch of the Francs : "Firstly they praise St. Martin with good wine and geese, until they are drunk. Unblessed is the house which has not a goose to eat that night; then they also tap their new wines which they have kept so far," to which he adds a description of a St. Martin's game : " In Franconia at that day people enclosed in a circuit or circle two boars, which tore each other to pieces. The meat was divided among the people, the best bits being given to the authorities."*
^ Bodonis Chronicon pict. in Leibnitz's Scriptores Br., III., 385, H. Pfannenschmid, Germanische Erntefeste, p. 500 ; and Uralte Sachsen Chronic by Caspar Abel, written about 1455 (^<i annum 1375). Pfannenschmid, Ibid.: "HertogOtto de bose to Getting halde eyn grot hop des Quecks uth Holt-Lande van der Wulfesborch, unde wolde darmidde driven in dat Lant to Getting, so legerde he sick under der Levenborch, unde was St. Martens- Avend, dar spysen se one myt alle sinen Volke, unde dem Quecke, des Morgens wolde he de Koste betalen, des wolden de van Schwichgelde nyn Gelt vore hebben, unde ereden sine Gnaden darmidde, do dreyff he sin Roffqueck in dat Lant to Getting, unde spisede dar sine Borghe midde, unde gaff do denen Schwichgelde vor de Woldad de Hartesborch to erve unde to egen, de worden so derna der Borch Goddes Friint, unde aller werlde vyent."
"^ Odo = November 13.
^Dr. Hartlieb, physician-in-ordinary to Duke Albrecht Oi Bavaria, in his BucA aller verboten kunst, ungelaubens und der zatiberei (1455) says: "Als man zu sant Martinstag oder nacht die gans geessen hat, so behalten die eltesten und die weisen das prustpain, und lassen das trucken werden bis morgens fru und schawen dan das nach alien umbstenden, vorn hinden und in der mitt. Darnach so urtailen si dan den winter wie er sol werden kalt warm trucken oder nass, und sind so vest des gelauben, das si darauf verwetten ir gut und hab." A hundred and fifty years later that habit was still in use, as we know from J. Golems, Calendarium Oeconomicum (1591), and from Olorinus Variscus and his writing on St. Martin's geese.
**'Nach dem kompt S. Martin, da jsset ein jeder Haussvater mit seinem Ilaussgesinde eine Ganss, vermag ers, kaufft er jnen Wein vnd Medt, vnd loben S. Martin mit voll seyn, essen, trincken, singen"; Heinrich Panthaleon, of Bale (1522-1595), writes in Der deutschen Nation Heldenbuch : "Die Leute pflegen zum Gedachtniss S. Martini in Deutschland mit
32
YULE AND CHRISTMAS
A writer of the same century says; "We Germans think Shrove Tuesday, St. Burkhard, and St. Martin, Pentecost, and Pasch, the times when people should be gay and banquet more than at other seasons of the year: on St. Burkhard's eve for the sake of the new must ; on Martinmas perhaps for the sake of the new wine; then people roast fat geese, all the world rejoicing." ^
Sermons of the seventeenth century, even when coming from Protestant pulpits, have many things to tell about Martinmas and its geese.^ The popular rime says :
" Auff Martini schlacht man feiste Schwein, Und wird allda der Most zu Wein."^
On the eve of Martinmas the devil had free play. On that night, in the
frohlichem Gemlith St. Martensnacht zu begehen, die Martensganss zu essen, und mit den Nachbaren und dem Hausgesinde frohlich zu sein, gleich als wenn aller Dinge Ueberflusi mit Sanct Martino der Armen Patron vorhanden say;" Jod. Lorichius, Aberglatiben, 1593, p. 52. Simrock, Martinslteder, xiv. ; Pfannenschmid, Germanische Erntefesie, pp. 500-1, where other proofs for similar festivities are given from Thomas Naogeorgus (Thomas Kirchmaier of Straubingen, 1511-1563), Regnum Papistkum, Lib. IV. ; Joannes Boemus Aubanus, De Omnitan Gentium Kitibus, 1520, fol. Ix. ; G. Forster, Frische Liedlein, II Parts, Nlirnberg, 1540, No. 5, and many items of later dates.
^ Scheible, Schaltjahr, II., 95, from Agricola.
^ Scheible, Schaltjahr, I., 187, Atis einer protestantischen Martinspredigt des 17. Jahrhunderts : " Und weil heute der Tag Martini gefallet, daran es die Ganse leider iibel haben, als will ich zu einer Martinsgans bitten, und dieselbe anatomieren und zutheilen. Nicht aber wie die Aberglaubischen, nach jetzt erzahltem heidnischem Gebrauch, von kiinftiger Winterwitterung aus dem Brustbein weissagen, sondern was wir bei einer Gans christlich zu lernen haben, anzeigen. Richtet ihr hierauf eure beharrliche Andacht. — Es isset Mancher eine Gans nach der andern, und ist und bleibet selbst eine Gans, versteht und weiss nicht, was Gott und die Natur uns an derselben zu studiren gegeben." Ibid., I., p. 194: "Ganse geben Speis, sonderlich um diese Martinszeit. Drum ihnen auch der Martinstag sehr gefahrlich ist. . . . Verstandige Koche wissen sie mit gutem Beifusz, Aepfeln und Kastanien zu fullen und zu einem lieblichen Schmack zu geben." On Martinmas gaieties a mass of material is contained in Mussard, Ceremoniae Ecclesiasticae, p. 117; Blumberg, Delineatio frateniitatum Calendarum, p. 155; Calvor, Ritual EccL, P. II., p. 362; Keisler, Antiquitates Stptentrionales ; Pirnische Chronick in Meiuken, II., p. 1554; Marks, Geschichte vom Martini- Abend und Martinsmann, Hamburg, 1772.
^Grasse, Des deutschen Landmanns Practica, Dresden, 1858, p. 27.
THE FEAST OF MARTINMAS 33
shape of a man dressed in a long wolf-skin coat, he appeared in 1594 at Spandau, Brandenburg, to a young fellow, and raged about in an indescrib- able fashion, so that all the persons possessed had to be brought to the high altar of the church for protection.^
In the seventeenth century the police began to become an important factor in the development, or rather suppression, of popular usages. So on February 4, 1605, and June 22, 1649, in old Frankfurt-on-the-Main, it was forbidden to bake Mertins hornichen for sale. The same happened to the other Martin's festivities at many places. 2 Yet some lived on for a considerable time as quasi legal institutions, e.g.^ in the Rhine country, where such festivities existed till after 1750,^ being given as a gift in return for the duties paid at that term.
^Scheible, Sckaltjakr, IV., 462, Schreckliche Zeitung.
2 "Die libel practicirten Martins- oder Herbsttriinke " (" Wurzburger Heibstinstruction " in Werndii Tractatus vom Zehntrechte, p. 324 (ed. de anno 1708) ; Pfannenschmid, Ger- manische Erntefesie, p. 224 ; comp. also Schiller's Glossar., p. 123.
^Al Oberaussem, in 1750, at the banqueting {Hofessen) on Sunday after Martinmas, the persons who took part were sixty-one. They dined at two tables, at a tisch auffm soller and a specktisch ifn haus (Armin Tille, Archiviibersicht, p. 102).
CHAPTER IV.
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR.
Martinmas is the earliest term occurring in the Anglo-Saxon Laws. Church-scot had to be paid then, so early as the seventh century/ though the church to which it was to be paid was that of the place at which a man stayed in the beginning of the calendar year,^ He who failed to do so was to forfeit sixty shillings, and render the church-scot twelve-fold.^
^ Haddan and Stubbs, Councils and Ecclesiastical Documents, Oxford, 1871, HI., 215, Laws of King Ine of Wessex, about A.D. 690 (688-693): "Be Ciric-sceattum. Ciric- sceattas sin agifene be Sancte Martines mgessan."
"^ Ibid., HI., 217, Ixi. : "Be ciric-sceatte. Ciric-sceat mon sceal agifan to tham healme and to tham heordhe the semon on bidh to middum wintra."
^ In Thorpe's Ancient Laws and Institutes of England (London), 1840, II., 460. In the canons enacted under King Edgar (ca. a.d. 967) it was enjoined that plough-alms were to be given xv. days after Easter ; and a tithe of young by Pentecost ; and of earth-fruits by All Saints; and Rome-"feoh" by St. Peter's mass; and church-scot by Martinmass ("serest sulh-oelmessan xv. niht onufan Estron. and ge&gudhe teodhunge be Pentecosten. and eordh-westma be Omnium Sanctorum, and Rom-fe6h be Petres-maessan, and ciric-sceat be Martinus-maessan "). This is the reading of MS. D, a small folio of the middle of the eleventh century. Corpus Christi, 201 (v. 18) ; X, a large octavo MS. of the tenth century, Bodleana, Junius, 121, has Ealra Hdlgena jnisssan instead of Omnium Sanctorum, and has the following sentence immediately preceding to the quoted Anglo- Saxon text : ";and riht is that man thisses mynegige to Eastrum. odhre sidhe to gang-dagum. thriddan sidhe to middan-sumera. thonne bidh msest folces gegaderod ; " whilst it adds after the above sentence from D the following: "and leoht-gesceotu thriwa on geare. serest on Easter-sefen. and odhre sidhe on candel-msesse sefen. thriddan sidhe on Ealra Halgena msesse sefen." Acta Conciliorum, Parisiis, 1714, Vol. VI., i, col. 657-8, Leges Ecclesiasticae Regis Edgari, ca. A.D. 967, iii. : " Quisque fetuum decimas omnes ante
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 35
This term lived on a considerable time. It was, at any rate, preserved as late as the reign of Henry the First (1100-1135), but probably a good deal longer.^ It is like many other Germanic institutes also found in
Pentecosten persolvito : terrae quidem fructuum decimas ante aequinoctium pendito : ipsas autem seminum primitias sub festum divi Martini reddito." Ibid., col. 659, iii. : " Et omnis decimatio juventutis reddita sit ad Pentecosten, et terrae frugum ad aequinoctium, et omne ciricsceattum ad festum sancti Martini, per plenam forisfacturam quam judicialis liber dicit." Ibid., col. 776 {Concilium Aenhamense, A.D. 1009) x. : "Jura Deo debita unusquisque annuatim recte pendito : aratri scilicet eleemosynam decimaquinta nocte a Paschate : fetuum seu novellorum gregum decimas, ad Pentecosten ; et terrae fructuum, ad festum omnium Sanctorum, xi. Census Romae debitus [quem denarium sancti Petri vocant] and festum sancti Petri ad vincula (alias Missam Petri) persolvatur : et ecclesiae census, qui cyrick sceat appellatur, ad Missam sancti Martini. xii. et. xiii. Luminarium census ter quotannis penditor". A mention of this institute occurs in Cnut's letter from Rome (Thorpe, Ancient Laws and Institutes of England {X^x^Aoxi) 1840, I., p. 104: " Et in festivitate Sancti Martini primitiae seminarum ad ecclesiam sub cujus parochia quisque degit, quae Anglice 'ciric-sceatt' nominatur") {Ibid, in the notes). Also Heming, 21 : " De cirisceato de Perscora dicit vicecomitatus, quod ilia ecclesia de Perscora debet habere ipsum cirisceattum de omnibus ccc. hidis, scilicet de unaquaque hida ubi francus homo manet unam summam annonse, et, si plures habet hidas, sint liberse, et si dies fractus fuerit, in festivitate Sancti Martini ipse, qui retinuerit det ipsam summam, et undecies persolvat abbati de Perscora, et reddat forisfacturam abbati de Westminstre quia sua terra est" {Ibid.). Cnut's letter is also printed in the Acta Coftciliorum.Vaiism, 1714, Vol. VI., I. col. 846, Epistola Canuti Regis ad Anglorwn proceres, A.D. 1031 : "Omnium debita, quae secundum legem antiquam debemus, sint persoluta: scilicet eleemosyna pro aratris, et decimae animalium ipso anno procreatorum, et denarii, quos Romam ad sanctum Petrum debetis, sive ex urbibus, sive ex villis, et mediante Augusto decimae frugum, et in festivitate sancti Martini primitiae seminum, ad ecclesiam sub cujus parochia quisque degit, quae Anglice Curcset nominatur." Bye and bye Easter creaps into the number of these terms {Acta Conciliortim, Parisiis, 1714, VI., I, col. 899, Leges Ecclesiasticae Canuti Regis, A.D. 1032), where the terms enumerated are: "a fortnight after Easter;" "Pentecost;" "All Saints," whilst, besides, Peter's penny is to be paid at Peter and Paul, and the firstlings of the seeds at Martinmas ; also three times a year the candle money has to be paid {Ibid., col. 899, xii.): "at Pasch," "at All Saints," and "at Mary's Purification"; Ibid., col. 908, xvi. ; col. 909, xvii., xix ; Thorpe's Ancient Laws, XL, 524.
1 Thorpe, Ancient Laws, I., 520 {Leges Regis Henrici Primi, xi., § 4): "Qui cyric- sceattum tenebit ultra festum Sancti Martini, reddat eum episcopo, et undecies persolvat, et regi 1. solidos." This institute is commented upon by Lingard, Altertiimer der angelsdchsischen Kirche, ed. by von Ritter, Breslau, 1847, p. 56. Compare David Wilkins, Concilia Magnae Britanniae et Hiberniae, London, 1734, I., pp. 59, 302; and Pfannen- schmid, Germanische Erntefeste, p. 204.
36 YULE AND CHRISTMAS
Spain, whither the Goths apparently had carried it,^ and in Germany, where it obtained all through the Middle Ages. There the church tax falling- due at Martinmas was even called St. Martin's penny,^ when the payment of it had been shifted to Christmas, a change which seems to have taken place in the thirteenth or in the beginning of the fourteenth century. A similar shifting of terms is to be observed in England in the eleventh century.^ Church-scot was not the only payment to be made at the beginning of winter. Then was also paid an instalment of the wages of female servants, which were due three times a year, viz., at the beginning of winter, at mid-Lent, and at the beginning of later summer.^ Ij In Germany Martinmas was legally recognised as a general term for paying
^ Vita Sancti Isidori Agricolae, Madriti in Castelia ( + 1 1 30), auctore lohanne Diacono (1275, comp. Potthast., Wegweiser, 766), No. 15 (Ducange-Henschel, Glossary, 1845, IV., 304): "Accidit, quendam virum ex eius curia ad colligendam exactionem regiam, quae vulgariter dicitur Martiniega, in tempore hiemis sub mense Decembri Majorinum certissime advenisse." H. Pfannenschmid, Girmanische Erntefeste, 1878, p. 466.
^Norrenberg, Geschichte der Pfarreien des Dekanates Miinchen Gladbach, Coin, 1889, p. 276, No. 23. A.D. 1324, Dec. 24: Among the revenues of the church at Giesenkirchen is named: "denarium, qui dicitur Mertyns pennynge."
^ John Earle, A Hand- Book to the Land- Charters and other Saxonic Docutnents, Oxford, 1888, pp. 344-345. Eadward (1042-1066), his Writ of Privileges to the Abbey of Ramsey, CO. Huntingdon (Manuscript of century xil., Cottoniana, Otho, B, xiv., f. 257): "and ealle dha gyltes dha belimpedh to mine kinehelme inne lol and inne Easteme and inne dha hali wuca set Gangdagas on ealle thingan al swa ic he6 meseolf ahe, and tolfreo ofer ealle Engleland, widhinne burhe and widhiitan, set gares cepinge and on sefrice styde, be wsetere and be lande ; habeant et omnes forisfacturas quae pertinent ad regiam coronam meam in natali dominico, in pascha, et in sancta ebdomada rogationum, in omnibus rebus sicut ipse habeo, et per totam Angliam infra ciuitatem et extra, in omni foro et annuls nundinis et in omnibus omnino locis per aquam et terram, ab omni telonii exactione liberi sint."
■* Thorpe, Ancient Laws, I., 436-7, "Rectitude Ancillae : ad hiemale companagium; ad quadragesimalem convictum ; and in aestate, i.e. to winter-sufle, to laengten-sufle, and on sumera." A similar state of things survived up to the present time. Notes and Queries, Ninth Series, February 4, 1899, p. 85, in a note on Pack Rag Feast by R. Hedger Wallace : ' ' The agricultural labourers in some of the North Derbyshire villages, among other old customs, retain that of having a social gathering on Old Martinmas Day (23 November), which is, not over politely, designated the Pack Rag Dinner. The name refers to the fact that the indoor menservants about the farms, who are changing masters at Martinmas, gather together their belongings for removal from one house to another."
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 37
\ duties about the time of Charlemagne, and in the ninth century it was in general use.^ In thirteenth century ordinances about leases, geese or fowls are mentioned which are delivered at Martinmas.^ We know of such gifts
^ to the clergy on saints' days from about the middle of the sixth century. ^
^ Anton, Geschichte der Landwirtschaft, I., p. 341, and Pfannenschmid, Germanische Emtefeste, p. 204, confuse the legal recognition of an existing status with the introduction of it.
^ Pfriindenordnung of the Monastery of Geisenfeld, ed. by Wittmann, Miinchen, 1856; 'Ltyitr, Mitlelhochdeutsches Ergdnzungsworterbuch, I., p. 736 ; Pfannenschmid, GertJianische Emtefeste, p. 205 : " leclichen hof und vourt unde sunderlich hus verzendet man mit eyme hune ze sente Mertinstage," Sachsenspiegel, ed. by Weiske, Landrecht, Book H., Art. 48, § 5 ; or, as another version has it : " Jeglichen hoff, odder wiiste hofstadt vnd sonderlich heuser, verzehent man mit einem hune, am S. Martinstag." "An St. Martinstag sind allerhand pfleg und zins verdient," Ibid. H., 58, from the thirteenth century, Middle Germany.
^ Acta Coiuiliorum, Parisiis, 1 7 14, Vol. HI., col. 352; Cotuiliujji Bracarense, I., xxi. : " Item placuit, ut si quid ex collatione fidelium, aut per festivitates martyrum, aut per commemorationem defunctorum offertur, apud unum clericorum fideliter colligatur ; et constituto tempore, aut semel, aut bis in anno, inter omnes clericos dividatur : nam non modica ex ipsa inaequalitate discordia generatur, si unusquisque in sua septimana quod oblatum fuerit, sibi defendat." Ample evidence on Martinmas as a term I have given in my Geschichte der deutschen Weihnacht, Leipzig, 1S93, PP- 23-28, and pp. 291-296. Mark's Geschichte vom Martini- Abend und Martins-Mann, Hamburg, 1772, contains on pp. 26, 27 a chapter (13), Von der Zahlungsfrist auf Martini, and mentions there documents of a.d. 1294, 1318, 1460, to w^hich are to be added those mentioned in E. J. Westphalen's Monumenta Inediia, Part IV., Preface, p. 95. Nicolaus von Werle, A.D. 1297, gave the town of Waren exemtionem ab angariis et petitionibus omnibus under the condition that the citizens every year at Martinmas would send him on a cart quantitatem seminum ; Georg Joachim Mark's Geschichte vom Martini-Abend und Martins-Mann, Hamburg, 1772, p. 49. Ibid., p. 81, mentioned that two generations earlier Graf Heinrich of Schwerin concluded a bargain with Archbishop Engelbrecht of Koln to the effect that the archbishop had to send him annually at Martinmas fifteen harradas or barrels of wine. That book is devoted to the question of the origin of the custom of the so-called Martinsviann at LUbeck, which, however, it fails to answer. The origin of the habit is unknown, but it is certain that in 1567 it was called an old habit, that on Martinmas the Town Council of the Imperial city of Liibeck sent a barrel of old Rheinweinmost to the Court of Schwerin. The story is firequently related in modern times, e.g., by Mark ; by Reimann, Deutsche Volksfeste, p. 28S ; by Pfannenschmid, Germanische Emtefeste, pp. 222, 223. As a gift in return for the duties paid — a kind of Germanic New Year's gift — the Town Council of Liibeck received some deer. The same was the case in Wiirtemberg monasteries. There the prelate was under obligation
\
38 YULE AND CHRISTMAS
That Martinmas was not merely a term like other terms, but the beginning of the economic year, is evident from the fact that, in the Middle Ages, accounts run from Martinmas to Martinmas, and appointments were made for the same period, which is also understood to cover a year of taxation.
As long as the duties were paid in natural products, the terms were the last days within which the duties had to be paid; but when the taxes were paid in money, they became the days on which the payment had to be made.^ As early as a.d. 1253 money payments for the whole of the year of taxation were made at Martinmas.^ In medieval Frankfurt a. M. the year for which officials were appointed ran from Martinmas to Martinmas, or the Sunday previous.^ So did the period of imperial taxation,* and so
to give the Martin's wine to all people of his place. In the provostship of Hellingen each holder of a tenure received a pint, each old man and each vk^oman half a pint, and male and female servants, and even the baby in the cradle, a quarter (Nork, Festkalender, p. 684 ; Reinsberg-Uiiringsfeld, Das festliche Jahr, p. 340). Wine was also given at Martinmas at the court of the Archbishop of Mayence, at Erfurt, about 1494 (Michelsen, Der Maimer Hof zu Erfurt am Ausgange des Mittelalters (1494), Jena, 1853, p. 26, Regu- lations for the Kiichenmeister (the highest economic official) : " Uff sanct Martins abent sal er wein, uff weihenachten opffergelt und uff das neue jhor zum neuen jhor geben, wie das rothbuchlein und die rechnung in belt, und auch christsemeln wie sich gehurth geben,") and about 1520 at Wiirzburg, as Martin Boemus tells us.
^ Arnold, Ztir Geschichte des Eigentums in den deutschen Stddten, Basel, 1 862, p. 68.
^Grimm, Deutsche KechtsaltertUmer, III., 607, Oeringen, A.D. 1253: " Swer dirre stete reht hat, der sol geben ze sante Mertins naht achte heller, und sol daz jar alles fri sin zolles halp."
^ Frankfurt a. M. Rechenbuch, 1358-59, fol. iG* : " Hartmude an fizscher porten synen jarlon 31b. 3SS. und geng sin jar an des suntags vor sant Mertinsdag." Ibid., 1374, fol. 93*: " Sabbato ipso die divisionis apostolorum Gultsmede dem wechter uff dem thorne zu Bonemesse l lb. sines penning lones unde 2 lb. fur 3 achtel komes unde ist da midde sines halben jarlons bezalet unde ged sin jar uz unde an uff sand Mertinsdaig unde pliget man yme eyn jar zuo gebin 61b. 6 achteil kornes unde i rog ;" Ibid., 1474, fol. 22^, under " Einzelinge Innemen:" "Item 2951b. 14 ss. i,\\i. han wir enphangen von Johan Heller, schriber im spitale zum heiligen geiste als der uns rechnunge getan hat von dem jare das Martini anno 74 ussgangen ist." Actum sabbato post dominicam Esto mihi anno 1475.
'^Frankfurt Kechenbuch, 1435, fol. 39% under "Einzelinge ussgeben:" "i 100 lb., 141b. mynner 3iss. han wir ussgeben und bezalt unserm gnedigen herren dem keiser keiser Sigmund die gewonliche des rijchs sture von sant Mertinstag izunt vergangen die im
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 39
it came that the tax had to be paid for all people who lived to see St. Martin's day, but not for those who died before it, because their lives did not cover the whole of the taxation year.^ On the other hand, the amount of corn and wine requisite for maintaining the owner and his servants from the date of paying the taxes till Martinmas was free from duty,2 Martinmas being the last term for paying the duties.^ In the medieval Frankfurt a. M., all through the fourteenth century, the taxes were raised in November or December,^ as they are now in Scotland, the taxation year beginning with Martinmas.^ In the thirteenth century the winter, during which all agricultural work was interrupted, was counted from Martinmas till Sf. Petri ad catJiedram (Feb. 22), for during that time, in 1297, the Ffahl- biirger of Frankfurt a. M. were required to have a household within the ramparts of the city.^ That on the banks of the Rhine Martinmas in the twelfth century also was considered as the beginning of the economic year is evident from a document of a.d. 1149, of Hirzenach, near Boppard, according to which once a year judgment was held, and the day of it proclaimed the day after Martinmas,'^ i.e., at the beginning of that year economic.
In medieval Tirol Martinmas began the business year for which all officials
Walther Swarzenberg, Heinrich vom Rijne und andere des rades frunde umb siner sunderlichen begerunge willen zu Pressburg zuvor bezalten und ussrichten uff sine quer- tancil. "
^ Bedebuch of Frankfurt a. M. of A.T>. 1476, Ob. 19'': "6ss. von siner swieger seligen wegen, die nach sant Mertins dage von dodes wegen abegegangen ist."
^Karl Blicher, Die Bevdlkerung von Frankfurt a. M., I., p. 263.
^ Ibid., I., p. 354, from the Citizettbook of 1378. He who does not pay his citizen money, "sal geben zuschen hie vnd sant Mertins dag neist kommet 10 lb. und 4ss. hell; wo he det nit entede, so mochte man sie uff in zun juden uff sinen schaden nemen ; " Ibid., I., p. 485, A.D. 1372, the inhabitants of the villages which are under the protectorate of the city have to give "eyme schultheizsen eynen schilling phennige vnd ein hun uff sant Martins dag"; Ibid., I., p. 486, A.D. 1383, "Und sal . . . dem schultheiszen sin recht uff sant Mertins dag."
■* Karl Biicher, Zwei mittelalterliche Steuerordnungen in Kleinere Beitrdge zur Geschichte von Dozenten der Leipziger Hochschule, Leipzig, 1894, p. 139.
^ Ibid., p. 150 : The duties are to be raised (a.d. 1474) " eyn iglich der nehst komenden drj' jare zu heben und zu sant Mertins dag schirst kommend anetzufahen. "
^Biicher, Bevdlkerung von Frankfurt a. M., I., p. 370.
'' Annalen fUr Rheinische Geschichtskunde, Vol. LXII., p. 39.
40
YULE AND CHRISTMAS
were elected,^ and for which duties and interests were paid.^ If anybody's property was burnt before St. Martin's day, he needed to pay no duties that year.3 Besides, Martinmas began the rustic winter in Tirol so late as the fifteenth and seventeenth centuries.* When drawing conclusions from his discussion of popular Martinmas, Heino Pfannenschmid, who firmly believes that the Germanic year began with the winter solstice, arrives at the result^ "that Martinmas forms the conclusion of the peasant agricultural year. This is also shown by the fact that most leases end about Martinmas, when the rent has to be paid, which might have been hard for many a one, though the harvest had been reaped and turned into money. To this refer the sayings, ' Martin is a severe man,' and ' Martin is a bad man.'^ So other payments are made, and accounts handed in, at Martinmas;'^ all sorts of duties in kind and money are paid to monasteries, churches, parishes ; church accounts are made up and paid. The conclusion of the old agricultural and crop year on Martinmas is finally marked by the changing of servants. With this end of the old agricultural and crop year a new one began. Then the new lease year begins both in Germany and England ; new servants are hired. ... In France Martinmas was considered the beginning of winter
^Zingerle, Tiroler IVeisiiimer, II., p. 173, a.d. 1580, of Nassereit and Torminz, Upper Engadine : '* Sollen drei erbare verstendige mannspersonen . . . jarlich an sanct Martins des heiligen bischofs tag . . . zu gwalthabern und dreierern . . . furgenomen, erwolt und erkiest warden;" Ibid., III., p. 258, A.D. 1607, of Latsch, Vintschgau : " Ein feldsaltner soil, wo es kann, auch am kassuntag, wo nit doch neyst darnach angenomen und seinen dienst, als hernach folgt unzt auf Martini zu verrichten schuldig sein ; " Ibid., III., p. 193, A.D. 1614, of Kortsch, Vintschgau : " Der messner allhier hat auch jahrlich am st Martinstag urlaub, und soil auch jahrlich vor der ganzen gemeinde stehen und um solches amt bitten ; " Ibid., II., p. 168, A.D. 1674; I., p. 79, A.D. 1727.
"^ Ibid., II., p. 310, A.D. 1303 ; II., p. 104, A.D. 1416; III., p. 351, A.D. 1427.
^ Ibid., III., p. 7, A.D. 1440, of Glurns, Vintschgau : " Und ob dann ainer verprent wurd oder verprunne vor sand Martins tag, derselb sol umb dieselben zins desselben jars ledig und los sein." A long list of other cases is given in the apparatus to my Geschichte der deutschen Weihnacht, p, 293.
^Zingerle, Tiroler Weistiiiner, IV., p. 33, a.d. 143 i, of Partschins ; Ibid., III., p. 65, A.D. 1630, of Burgeis.
* Germanische Erntefeste, Hannover, 1878, p. 237.
^ Nork, Festkalender, p. 683, and Simrock, Martins Lieder, xv.
' Leoprechting, Aus dent Leckrain, p. 200; Birlinger, Aus Schwaben, II., 132.
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 41
and of the new year, the legend of St. Martin explaining the latter fact as a proof of the high esteem and reverence in which that Saint was held. But in Germany also the time about Martinmas must have been considered as, in a certain respect, the beginning of the year. As the day began with the eve, so for the ancient Germans the new year could begin with the beginning of winter at Martinmas. After the time of Gregory of Tours, even a new era was computed from St. Martin's death, just as an era was computed from Christ's death." ^ Montanus^ noted that formerly every- where, and in his time still on the left bank of the Rhine, the lease year and agricultural year closed with Martinmas, to which Pfannenschmid' added that that was also the case in Lower Saxony and in other provinces of Germany. The Church Ordinance of Hoya of 1573* fixed Martinmas as the date for the elders of the church to lay the accounts before the officials in presence of the minister. Pfannenschmid gives a long list of facts in support of Martinmas being an old term. Male and female farm servants changed their places at Martinmas.^ About Martinmas, at Seelze near Hannover, male farm servants changed.^ Elsewhere female farm servants changed,^ and in other regions all servants did.^ In the Havel country it was so till a short time ago. Now they change at Christmas.^ Also the lease year began at Martinmas in Germany and England.^" Even Grotefend (who has been completely led astray by Weinhold's theories, according to which the Germanic year was divided by solstices and equinoxes) has to confess" that among the country folk in various districts Martinmas means the beginning of winter, where a dual division of the year prevails.
^ Pfannenschmid, Germanische Erntefeste, p. 511.
^ Volksfeste, I., 55. ^ Germanische Erntefeste, p. 511.
^Richter, H., 359; Pfannenschmid, Germanische Erntefeste, p. 511.
^Schambach, IVorterbuch, 131. ^Waldmann, Eichsf elder Gebrduche, 15.
^Danneil, Altmdrkisckes IVorterbuch, 132.
^Kuhn und Schwartz, Norddeutsche Sagen, H., 401 ; Birlinger, Aus Schwabeti, H., 132.
^Ibid.
^"Nork, Festkalender, p. 683; Simrock, Mythologie, 574; Miilhause, Urreligion, p. 308; Rochholz, Wandelkirchen, p. 14.
^^ " Martinstag ist bei der Zweiteilung des Jahres provinciell als Anfang des Winters gebrauchlich " (Zeitrechnung, I., 1 19).
42 YULE AND CHRISTMAS
In German folk-belief there is still a faint recollection of Martinmas being the old Germanic New Year. It is in some phrases still used as identical with year, or with winter, as the phrase went previously. Instead of: "a man has lived through many years," the folk say : The man has helped to eat many a St. Martin's goose.^ The old hexameter,
" Iss Ganss Martini, trink Wein ad circulum anni,"
also alludes to Martinmas as the beginning of the year, when people drank good luck to a new annual course.
Beside the legal institution of termly duties stands that of regular assemblies, the so-called not-ordered law courts. It was shown before that there existed among the Germanics sometimes three (originally held at mid- November, mid-March, and mid-July),^ sometimes two (originally held at mid-November and mid-May), according to a dual division or a tri-partition of the year. It is a fact of great weight that they adhere to these terms, even when one or two of them disappear.
The only foreign impost imposed by the Church upon her believers was the Peter's penny. It was by a perfectly voluntary act that the payment of it was fixed at St. Peter's mass. Thus it has nothing to do with ancient Germanic institutions. Its name Rome-feoh on British ground showed the foreign origin and purpose of it only too clearly. Keeping this in mind, we cannot fail to see that the ordinary taxes as well as duties like light-scot
^Grimm's Wdrterbuch, IV., i, 1262. The explanation of that phrase, given there by Rudolf Hildebrand, is wrong. The time from one St. Martin's goose to the next is properly regarded as a complete year.
^A curious interpretation of the word Trithinga, which shows that its meaning had been forgotten entirely, is given by Fleta seu Commentarius Juris Anglicani sic nuncupatus, sub Edwardo Re^e primo seu circa annos abhinc CCCXL. ah Anonymo conscriptus, atque e Codice veteri, autore ipso aliquantulwn recentiori, nunc primum typis editus, Londini, 1647, p. 134, Lib. II., chap. Ixi., § 23: " Ue tritingis, sciendum quod aliae potestates erant super wapentakia quae tritinga dicebantur, eo quod erat tertia pars provinciae ; quia vero super eos dominabantur, trithingreves vocabantur, quibus differebantur causae quae non in wapentakes poterint diffiniri in Schiram, sicque quod Anglici vocant hundredos, jam, per variationem locorum et idiomatis, wapentakia appellantur, et tria, vel quatuor, vel plura hundredi solebant Trithinga vocari ; et quod in Trithingis non potuit diffiniri, in Schiram id est, in comitatum differebatur terminandum : modernis autem temporibus pro uno eodemque habentur apud homines hundredi, wakentakia, et Trithinga."
UNIVERSITY
Of
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 43
had to be paid three times a year, though their exact terms vary a little in the several districts. Under King Ethelred (991-10 16) plough-alms were to be paid xv. days after Easter, and a tithe of young by Pentecost, and of earth fruits by All-hallows' mass (and Rome-feoh by St. Peter's mass), and light-scot thrice in the year,^ one of the days of payment for them being Candlemas.^
In the Rectitudo Ancillae^ mid-Lent is named as the second of the three terms of the year, the term used being to losngten-sufle and ad quadragesi-
^Thorpe's Ancient Laws, I., p. 306: "xi. And gelaeste man Codes gerihta georne EEghwylce geare. that is. sulh-selmessan xv. niht on ufan Eastran. and geogodhe teodhunge be Pentecosten. and eordh-wsestma be Ealra Halgena mgessan. and Rom-feoh be Petres msessan. and le6ht-gescot thriwa on geare"; and Ibid., I., p. 318, in the resolutions passed at the Council of Enham : "xvi. And gelaeste man Codes gerihta. s^hwilce geare rihtlice georne. that is. sulh-aslmessan huru XX. niht ofer Eastron : xvii. And ge6godhe teodhunge be Pentecosten. and eordh-waestma be Ealra Halgena msessan : xviii. And Rom-fe6h be Petres msessan. and ciric-sceat t6 Martinus msessan : xix. And leoht-gescot thriwa on geare"; Ibid., I., 338, under King Ethelred: " iv. Et praecipimus, ut omnis homo, super dilectionem Dei et omnium sanctorum, det cyricsceattum et rectam decimam suam, sicut in diebus antecessorum nostrorum stetit, quando melius stetit ; hoc est, sicut aratrum peragrabit decimam acram. "
"^Ibid., I., p. 342 (also under King Ethelred): "ix. And si selc geogudhe teodhung gelsest be Pentecosten be wite. and eordh-wsestma be emnihte. oththe huru be Ealra Halgena msessan." {Here the equinox was probably substituted for the older All-Hallows term.) "x. And Rom-fe6h gelaeste man seghwilce geare be Petres maessan. and sethe that nelle gelaestan sylle thar-t6-eacan. xxx. peninga. and gilde tham cyninge cxx. scillingas. xi. And ciric-sceat gelaeste man be Martinus-maessan. and sethe that ne gelaeste for gilde hine mid twelffealdan. and tham cyninge cxx. scillingas. xii. Sulh- aelmessan gebyredh that man gelaeste be wite aeghwilce geare. thonne XV. niht beodh agan ofer Easter-tid. and leoht-gescot gelaeste man to Candel-maessan. d6 oftor sethe wille." The same is ordained in the Laws of King Cnut, 1017-1042 (Ibid. I., 366 : " viii. And gelaest man Codes gerihta aeghwylce geare rihtlice georne. that is. sulh- selmesse huru fiftene niht ofer Eastran. and geogudhe teodhunge be Pentecosten. and eordh-waestma be Ealra Halgena maessan. ix. De Nummo Romano. And Rom-fe6h be Petres maessan. x. De Primitiis Seminum. And cyric-sceat to Martines maessan. xii. De Pecunia Pro Lucernis. And leoht-gesceot. thrtwa on geare. aerest on Easter- aefen healf-penig-wurdh wexes aet aelcere hide, and eft on Ealra Halgena maessan eall swa mycel. and eft to thaem Sanctan Mariam claensunge eal swa" (Feb. 2); Ibid., I., 434-35 : " Et det suum cyric-sceatum in festo Sancti Martini (and sylle his cyric-sceat to Martinus maessan "), {Rectitudines Singularum Personarum).
'Thorpe's Ancient Laws, I., 436-7.
44
YULE AND CHRISTMAS
malem convictum. The fourteenth century Rhenish Urbary of St. Victor, Xanten,! gives St. Gertrudis Day (March 17) as the spring term, whilst Tirol documents of the fifteenth century call it simply mitte merzenf whilst, two centuries later,^ summer was reckoned to begin on April 23,* the beginning of winter being in both cases Martinmas. In Flanders St. Gertrudis Day (March 17) is called Sommer tag/^ Just as the mid-May term, which halved the year beginning at Martinmas, was easily replaced by the Rogation Days, and afterwards by Pentecost, so the nearness of the Christian festival of Easter could scarcely fail to become detrimental to a mid-Lent term, or rather to a mid-March term. The earliest date on which Easter could fall was March 22, a date only a week distant from March 15. Grimm has shown ^ that the three old Germanic offering-tides coincided with the Thing tides, nay, represented one side of them. There is no doubt that the autumn Thing and the spring Thing were the most important, while the summer Thing could not be so significant, for the very simple reason that, in time of war, almost all men able to bear arms were away. A festival about the middle of July I should be the last to deny to Ger- manics, Slavs, and Celts, but that it had any early relation to a summer solstice which fell about three weeks earlier must be most emphatically gainsaid. It is true the festivals which appear in medieval poetry are almost all celebrated either at Pentecost or ze einer sunnenwenden (about June 24). But to that date the festival of the beginning of late summer at mid-July had been shifted in the tenth or eleventh century, whilst the legal term in many places remained where it had been. As the expression ze einer sunnenwenden is foreign, so the date itself is foreign. Were it otherwise, and had June 24 been an old Germanic date (so that the other two had originally fallen on October 24 and February 24), it would be quite inexplicable how the term could have moved away from it to mid-July, which was equally out of keeping with October i, with Christmas, and with
^ State- Archive, Diisseldorf, under " Stift Xanten," R. No. 8% leaf 8*.
^Zingerle, Tiroler Weistumer, IV., p. 33, A.D. 1431, at Partschins.
^A.D. 1630, at Burgeis. ^Zingerle, Tiroler Weistiimer, III., p. 65.
"Grotefend, Zeitrechnung des deutschen Miitelalters, I., 178.
^Deutsche Rechtsaltertumer, 821 ss. 245, 745.
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 45
Easter. Besides, there is no trace of sun-worship whatever in Germanic religion, 1 and in the Scandinavian North the existence of a midsummer Thing is as well vouched as any fact of Old Norse history, without it showing the slightest trace of a relation to a sun-cult. Quite in accordance with the fact that the Scandinavian year began between October 9 and 14, and had a Gbiblbt between February 9 and 14, the summer Thing was held between June 9 and 14. In Germany, where the terms of both the beginning of winter and of the beginning of early summer fell one month later, the festival at the beginning of later summer must have been held about July 15. In the Rhenish Urbary of St. Victor, Xanten,^ July 12 (St. Margaret's day) was marked, not only as the beginning of a new season, but even as the beginning of a new year. At other places July 1 7 appears as the joyous day. So in Swabia children and gilds received gifts — the latter of wine.^ It is also the day of old bonfires in Villingen, Swabia."* In the Netherlands the old season of four months, from March 15 to July 15, was till very late called May,^ just as in Germany a Roman quarter of a year was taken as identical with spring, and counted from February 22 to May 25.^ That that popular summer festival had originally nothing to do with a solstice appears from the fact that even July 25 was called te midzomer (1419) or na midden-somere (1351).^ In 1461 it was still taken as corresponding to Christmas, and was the occasion of a local fair like that festival.^ It was still regarded also as one summer term in the Tirol of the sixteenth century. It divided the time of pasture into two halves, and he who, at Laatsch, Tirol, in 1546, bought oxen after that term, and put them
^ Holtzmann, Germanische AltertUmer, Leipzig, 1873, PP- 127 and 173; Grimm, Deutsche Mythologie, 591.
2 State-Archive, Dusseldorf, under *' Stifc Xanten," R. No. 8% leaf 8*.
^Birlinger, Aus Schwaben, H., p. 1 18.
*Ibid.; and Mone, Quellensammhmg, II., 88". ~"
' Tijdschrift v. nederl. Taalk, IX., 134 ; Grotefend, Zeiirecknung, I., 1 16.
^ Baltische Studien, XIX., 49: "de Mey beginnet in sunte Peters daghe, de summer in sunte Urbans daghe" (Grotefend, Zeiirecknung, I., 116).
■^Grotefend, Zeiirecknung, I., 87.
^ Annalen des historischen Vereins fiir den Niederrkein, Vol. LXV., p. 42 (Town-Archive of Kempen, Docs. Nos. 367 and 387). It there appears as St. Jacob's day.
46 YULE AND CHRISTMAS
on the common pasture ground of the community, had to pay to the community four Bernese pounds for each pair,^ whilst, a century later (in 1647), the term had been shifted to St. Vitus Day, June 15.2 In the Anglo- Saxon Rectitiido Ancillae^ this term is called on sumera or in aestate (an expression corresponding to Old Icelandic at sumri, i.e., June 9 to 14), meaning a day about mid-July. When it had been shifted to June 24, it was called midsummer. Midsummer was a later Anglo-Saxon term recognized by law : " A sheep shall go with its fleece until Midsummer, or let the fleece be paid for with two pence," is a doom in King Ine's Laws.* But occasionally the regular law courts and assemblies were held still later. More than once the day of the beheading of St. John the Baptist (August 29) is fixed as the term at which the bishops and reeves are to adjudge the king's commands to all whom it behoveth.^
The tri-partition of the year — Martinmas, mid-March, mid-July — was, till late in the Middle Ages, more than an artificial division of the year carried on by tradition without apparent reason. It was deeply rooted in economic life, and in conditions affecting pasture and agriculture. In connection with the keeping of domestic animals, as well as with the ploughing of the fields, traces of an old tri-partition have come down to us. The anonymous Anglo- French Seneschaucie of the thirteenth century*' ordains : "The bailiff" ought, after St. John's Day, to cause all the old and feeble oxen with bad teeth to be drafted out, and all the old cows, and the weak and the barren, and the young avers that will not grow to good, and put them in good pasture to fatten, so the worst shall then be worth a better. And he ought, three times a year, to cause all the sheep in his charge to be inspected by men who know their business — that is, to wit, after Easter, because of the disease of May, and later, for then sheep die and perish by the disease; and all
^Zingerle, Tiroler Weistiimer, Vol. III., p. 103.
2 In the neighbouring community of Schleiss (Zingerle, Tiroler Weistiimer, Vol. III., p. 89). ^Thorpe's Ancient Laws, I., 436-7.
* Thorpe, Ancient Laws, I., p. 146: "odh midne sumor."
'^ Laws of King Athelstan, I., Thorpe's Ancient Laws, I., p. 194: "ond thses sie to thsem dseg thser beheafdunges Seint lohannes thaes fulhteres."
^Ed. by Elizabeth Lamond in Walter of Henley's Husbandry, London, 1890, p. 97.
MARTINMAS, AND THE TRI-PARTITION OF THE YEAR 47
that are found so, by the sure proof of killing two or three of the best, and as many of the middling, and as many of the worst, or by proof of the eye or of the wool, which separates from the skin, let them be sold with all the wool. And again, let all the old and weak be drafted out before Lammas, and let them be put in good pasture to fatten, and when the best have presently mended and are fat, let them be sold to the butchers; so can one do well, for mutton flesh is more sought after and sold then than after August; and let all the rest of the draft beasts which cannot be sold then be sold before Martinmas. And the third time, at Michaelmas, let all the sheep be drafted out."^ In the same century elsewhere the day of testing the health of wethers was October 28. Walter of Henley, in his Husbandry^ laid it down that two of the best wethers, two of the middling, and two of the worst should be killed on that day. If they were found not to be sound, a part was to be sold by true men for good security, until Hockday (Thursday after Easter), and then replaced.^ Almost the same way of denoting a third of a year is found in connection with sheep-keeping ; from Martinmas to Pasch sheep were to spend their nights under shelter.^
In agricultural life in the middle of the thirteenth century the three terms were known in low Latinity as Hibernagium, Tratmesium, and Warectum. They then continued to determine the ploughing times, which seem to have occupied the latter half of each of the old three seasons.'*
^"E tote le remeignant de creim ke ne put estre uendu adonkes seit vendu deuant la seint martin."
2 Walter of Henley's Husbandry, ed. by Elizabeth Lamond, London, 1890, p. 33 : "A la seynt symon e seyn lude facet tuer deus de meylurs e deus de myuueyns e deus de pyres e si vos trouet ke eus ne seyent mye seyens fetes vendre vne partye a lele genz par bone surte iekes a la hokeday e done fetes releuer autres."
^ Fleta seu Commetitarius Juris Anglicani, London, 1647, p. 167, Lib. IL, chap. Ixxix., § 7 : " Inter festa autem sancti Martini et Paschae, infra domum oves expedit noctanter custodire, nisi terra sicca fuerit, ovileque bene reparatum, tempusque serenum."
* Registrum sive Liber Irrotularius et Consuetudinarius Prioratus Beatae Mariae Wigorniensis : with an Introduction, Notes, and Illustrations by William Hale Hale, Londini, Sumptibus Societatis Camdensis, 1865 ; Redditus Prioratus Wigomiae Anno Incarnationis Domini MCCXL., p. 14'': " Praeterea arabit ad yvemagium, tramesium, et ad warrectum, per unum diem, excepto opere, et vocatur * benherthe.' " Ibid., p. 14'': "Praeterea arabit et herciabit I. die ad yvemagium ; et Prior inveniet semen, et si necesse
48 YULE AND CPIRISTMAS
fuerit quaelibet herciabit pro opere, donee perventum fuerit ad carueas, praeterea arabit uno die ad tramesium, et uno die warectabit," etc., ut supra. Ibid., p. iS*" : "Nova assisa de vilenagio de Mora. In hoc manerio sunt xxvil. dimidiae virgatae Quarum quaelibet ad firmam posita reddit ad quemlibet terminum ll°^ solidos et in Purificatione I quarter. avenae. Quaelibet etiam debet x'^''"^ summagia apud Wygorniam et terram arare sicut sibi arat ; scilicet semel ad yvernagium et ad tratmesium et ad warectum et debent sarclare et metere et intassare una cum cottariorum operibus et aliorum in autumpno totum bladum de dominico, et debent Thac et Thol et pannagium et gersummationem prolis et hujusmodi." Ibid., p. \<^ '■ " Et dat auxilium, scilicet xviil. denarios Et in Purificatione dimidiam quarterium avenae et facit III" aniras scilicet ad yvernagium ad tratmesium et ad Warectum et iii^= Benrip."
CHAPTER V.
MARTINMAS, AND THE DUAL DIVISION OF THE YEAR.
In Germany mid-May and Martinmas appear as the two half-yearly terms as late as 1525. In that year the peasants of his district reproached the Count of Fiirstenberg for having increased the taxes, the same being then raised twice a year as May tax and autumn tax. In reality these taxes had, as is shown by the Fiirstenberg Urbaria, been raised almost regularly through the whole of the fifteenth century. The holders petitioned the Count to reduce these two taxes to one, which should be paid at Martinmas.^ In the Exchequer Rolls of Scotland ^ almost all accounts of provosts of burghs are dated Whitsunday and Martinmas, while the dates of the custumars' accounts vary considerably. Even accounts "for four years ending Martinmas, 1331 " occur. Martinmas appears considerably oftener than Whitsunday, thus being shown to have been the more important term, at which not only half a year but a whole year ended. In the second volume, which covers the years 1359 to 1379, Whitsunday and Martinmas are also the two main terms occurring in accounts.^
^ The application of the small-holders is printed by Baumann, Akten zur Geschichte des deutschen Baiiernkrieges aus Obersckioaben, Freiburg i. B., 1877, ^"^^ is commented upon by Hossler, Zur Entstehungsgeschichte des Bauernkrieges in Siidwestdeutschlattd, Leipzig Dissertation, 1895, p. 49.
^ Rotuli Scaccarii Regum Scotortan, edited by John Stuart and George Burnett, Vol. I. , 1 264- 1 359, Edinburgh, 1878.
^On pages viii., x., xi., xii., xv., xvii. to xx., xxiii. to xxv., long lists of accounts, beginning and ending at those days, are enumerated, just as in the contents of Vol. III.,
D
5°
YULE AND CHRISTMAS
In England this state of things was codified by Edward the Confessor ( 1 042-1 066), perhaps with a slight alteration of the existing usage. Whilst mid-May or Rogation Days as a rule appear as the old legal term, Edward ordained that the great assembly of his people was to take place on May i.^ As regards the Franks^ it has been finally proved that their great annual assembly took place in the middle of May. Can the name of this assembly Campus Martins (to be translated May-field) suggest that previously, when the tri-partition prevailed and there were three such meetings, the most important of them was held in the middle of March, and thus has nothing to do with Afars = Ziu} This institution extended all over German soil as far as to the Italian frontier. As late as 1281 the community of Fleims, in the secular territory of Trient, reserved to itself the right of keeping two placita — the one in May and the other in November.^ It took a very long time to uproot this institution, and replace it by meetings held according to Roman quarters of years and Christian high festivals.
Legal institutions were not the only form in which the Germanic terms survived. Very early the Christian Church in Gaul was compelled to make considerable concessions to the Germanic mode of dividing the year, which,
which covers the time from 1379-1406. The most frequent phrases of Vol. II. are: "de termino heati Martini ultimo preterito" (p. 475); "de termino Sancti Martini" (p. 621 twice; p. 281) ; "de duobus terminis huius compoti, videlicet Pentecostes et Sancti Martini" (pp. 72, 73) ; " de termino Pentecostes ultimo preterito " (pp. 72, 73) ; " de terminis Pentecostes et Sancti Martini" (pp. 72, 73 twice); "de eodem termino Pentecostes" (pp. 72, 73 twice).
^ In capite Calendarum Maji, Hampson, II., 94, Grotefend, Zeitrechntmg, I., 20*. The spring term appearing probably with most frequency in other connections is the Rogation Days. "And we ordain that every 'burh' be repaired xiv. days over Rogation Days." Laws of King Athelstan, Thorpe's Ancient Laws, I., p. 206, No. 13: "xilll. niht ofer Gang-dagas." "And every man that will may make 'bot' for every theft with the accuser, without any kind of ' wite,' until Rogation days ; and be it after that as it was before." Ibid., p. 222, Laws of King Athelstan, iii. : "odh Gong-dagas."
^ H. L. Ahrens, Uber Namen und Zeit des campus Martins der alien Franken, Hannover, 1872.
'^Pontes Keriwi Austriacarum, Second Division, Diplomata, Vol. V., No. 212, p. 417: " Allegando, quod ipsi homines et comniunitas de Flemmis sicut de jure et ex antiquo est observatum, nisi bis in anno quolibet non debeant conveniri in foro temporali et juri parere in civilibus et sub judicio esse, videlicet ad placitum in festo s. Martini et in placito in Majo. . . ."
MARTINMAS, AND THE DUAL DIVISION OF THE YEAR 51
about the year 500 a.d., found expression in processions and litanies at the two terms of mid-May and mid-November. Two closely corresponding^ Church celebrations were held at these tides, so that their relationship cannot fail to appear. In 511 a.d. the Council of Orleans instituted the so-called Rogations before the Ascension day,* and decreed a three days' liberation from all work for servants of both sexes. Because of a suspicion that the clergy might try to ignore this new institution, as too great a concession to the Germanic field processions about mid-May, a special canon was added, threatening them with punishment in case of non-com- pliance with the command of the Church. Bye and bye some change took place in the date of the praying processions round the fields, further concessions being made to the Germanic celebration, or uoba, in the corresponding tide of the year in November. The Synod of Gerunda of June 8, 517, at which six bishops and one archbishop were present, ordained in its second and third canons that litanies and fasts should be held in the weeks subsequent to Pentecost and to the Calends of November respectively,^ so that the latter were held between November i and November 9. That this was a permanent institution appears from Canon VI. of the second Synod of Lyon in 567,^ from which we also learn that, between 517 and
^Acta Conciliorum, Parisiis, 1714, Vol. II., col. 1011-12; Concilium Aurelianense, I., A.D. 511, Canon xxvii. : " Rogationes, id est, litanias ante Ascensionem Domini ab omnibus ecclesiis placuit celebrari ; ita ut pi-aemissum triduanum jejunium in Dominicae Ascensionis festivitate solvatur : per quod triduum ser\i et ancillae ab omni opere relaxentur, quo magis plebs universa conveniat: quo triduo omnes abstineant, et quadragesimalibus cibis utanlur." Canon xxviii. : "Clerici vero qui ad hoc opus sanctum adesse contempserint, secundum arbitrium episcopi ecclesiae suscipiant disciplinam."
^Acta Conciliorum, Parisiis, 1714, Vol. II., col. 1043; Concilium Gerundense, A.D. 517, ii. : "De litania, ut expleta solemnitate Pentecostes, sequens septimana, a quinta feria usque in sabbatum, per hoc triduum abstinentia celebretur. iii. Item secundae litaniae faciendae sunt Kalendis Novembris, ea tamen conditione servata, ut si iisdem diebus Dominica intercesserit, in alia hebdomada, secundum prions abstinentiae observantiam, a quinta feria incipiantur, et in sabbato vespere missi facta finiantur. Quibus tamen diebus a carnibus et a vino abstinendum decrevimus."
^Acta Conciliorum, Vol. III., col. 355; Concilium Lugdunense, II., A.D. 567, \A.'. "Placuit enim universis fratribus, ut in prima hebdomada noni mensis, hoc est, ante diem Dominicam, quae prima in ipso mense illuxerit, litaniae, sicut ante Ascensionem Domini sancti patres fieri decreverunt, deinceps ab omnibus ecclesiis, seu parochiis celebrentur. "
52 YULE AND CHRISTMAS
567 A.D., the Church had returned to the older date of the spring litany, to the days preceding Ascension day, which apparently almost coincided with the ancient Germanic May term. The autumn litany was then ordained to be held before November 7 — which implied almost no alteration.^ In 517 there existed neither a forty days' fasting-tide from Martinmas to Christmas nor even Martinmas itself. Martin of Tours had died in 401, but he was not made St. Martin till long after. His feast appears first in a Sacra- mentarium by Pope Gelasius I. (492-496), and in the Liber Sacramentorum of Gregory the Great (590-604) ;2 but it was only Pope Martin I. (649-654) who made it a great Church festival.^ So the excuse is not possible that the praying procession was instituted as a preparation for St. Martinmas, or the fasting-tide beginning with it, as has been suggested by a man so ingenious as Heino Pfannenschmid.^ For, according to his own words, the first certain traces of an Advent-tide are found in some homilies,^ probably written by Caesarius of Aries, who died in 542. Yet even there it is only the question of a general preparation for Christmas, and by no means of a prevailing custom or an ecclesiastical statute.*' According to Gregory of Tours, '^ who died in 595, it was Bishop Perpetuus of Tours (who died as Bishop of Toulouse in 506) who ordained for his diocese a fast of three days a week, from St. Martin's burial day till Christmas.^ From the bishopric of Tours the habit of keeping an Advent- tide seems then to have spread over the whole territory of the Church.
First the new fast-tide referred to monks only. In the second Synod of Tours in 567 (where nine bishops were gathered, and among them those of Tours, Rouen, and Paris), with the consent of King Charibert, a daily
^ About no7ius mens = Nov ember, comp. Eccard, Co?nmenlarius de Rebus Franciae, I., 131.
"^ Pfannenschmid, Gennanische Erntefeste, p. 464.
3 Wandalbert, Martyrologium in d' Archery, Specilegium veterum Scriptorum, T. II.; Pfannenschmid, Ibid., p. 465.
^ Gennanische Erntefeste, p. 515.
^ In Appendix Augustianus, Tom. V. Operum St. Atigustini, nova edit.. No. 1 15 et 1 16.
SBinterim, Denkwurdigkeiten, V., L, 164. ''Lib. X., vYwA, chap. xxxi.
8" A depositione domini Martini usque ad Natale domini."
MARTINMAS, AND THE DUAL DIVISION OF THE YEAR 53
fast was decreed for monks during December up till Christmas day, a period bye and bye extended to forty days, and made applicable to laymen also. The first testimony as to a general celebration of an Advent-tide is an ordinance on fasting for laymen given by the first Synod of Macon, which was called in 581 by the Frankish king, Guntram, and was attended by twenty-one bishops from various provinces of the Church, among others by the bishops of Lyon, Vienne, Sens, and Bourges. Its Canon IX. runs as follows : " From St. Martin's day till Christmas every Monday, Wednesday, and Friday is to be a day of fast."^ At the end of the sixth century Rome took over that forty-day fast-tide preceding Christmas, and in the seventh century it was kept all over Italy, Spain, and England. In Germany it was ordained by the Synods of Aix-la-Chapelle in 836, and at Erfurt in 932. The period of fast as well as its strictness varied, however, considerably. Finally, the whole fortnight preceding Christmas was declared a continuous fast-tide, and the week preceding Christmas a time void of any legal process. In 1022 it was even decreed that from the beginning of Advent till Epiphany nobody was to marry.^
In the beginning of the seventh century the Calends of November received a new ecclesiastical significance. About the year 608 the Pantheon of Rome, which until then had been devoted to the service of all Roman gods, was by Boniface IV. dedicated in honour of "the holy Mother of God, and of all Saints;" and it was ordained that a com- memoration of them should be observed during the Kalends of November.^ This feast was received through all Gaul by the authority of the Emperor, Louis the Pious (a.d. 835).* In 694 the seventeenth Council of Toledo extended the litanies, which so far had been held twice a year — in May
' " Ut a feria sancti Martini usque ad Natale Domini, secunda, quarta, et sexta sabbali jejunetur, et sacrificia quadragesimali debeant ordine celebrari " {Ada Condliorum, Parisiis, 1714, Vol. III., col. 451).
'^ Pfannenschmid, Germanische Ertitefeste, p. 514, Hefele, IV.. 565, 564, 640.
' Alcuin, De Divitto Officio.
* Sigeberti Gemblacensis, Chronicon ab antio 381 ad 1113, under a.d. 835: "Monente Gregorio papa et omnibus episcopis assentientibus Ludouicus imperator statuit ut in Gallia et Germania festiuitas omnium sanctorum in Calen. Nouemb. celebraretur, quam Romani ex instituto Bonifacij papae celebrant."
54
YULE AND CHRISTMAS
and in November — over the whole twelve months of the year, at the same time specifying the reasons for the change.^
If proof were requisite that the Rogation Days took the place of a most important Germanic festival, it would be supplied by the descriptfon of their celebration given by the Council of Cloveshou, II., a.d. 747, in the canons of which games, horseraces, and extensive dinners were named as the characteristics of that tide.^ In the ninth and tenth centuries these regulations were more than once repeated by Councils and Synods.^
^ Acta Conciliorum, Parisiis, 1714, Vol. III., col. 1815, Concilium Toletanicm,'XN\\., vi. : " Quamquam priscorum patrum institutio, per totum annum, per singulorum mensium cursum, litaniarum vota decreverit persolvendum, nee tamen specialiter sanxerit pro quibus causis idipsum sit peragendum : tamen, quia cooperante humani generis adversario, multa inolevit oberrandi consuetudo, et jurisjurandi transgressio ; ideo secundum evangelistam, qui ait : Vigilate et orate, ne intretis in tentationem ; in commune statuentes decernimus, ut deinceps per totum annum, in cunctis duodecim mensibus, per universas Hispaniae et Galliarum provincias, pro statu ecclesiae Dei, pro incolumitate principis nostri, atque salvatione populi, et indulgentia totius peccati, et a cunctorum fidelium cordibus expulsione diaboli, exhomologeses votis gliscentibus celebrentur : quatenus dum generalem omnipotens Dominus afflictionem perspexerit, et delictis omnibus miseratus indulgeat, et saevientis diaboli incitamenta ab animis omnium procul efficiat."
"^ Acta Conciliorum, Parisiis, 1714, Vol. III., col. 1956, Co7tcilium Cloveshoviae, II., A.D. 747, xvi. : "Sextodecimo condixerunt capitulo, ut litaniae, id est, rogationes, a clero omnique populo his diebus cum magna reverentia agantur : id est, die septimo Kalendarum Maiaium [this date must mean May 7], juxta ritum Romanae ecclesiae : quae et litania major apud earn vocatur. Et item quoque, secundum morem priorum nostrorum, dies ante ascensionem Domini in coelos cum jejunio usque ad horam nonam, et missarum celebratione venerentur ; non admixtis vanitatibus, uti mos est plurimis, vel negligentibus, vel imperitis : id est, in ludis, et equorum cursibus, et epulis majoribus : sed magis cum timore et tremore, signo passionis Christi, nostraeque aeternae redemptionis, et reliquiis sanctorum ejus coram portatis, omnis populus genu flectendo divinam pro delictis humiliter exorat indulgentiam."
^ Acta Conciliorum, Parisiis, 1714, Vol. IV., col. 1014-5, Concilium Moguntiacuin, A.D. 813, xxxiii. : " Placuit nobis, ut litania maior observanda sit a cunctis Christianis diebus tribus, sicut legendo reperimus, et sicut sancti patres nostri instituerunt, non equitando, nee pretiosis vestibus induti, sed discalceati, cinere et cilicio induti, nisi infirmitas impedierit." Ibid., Vol. IV., col. 1395, Conciliutn Aquisgratiense, II., a.d. 836, x. : ''Do Litania quoque maiore atque de Rogationibus ventilatum est : sed communi consensu ab omnibus electum atque decretum, juxta morem Romanum, vil. Kalendas Maii illam celebralionem, secundum consuetudinem nostrae ecclesiae non omittendam." Ibid., Vol. V., col. 456, llerardi Turonensis Capitula, A.D. 858, xciv. : " De Letania Romana vil. Kalendis Maii, ut rememoretur. " xcv. : " De diebus Rogationum, ut reverenter ac studiose absque turpibus
MARTINMAS, AND THE DUAL DIVISION OF THE YEAR
55
Just as the Church sanctified the older Germanic celebration of mid-May and raid-November by special litanies, so it took over the meetings wont to be held at those terms. In 578, at the Synod of Auxerre,* it was decreed that every year priests should meet at mid-May and abbots at November i. When (in 589, at the Council of Toledo) it was resolved that the Synods, instead of meeting twice a year (on mid-May and November i), were to meet only once, November i was fixed for that meeting ^ — a date observed for more than a century.^
Not before a.d. 755 were these terms superseded by March i and October i,* but even after the middle of the ninth century the annual meetings of the priests of the Church were held in the beginning of November.^ In Great Britain the Rogation Days were, under the name of
jocis et verbis celebrentur. Ut nullus in eis prandia, comessationes diversasque potiones per diversa loca facere praesumat." Ibid., Vol. VI., i., col. 606, Concilium Engilenheimense, A.D. 948, vii. : " Ut litania majore jejunium, sicut in Rogationibus ante Ascensionem Domini exerceatur."
^ Acta Conciliorutn, Parisiis, 1714, Vol. III., col. 444, Syttodus Autissiodorensis, Canon vii. : '* Ut medio Maio omnes presbyteri ad synodum in civitatem veniant, et Kalendis Novembris omnes abbates ad concilium conveniant."
"^ Acta Cottciliorum, Parisiis, 1714, Vol. III., col. 482, Concilium Toletanutn, III., A.D. 589, xviii. : " Praecipit haec sancta et veneranda synodus, ut stante priorum auctoritate canonum, quae bis in anno praecipit congregari concilia, consulta itineris longitudine, et paupertate ecclesiarum Hispaniae, semel in anno in locum quem metropolitanus elegerit episcopi congregentur : judices vero locorum, vel actores fiscalium patrimoniorum, ex decreto gloriosissimi domini nostri simul cum sacerdotali concilio, autumnali tempore, die Kalendarum Novembrium in unum conveniant."
^The decree was repeated at another Council of Toledo in 681. Acta Conciliorum, Parisiis, 1 7 14, Vol. III., col. 1725, Concilium Toletanum, XII., A.D. 681, xii. : " Placuit huic venerando concilio, ut juxta priorum canonum instituta, episcopi singularum provinciarum annis singulis in unaquaque provincia Kalendis Novembribus concilium celebraturi conveniant. Quisquis autem in praedictis Kalendis Novembribus pro celebratione synodi venire distulerit, excommunicationi debitae subjacebit."
*^Acta Cottciliorum, Parisiis, 1714, Vol. III., col. 1995, Concilium Veronense, A.D. 755, iv, : " Ut bis in anno synodus fiat. Prima synodus mense primo, quod est Martiis Kalendis, ubicumque domnus rex jusserit, in ejus praesentia. Secunda synodus Kalendis Octobris, aut ad Suessiones, aut alibi, uti in Martiis Kalendis inter ipsos episcopos convenit."
^ Acta Conciliorum, Parisiis, 1 7 14, Vol. V., col. 391, Hincmari Archiepiscopi Remensis Capitula, A.D. 852, i. : "Anno DCCCLii. Kalendis Novembris, conventu habito presbytorum in metropoli civitate Remorum," etc.
56 YULE AND CHRISTMAS
Gangdagas^ as popular as on the cohtinent, they being one of the great tides of the year by which people computed time.
In the Parker ms. (A) of the Saxon Chronicle, in the same year in which Martinmas appears for the first time (a.d. 913), there are also found the Rogation Days : betweox gangdagum and niiddurn sumera. They are held to be fixed on Monday, Tuesday, and Wednesday of the Ascension week, and appear once more in 921 to gangdagum, and in 922 betweox gangdagum and middan sumera.^ In the other group of chronicles represented by the Laud MS. (E), the entry of 913 a.d. is the same, whilst the next two are lacking. Then the Rogation Days appear in 1016.^
The term denoting at once the beginning of the Germanic year, and of the winter season, varies from the Calends of November to mid-November, thus keeping clearly within the time which had to be assumed as the beginning of the old Germanic liuleis tide.^
^ The only similar term other than Martinmas and gangdagas appearing is hlafinassa, the later Lammas ; it is the first of August, Sti Petri ad viiuiila (Augusti), frequently abbreviated as Gula Augusti, which, of course, has nothing to do with gula (the throat or palate), but is merely a mutilation of vhuula, C and g are fiequently interchanged ; comp, Gutnplete for Completa ; Grotefend, Zeitrechnung des deutschen Mittelalters, I., p. 78, where the current mistaken explanation of Gula Augusti is also found. Lammas appears in 921 {betwix hlafmcessan atid middum sumera) ; whilst in the little interpolation of later date in B and C, which is styled by Earle The Atmals of ALthelflccd, it occurs as early as 913 and 917 {thces foran to hlafmassan send, foran to hlcefmassan lespectively). It seems to have been the term dividing the economic summer-tide, instead of July 15.
'^ Hlafmmssa appears in 917, in 1085, in iioo, in iioi, and 1 135, whilst ane dtcge icr sanctes Petres massan afene and on sanctes Pet res mcessa dccg are mentioned in 1048, I131, 1132.
^It is a mere exception when the term is shifted back as far as October 18. In the thirteenth century it was in the south of England usual and right that plough beasts should be in the stall between the feast of St. Luke (October 18) and the feast of the Holy Cross (on May 3), five-and-twenty weeks (Walter of Henley's Husbandry, ed. by Elizabeth Lamond, London, 1890, p. 13 : "Custume est edreyt ke bestes des charues seyent a la creche entre la feste de seynt luc e la feste de la seyt croys en may par vint e cynk semeynes "). At the same time sheep were kept in houses between Martinmas and Easter [Ibid., p. 31 : " Veet ke vos berbyz seyent en mesun entre la seynt martyn e pasche "). Even in these five-and- twenty weeks the wintry half of the year is clearly recognisable.
CHAPTER VI.
MARTINMAS AND MICHAELMAS.
In Professor Weinhold's opinion Michaelmas is an older term and festival than Martinmas — a view not tenable for a moment, as he easily might have seen himself. For, in the text of his book on the division of the German year,i he says that the four not-ordered law courts, mentioned in some legal documents, are Michaelmas or Martinmas, Epiphany, Easter, John Baptist's Day ; and when, in the apparatus,^ he has to give the proof, he instances seven cases, in five of which Martinmas appears, whilst no other occurs so often, and Michaelmas and John Baptist's Day only in two. If any generalisation is to be gathered from these facts, it is that, even when the Roman quartering had superseded the Germanic tri-partition of the year, for a long time Martinmas by far prevailed over Michaelmas. He further talks ^ of all kinds of usages and customs having been transferred from Michaelmas to Martinmas, without giving a single historical instance of such a transference j nay, even without attempting any proof of the assertion that they were found earlier at Michaelmas than at Martinmas. The fact is that, after eighteen hundred years of effort to force upon the Germanics the quartering of the year and a beginning of the winter on September 29, the attempt has succeeded so little, that up to this day Martinmas has in many places, preserved its character as the popular beginning of the winter. Grotefend* shares the view of a shifting of the beginning of winter from
^ Uber die deutsche Jahrteilung, I-Ciel, 1862, p. 10. "^Ibid,, p. 19. "^ Ibid., p. 5.
^ Zeitrechnung des deutschen Mittelalters, 1891, I., 89, JahreszeiUtt.
58 YULE AND CHRISTMAS
September 29 to November ri, and adds to it an imaginary shifting of the beginning of summer from April or Easter to the middle of May, whilst in reality the middle of November and the middle of May are the most ancient Germanic terms, and in Scandinavia (as Weinhold and Grotefend know very well) a shifting by one full month has taken place, so that October 14 and April 14 divide the year.
In the Saxon Chronicle Martinmas appears first in 913, Michaelmas first in 1014; but after 1066 the mentions of the latter quickly outnumber those of the former. From the Parker ms. of the Saxon Chronicle Martinmas appears to have existed long before Michaelmas. We have in 913 ymb Marlines maessan, in 918 and gig foran to Marlines mcessan, in 921 thces ilcan geres foran to Marlines mcsssan, whilst Michaelmas does not occur a single time ; in tlie Laud ms.^ things are a little different. Its oldest part was written in the tenth century, so that it is quite irrelevant that under A.D. 759 appears a solitary to sancle Michaeles tyde. This can only be a dating after a later fashion. Then Martinmas is mentioned under 913, 915, 971 (B), 1009, 1021, 1089, 1097, 1099, HOC, 1 1 14, whilst Michaelmas appears again as late as 1014 (also in ms. C) ; but its occurrences become very frequent after 1066.^
Nobody will deny that Dr. Heino Pfannenschmid, author of Germanische Ernlefeste* is the first authority on everything connected with the festivities held in autumn on Germanic ground. His book, though written more than twenty years ago, is still the best on the subject, and unparalleled by another book on cognate matter. By the most thorough investigation he was led to the conclusion that only very slight traces of a thanksgiving for the corn harvest can be discovered in the later Michaelmas, and that it almost exclusively, in Christian times, bears the character of a celebration for the sake of the dead and of a festival in honour of angels,'^ whilst " an abundance
^ Two of the Saxon Chronicles parallel, ed. by John Earle, Oxford, 1865.
'•They are 1066, 1086, 1089, 1091, 1095, 1097 (twice), 1098, 1099, 1 100, iioi, 1102, 1103, 1 106, 1 1 19, 1125, 1 126 (twice), 1 129 (twice). Asserius, De Rebus Gestis Ailfredi, Afonutnenta Historka Britannica, I. , p. 492, has in venerabilis Martini festivitate as early as between 886 and 893.
■* Hannover, 1878. *Ibid., p. 193.
MARTINMAS AND MICHAELMAS
59
of customs, which point to the ancient heathen autumn-festival celebrated in November, have clung round the festival held in honour of St. Martin." The expression Herbstfeier (autumn-festival) is perhaps the only thing in this statement which might be improved upon ; it should be : Festival of winter's beginning and summer's close.^
Whilst Martinmas can be proved to have been a popular festival in 578, when the banqueting at Martinmas eve was forbidden by the Synod of Auxerre, it was not before the ninth century that the Church made an attempt to give to the end of the third quarter of the Roman year a special importance by a festival — that of St. Michael and of the angels and guardian angels in general — called in Germany Engelweihe or Fest der Engel.^ It was the Council of Mayence of 813 which added that angel-festival to two others (on March 15 and on May 2).^ Round this festival there gathered from that time a number of habits and customs, all of them inaugurated by certain Church practices, but becoming a little more popular with every century, although their popularity cannot, even so late as the seventeenth century, compete with that of Martinmas. Had there been any Germanic festival about that time which the Church thought it worth while to absorb and use for its own purposes, it would long before the ninth century have instituted some saint's day of special prominence in that part of the year. The payment of a tax or duty at Michaelmas cannot be proved before the tenth century, when the Anglo-Saxons paid the fruit-
^ When, in 1893, dealing with this matter in my book, Die Geschkhte der deutscheft Weihtmcht, I did not suppose that any folklorist could be unfamiliar with the results of Dr. Pfannenschmid's book, and, consequently, endeavoured solely to supplement his arguments, instead of restating them and summing them up. But Professor Weinhold seems really to have overlooked them. Otherwise he could no longer be in favour of Michaelmas as the ancient Germanic festival of winter's beginning.
^ II. Pfannenschmid, Germanische Erntefeste, p. 169 ; Ducange, Glossariuin under Festum S. Michaelis: "Est ilia dies, inquit Honorius Augustod., Lib. III., cap. 167, qua populus Christianus cum paganis pugnavit, et victoriam per S. Michaelem Archangelum obtinuit ; Cathiulphi Epistitla ad Carolum Magnum, Vol. II. ; Historia Francontim, p. 667; Beletus, c. cxxix., cliii. ; Durandus, Lib. VII., c. xii."
^ Acta Cotuiliorum, Parisiis, 1714, Vol. IV., col. 1015; Pfannenschmid, Ibid., P- 175-
6o YULE AND CHRISTMAS
tithe to the Church on St. Michael's day/ and lease-rents seem not to have been paid in England at Michaelmas prior to the fifteenth century. The rent day is marked in the fifteenth and sixteenth centuries by the fact that the landlords used to invite their tenants for Michaelmas, a roast goose being the festive dish. 2 Under King Edward IV. (1461-1483), John de la Hay paid to William Barneby of Lastres, in the county of Hereford, among other things as rent for part of his land, a goose for Michaelmas.^ We still are able to trace the way in which the quarterly division of the Roman year was made popular by the Church. In England the four quatembers or ember days were introduced by Gregory the Great ( + 604), in the Prankish empire in the Staluta Bonifacii,^ and emphasised by Charle- magne's Capitulare of 769, chapter xi., and the Synod of Mayence of 813,'^ whilst it is not earlier than about the year 1000 that a fast-tide is brought into connection with Michaelmas.*'
^ Lingard, Altertiimer der angelsdchsischen Kirche, Aus dem Englischen ilbersetzt von Ritter, p. 55.
'^N. Drake, Shakespeare atid his Times, p. 165.
^ Fragmenta Antiquitatis ; Antient Tenures of Land, and Jocular Customs of some Manners, by T[homas] B[lount], London, 1679, p. 8 : "Johannes de la Hay cepit de Will. Barneby Domino de Lastres in Com. Heref. imam parcellam terrae de terris Dominicalibus. Reddend. inde per annum xx. d. et unam Aucam habilem, pro prandio Domini in Festo S. Michaelis Archangeli, Sectam Curiae et alia Servitia inde debita, etc. i. Paying a Goose fit for the Lord's dinner on Michaelmas day."
*"Doceant presbyteri populum quatuor legitima temporum jejunia observare, hoc est in mense Martio, Junio, Septembri et Decembri."
^ Acta Conciliorum, Parisiis, 1714, Vol. IV., col. 1015 ; Pfannenschmid, Germanische Erntefeste, 425.
^ Acta Cotuiliorum, Parisiis, 1714, Vol. VI., i., col. 794 ; Leges Ecclesiasticae Aethelredi regis circa annum 1012 apud ILaba?n conditae, chap. ii. : " De jejunio et feriatione trium dierum ante festum Michaelis," etc. " Et instituimus, ut omnis christianus, qui aetatem habet, jejunet tribus diebus, jejunet in pane, et aqua, et herbis crudis, ante festum Sancti Michaelis. Et omnis homo ad confessionem vadat, et nudis pedibus ad ecclesiam ; et peccatis omnibus abrenunciet emendando et cessando. Et eat omnis presbyter cum populo suo ad processionem tribus diebus nudis pedibus, et super hoc cantet omnis presbyter triginta Missas, et omnis diaconus et clericus triginta psalmos : et apparetur tribus diebus corrodium unuscujusque sine came in cibo et potu, sicut idem comedere deberet, et dividatur hoc totum pauperibus. Et sit omnis servus liber ab opere illis tribus, quo melius jejunare possit : operetur siljimet quod vult. Hi sunt illi tres dies ; dies Lunae, dies
MARTINMAS AND MICHAELMAS 6l
The first time that Michaelmas appears alongside Martinmas is in 813, in the decrees of the Council of Mayence,i where also the birthday of John Baptist (June 24) and the day of Peter and Paul (June 29) appear to mark the ultimate quartering of the year according to solstices and equinoxes. In 858 the list considerably differs from that prevalent before,^ but even then sometimes Michaelmas is not in the list, whilst the Rogation Days are.^
In England St. Michael's day seems not to have taken root much before the end of the tenth century, King Ethelred's Laws being the first collection of institutes to contain an ordinance for keeping it, while so far only the Apostle's days, the Mary's days, and Martinmas had been kept, besides the three great Church festivals of Christmas, Easter, and Pentecost.* Some- times both terms appear together.^ The mentions of Michaelmas became
Martis et dies Mercurii proximi ante festum sancti Michaelis." This ordinance bears the complete stamp of being a mere church invention.
Mf/a Conciliortim, Parisiis, 1714, Vol. IV., col. 1015, Concilium Moguntiacum, A.D. 813, xxxvi. : " Festos dies in anno celebrare sancimus. Hoc est, diem Dominicum Paschae, cum omni honore et sobrietate venerari : simili modo totam hebdomadam illam observari decrevimus. Diem Ascensionis Domini pleniter celebrare. Item Pentecosten similiter ut in Pascha. In natali apostolorum Petri et Pauli diem unum, Nativitatem sancti loannis Baptistae. Assumptionem sanctae Mariae, dedicationem sancti Michaelis, natalem sancti Remigii, sancti Martini, sancti Andreae. In Natali Domini dies quatuor, octavas Domini, Epiphaniam Domini, Purificationem sanctae Mariae. Et illas festivitates martyrum, vel con- fessorum observare decrevimus, quorum in unaquaque parochia sancta corpora requiescunt. Similiter etiam Dedicationem templi."
"^Acta Conciliortitn, Parisiis, 1 7 14, Vol. V., col. 454, Herardi Turonensis Capitula, A.D. 858, Ixi. : " De festivitatibus anni, quae feriari debeant, id est, Natali Domini, sancti Stephani, sancti loannis, et Innocentium, octavas Domini, Epiphania, Purificatione sanctae Mariae, et Assumptione, Ascensione Domini et Pentecoste. Missa sancti loannis Baptistae, Apostolorum Petri et Pauli, sancti Michaelis, atque omnium sanctorum, sancti Martini, et sancti Andreae, et eorum, quorum corpora ac debitae venerationes in locis singulis peraguntur. "
^ Ibid., v., col. 462, Walterii Aurelianensis Capitula, xviii., about A.D. 850.
"^ King Ethelred's Laws (991-1016) in Thorpe's Ancient Laws, I., viii., p. 337, ii. (Acta Conciliorum, Parisiis, 1714, Vol. VI., I., col. 793-4, Leges Ecclesiasticae Aethelredi Regis, ca. a.d. 1012, ii.); and in the concluding passage vii. in p. 339: " Et reddatur pecunia eleemosinae hinc ad festum Sancti Michaelis, si alicubi retro sit, per plenam witam," etc.
'Thorpe, Ancient Laws, I., 479, xxviii. of the Laws of William the Conqueror: " De qualibet hida in hundredo iiii. homines ad stretwarde invenientur a festo Sancti Michaelis, usque ad festum Sancti Martini."
62 YULE AND CHRISTMAS
more frequent after the middle of the eleventh century, the time when the quartering of the year according to Roman custom had been effected all over the realm governed by the Church, From a.d. 877 the term intra tres menses appears in the Councils of the Church,' which shows that quarters of years had become the unities to be reckoned with.
Probably the institution of a quartering term at the division between September and October would not have been so easy had it not had a certain economic basis in an important change which took place contemporaneously. Up to the third and fourth centuries of our era the Germanics had, for their livelihood, almost entirely depended upon pasture. It was only in the Carolingian age that the cultivation of meadows began to develop ; and in consequence of the vast increase in produce of cattle, the continental Germanic tribes grew quickly in numbers. But for several centuries the improved cultivation of meadows for the purposes of pasture continued to surpass agriculture in importance, and it seems to have been not much before a.d. iooo that agriculture took equal rank with pasture as a means of livelihood. The pasture time did not end before the beginning of actual winter. After Martinmas it was no longer considered possible to pasture foals, so that before a.d. 800 it was not customary to let them be out at pasture after that term. 2 For the same length of time swine were kept in the oak forest in the Westphalia and Tirol of the fifteenth century.^ The time in which no pasture was possible extends there down to the sixteenth century von St. Martanstag bis auf mitten meien^
^ Acta Conciliorut?t, Parisiis, 1714, VI., i, col. 185, Synodus Ravennae habita, A.D. 877, i., ii.
"^ Capitulare de Villis, by Charlemagne : " Ut poledros nostros missa sancti Martini hiemale ad palatium omnimodis habeant."
^ "Op S. Remigy dach (Oct. i) in tho driven in den wolde twelff schwine vnd een beer vnd die Martin wieder vuith tho driven ; weer saecke die beer daer nicht mede en ist, mach men die schwine uthschiitten" (a.d. 1465, Speller Waldweistum , Westphalia), Freiherrvon Low, Uber die Markgenos sense haft eft, Heidelberg, 1829, p. 99; Piper, Beschreibungdes Markenrechts in Westphalen, Halle, 1763, pp. 158, 159.
*Zingerle, Tiroler Weistiimer, III., pp. 72, 73, A.D. 1542; on the meadows which were the common property of the communities of Mais and Burgeis. A long list of cases from Tirol legal documents, which shows that Martinmas was throughout the beginning of
MARTINMAS AND MICHAELMAS 63
At least as regards the beginning of winter, similar conditions ruled the economic year of the Anglo-Saxons of the eleventh century. Just as Tirol documents of the fifteenth century considered the season of the year to be summer from mitte merzen to Martinmas,^ the Reciitudo Geburi treated the season of winter as from Martinmas till Easter, Martinmas being given as the date at which the ploughing of the fields came to an end, and the time between February 2 and Easter being denoted as no less busy than the harvest-tide.2 It is in this latter state, however, that a change is contained. Whilst the real pasture time does not begin much before mid-May, the field work sets in about two months earlier, though in Germany nowhere at the beginning of February.
But soon enough in autumn also a change was wrought by the spreading of agriculture. All grain and aftermath are stored in the bams towards
winter, is given in the apparatus to my Geschichte der dentschen VVeihnacht, Leipzig, 1893, pp. 291-3. As late as the fifteenth century the I'ibe Herr sant Martein is addressed as the keeper and patron of cattle (Grimm, Deutsche Mythologie, 1189).
" Ich treip heut aus in unser lieben frauen haus, in Abrahames garten der lieber herr sant Martein der sol heut meines vihes warten."
Karl Godeke, Deutsche Dichtung im Mittelalter, Dresden, 1871, p. 243-5. Hirtensegen from a fifteenth century MS.
^Zingerle, TiroUr Weisiiimer, IV., p. 33, A.D. 1431, of Partschins, whilst the seven- teenth century began it on April 23, and ended it on Martinmas. Zingerle, Ibid., III., p. 65, A.D. 1630, of Burgeis.
^ 1 horpe's Ancient Laws, I., 434-435, : " Rectitudines Singularum Personarum : Geburi consuetudines inveniuntur multimodae, et ubi sunt onerosae et ubi sunt leviores aut mediae. In quibusdam terris operatur opus septimanae, 11. dies, sic opus sicut ei dicetur per anni spatium, omni septimana ; et in Augusto HI. dies pro septimanali operatione, et a festo Candelarum ad usque Pascha III. Si averiat, non cogitur operari quamdiu equus eius foris moratur. Dare debet in festo Sancti Michaelis X. denarios de gablo, et Sancti Martini die xxiii., et sestarium ordei, et II. gallinas. Ad Pascha i. ovem juvenem, vel II. denarios. Et jacebit a festo Sancti Martini usque ad Pascha ad faldani domini sui, quotiens ei pertinebit. Et a termino quo primitus arabitur usque ad festum Sancti Martini arabit unaquaque septimana i. acram, et ipse parabit semen domini sui in horreo. Ad haec 11 1, acras precum, et duas de herbagio. Si plus indigeat herbagio, arabit proinde sicut ei permittatur." Here August is in the Anglo-Saxon text corresponded by hcerfest.
64 YULE AND CHRISTMAS
the end of September. In Anglo-Saxon time August was the month of harvest.^ Except Professor Weinhold, nobody doubts any longer the late origin of the harvest festivals. Professor Mogk agrees with Heino Pfannen- schmid, author of Germanische Erntefestc, who maintains that Michaelmas is rooted in economic conditions, the existence of which the Germanics owe to the Romans.2 In the same degree as, in the centuries which followed, agriculture excelled pasture as a means of producing food, Martinmas was bound to decay in favour of Michaelmas, which was bound to receive ever new stress. But another economic force also set in, with a tendency destructive of a Martinmas celebration, though without anything in it to raise the significance of Michaelmas. In olden times it had been the most economical course to leave cattle, swine, sheep, and horses on the pasture grounds till the actual winter came, and then at once to kill all such of those animals as could not be kept over the winter. Thus, in the first half of November, a great killing time for the domestic animals had begun, which was apparently distinguished by the festival at the beginning of winter. With the improvement which took place in the cultivation of meadows in the Carolingian age, the quantity of hay produced annually
^This appears plainly from Thorpe's Ancient Laws, I., 432-33: " Recti tudines Singu- larum Personarum : Cot-setle rectum est juxta quod in terra constitutum est. Apud quosdam debet omni die Lunae, per anni spatium, operari domino suo, et tribus diebus unaquaque septimana in Augusto. Apud quosdam, operatur per totum Augustum, omni die, et unam acram avenae metit pro diurnale opere." The word corresponding to August in the Anglo-Saxon text of this Rectitudo is again hcerfest : " Kote-setlan riht. be dham dhe on lande stent. On sumon he sceal selce Mon-daege ofer geares iyrst his laforde wyrcan. odhdh III. dagar selcre wucan on hserfest ne dhearf he land-gafol syllan." When Jacob Grimm explains evenmant (September) as meaning oats-month (from Latin havaia, Geschichte der deutschen Sprache, 1848, L, p. 87), he is probably wrong, for oats were reaped before September. The term, which is of very late origin, is rather to be put beside even-naht (equinox), and means the month of equinox on Frisian ground. So is Grotefend wrong {Zeitrechnung, L, p. 54). From the quotations given there it is apparent that the term Evenmaend is confined to the Nether-Rhine up to Cologne.
'^"Auch auf deutschem Boden scheinen wir noch Uberreste dieser alten Sommer- und Herbstopfer zu haben : jener in der Hagelfeier, dem Johannisopfer, an dem es besonders gait, Menschen, Vieh und Erzeugnisse des Bodens vor bosen Geistern zu schiitzen, dieser in den Erntefesten oder den Martinsschmausen, doch sind die Nachrichten auf diesem Gebiete mit Vorsicht fiir altgermanischen Kult zu verwerten, da sie in Kulturverhaltnissen
MARTINMAS AND MICHAELMAS 65
was increased, and consequently it was no longer necessary to slaughter at once all the domestic animals designed for food. They could now in part be kept for some time in the stable, and fattened whilst they did not move about very much. Thus the great killing time slowly advanced further into winter — to St. Andrew's day (November 30)^ or St. Nicolas day (December 5).^
So late as the time of King David I. of Scotland (11 24-1 153) the usual time of slaughter for cattle, swine, and sheep was from Martinmas till Christmas, and these forty-four days were in legal language called "tyme of slauchter."^ Almost contemporaneously a pig appeared as a duty
ihre Wurzel haben, die wir hauptsachlich den Romern verdanken" {Mythologie, 1 127, in Paul's Gnindriss der germanischen Fhilologie, Strassburg, 1891, Vol. I.). This argument does not, however, hold good for the Martinsschmduse ', for the impossibility to pasture cattle, sheep, and horses beyond Martinmas is much older than the Roman influence upon the Germanics is.
*A St. Andrew's feast is mentioned e.g. by Melchior Goldast of Haiminsfeld, Rerum Alamannicarum Scriptores Aliquot Vetusti, Francofurti, 1661, I., p. 97, in Ephetnerides Monasterii S. Galli: "Andreae Apostoli. Eodem festo dat Hospitarius X. fercula, scilicet bis cames, bis pisces, bis caseos, bis ova, duos ciatos, et unum stuopum, maximum leibonem, et minorem leibunculum, et in vespera stuopum, lunulas et oblatas de Linkinwiller. "
^Thorpe, Aiuient Laws, I., 461, Leges Regis Edwardi Confessoris (1042-1066): " De Occisionibus Animalium contra natale. xxxix. Cum autem dictum est, quod non emerent animalia praeter pl^ios, clamaverunt macecrarii, qnos Angli vocant fleismangeres, de civitatibus et burgis, quod quaque die oportebat eos emere animalia, occidere et vendere [L. add : nam in occisione animalium erat vita eorum]. Clamabant etiam cives et burgenses pro consuetudinibus suis, quod circa festum Sancti Martini emebant animalia [L. instead : consueverant animalia in foro mercari] sine plegiis, ad faciendas suas occisiones contra Natale Domini, quas consuetudines justas et sapienter ductas non auferimus eis, tamen in mercatis emptis cum testibus et cognitione venditorum." As to the masting of swine and the varying thickness of their fat, compare the Lmws of King lite in Thorpe's Ancient Laru's and Institutes of England, I., p. 133 (xlix.), where different fines are prescribed for taking forbidden mast, according to the thickness of the fat of the swine. There were rules laid down for the swine-herd, how many swine of each class had to be slaughtered every year (Thorpe, Ancient Laws, I., pp. 436 437 : " Gafolswane, id est, ad censum porcario, pertinet, ut suam occisionem det secundum quod in patria statutum est. In multis locis Stat, ut det singulis annis xv. porcos ad occisionem, x. veteres et V. juvenes ; ipse autem habeat super-augmentum "), though the exact time of killing the swine is not stated in any Rectitude of the Anglo-Saxon period.
* Leges Burgorum Scocie, or luges et Consuetudines Quatuor Burgorum Berewic Rokisburg Edinburg et Strivelin, constitute edite ac confirmate per Regent David, titulo
£
66 YULE AND CHRISTMAS
to be paid at Christmas in Germany,^ which, of course, meant that it was to be killed at once.
In Scotland Martinmas was, so late as 1800, "the term at which beeves are usually killed for winter." This was "commonly called Martlemas in England, whence the phrase mentioned by Serenius,^ ' Martlemas beef.'"^ Brand's Popular Antiquities'^ tell of a little later time : " Two or more of the poorer sort of rustic families still join in purchasing a cow, etc., for slaughter at the time (called in Northumberland a Mart), the entrails of which, after having been filled with a kind of pudding meat consisting of blood, suet, groats, etc., are formed into little sausage links, boiled, and sent about as presents, etc. From their appearance they are called Black Puddings." Jamieson^ mentions that the Black Puddings were, at the beginning of our
LXIV. De Officio Carnificum (in The Acts of the Parliaments of Scotland, Vol. I., A.D. 1 124 to 1423 ; 1844, p. 346) :
LXIV. "De Officio Carnificum "Of Fleschewaris in the Burgh
" Quicunque carnes vendere voluerit " Quha that wyl sell flesche he sal sell
vendat bonas carnes scilicet bovinas ovinas gude flesche beyff muttone and pork eftir
et porcinas et vendat secundum considera- the ordinans of gud men of the toune and
cionem proborum hominum ville et ponat he sal sett his flesche opynly in his wyndow
eas in fenestra sua ut sint communes omni- that it be sene communly till al men that
bus emere volentibus Carnifices vero will tharof And fleschewaris forsuth sal
servient burgensibus tempore occisionis serve the burges in tyme of slauchter that
scilicet a festo sancti Martini usque ad is to say fra the fest of sayncte Martyne
natale Domini de carnibus suis preparandis quhil yhule of the flesche in thar lardyner
et conficiendis in lardariis Si vero carnes to be graythit and dycht And gif the
male preparentur carnifex restituet ei fleschewar graythis ivil flesche he sal restor
dampnum suum cuius erant animalia hym the scathis that aw the bestys And
Carnifices dum serviunt burgensibus come- the fleschewaris quhilis thai serve thaim
dent ad mensam illorum scilicet cum thai sal ete at thair burde wyth thair ser-
servientibus eorum Et habebunt pro uno vandis And thai sal hafe for a cow or ane
marto obolum pro quinque ovibus obulum ox a halpeny and for v shepe a halpeny
pro uno porco obulum " and for a swyne a halpeny "
^ Grimm, Deutsche Rechtsaltertiimer, V., 537, Baubach, Lower Alsace, A.i). 1 143, § 5 : "Ipse villicus mansum cum omnibus iustitiis habebit, porro in natale domini curiam visitabit, 12 panes, 4 sextaria vini et unum porcum, quem pascalem vocant, apportabit."
''■English and Swedish Dictionary, Nykoping, 1757.
^Jamieson, Etytnological Dictionary of the Scottish Language, under "Mart."
*P. 355- ^Jbid.
MARTINMAS AND MICHAELMAS 67
century, as they still are, an appendage of the Mart in Scotland. They were made of blood, suet, onions, pepper, and a little oatmeal.^ A cow or ox which was fattened, killed, and salted for winter provision was at least from the fifteenth to the nineteenth century in Scotland called a Mart (Gaelic = cow), Marie, or Mairt?
The sculptured capitals of the choir-pillars of Carlisle Cathedral present a probably unrivalled fourteenth century series of figures, depicting the occu- pations of the seasons.^ While June is there described as the month of hunting with a hawk, and July as the time of mowing with a scythe, the representative of August, holding in one hand a crutch and in the other a weed-hook, is cutting off with the latter the thick succulent stalk of a thistle- leaf which borders the opening ; and it is September which is denoted as the month of grain-harvest, its symbol being a man in a field of wheat, holding a handful in his right hand, and cutting it with a sickle in his left. October is the tide of grape-harvest, the bunch of grapes in the left hand,
^ The eighteenth century song says :
' ' It fell about the Martinmas time, And a gay time it was than, When our gudewife got puddings to mak', And she boil'd them in the pan." (_The Songs of Scotland chronologically arranged, London, 2nd ed., p. 158; "Get up and bar the door," from Herd's Collection).
^ Jamieson, Etymological Dictionary of the Scottish Language, under ' ' Mart. " He gives the following instances: "Of fleshers being burgesses, and slaying mairts with their awin hands" {Chalmerlan Air, c. 39, s. 68). "That all — martis, muttoun, pultrie, — that war in the handis of his Progenitouris and Father — cum to our Souerane Lord, to the honorabill sustentation of his hous and nobill estate" (Acts of James IV., 1489, c. 24, edit. 1566; Skene, Laws and Acts of Parliament, Fol., Edin., 1597, c. lo). "In 1565, the rents were £'2.(>'i 19s. 2d. sterling, — 60 marts or fat beeves, 162 sheep," etc. {Statistical Account, v., 4). The same word is also used metaphorically to denote those who are pampered with ease and prosperity: "As for the fed Marts of this warlde, the Lord in his righteous judgement, hes appoynted them for slaughter" (Bruce's Eleven Sermons, 1591, A. 4 a., Jamieson, Ibid.). There can be no doubt that this Celtic word Mart -covf was very early brought into connection with Martinmas.
^ Described by James Fowler in the Transactions of the Cumberland and Westviorland Antiquarian and Archaeological Society, IV., p. 280, which description is extracted in R. S. Ferguson's Guide to Carlisle, Carlisle, 1890, pp. 45-46.
68 YULE AND CHRISTMAS
the hooked knife in the right hand of the vine-dresser, and the basket upon the ground by his side denoting it as such. It is November and December which bring in the domestic animals. The emblem of November is a man in boots, sowing corn broadcast with his right hand out of a wicker basket hanging at his left side, suspended by a strap from his right shoulder. Oak-leaves and acorns are on the bell of the western member of this capital ; on the eastern member of the next capital is seen a swine- herd in the midst of oak-leaves and acorns tending a herd of swine feeding, one of the swine having its head raised as if to catch a falling acorn. December is a man with an axe grasped by the handle in both hands, raised, and with the back of it about to fall on the forehead of an ox, which is held fast by its horns by a man, in similar costume to the first, standing behind it. January is the time of gay drinking, its representative having three smooth unbearded faces under one skull cap, drinking by the right and left mouths out of shallow cups held respectively in the right hand and in the left, and with the central face looking impassively forward. A jug wherewith to replenish his cups stands on the ground at his left side. February is nothing but the month of cold and wet weather ; March digs up the ground round still leafless trees; April, with a crooked knife, cuts dry branches down from them ; and May is the gay month of young foliage and flowers, its symbol being a woman holding in each hand a ^^eur-de-fys-shsL^ted. bunch of sprouting foliage, and presenting them to a young man, who, by his right hand, takes from her the bunch in her hand.
On the Nether Rhine, about a.d. 1400, the killing time of swine was about Christmas,^ just as in England, a little later, December was the principal month of slaughter.^ In the Germany of the sixteenth century, under the
^ Annalen des Historischett Vereins fiir den Niederrhein, Instalment LIV., 1892, p. 12; Book of Expenditure of Herr von Drachenfels, 139S, p. 21, Jan. 4, 1396, No. 56 : " i alb um spiskraut, i alb um eier up dat huis, doe man die verken affdeide ; " p. 37, Nov. 29, 1396 : "Ich haen Heynen Volrait gegen 55 m van den verken die zuo jair up vur kirsnacht wurden gegulden."
^ Bartholomaeus Anglicus, IX., chap. xix. (ed. 1488): "De Decembre. In hoc mense propter asperitatem frigoris sunt altilia et animalia domestica multae quietis et parvi motus, et ideo plurimum impinguantur. Unde tunc temporis interficiuntur potissime et mactantur;
MARTINMAS AND MICHAELMAS 69
influence of agricultural progress, the killing time of pigs extended to Epiphany,^ and shortly afterwards to St. Anthony's Day (January ly),^ thus reaching the second half of January.^ To February it was shifted not earlier than the seventeenth century, when the cultivation of potatoes had become popular and productive enough to keep especially swine through the greater part
propter quod depingitur tanquam camifex qui cum securi percutit et mactat porcum suum.' That the slaughter began in November is shown by the following passage : " Of our tame boars we make brawn, which is a kind of meat not usually known to strangers. . . . With us it is accounted a great piece of service at the table from November until February be ended, but chiefly in the Christmas time. With the same also we begin our dinners each day after other ; and, because it is somewhat hard of digestion, a draught of malvesey, bastard, or muscadel is usually drank after it." . . . {Elizabethan England: from "A Description of England," by William Harrison (in Hollinshed' s Chronicles), edited by Lothorp Withington, with Introduction by F. J. Fumivall, London, The Scott Library, p. 658).
^ Lauterbach Document of 1589, Grimm, Deutsche Rechtsaltertiiiner, III., 369: a young pig which had not reached maturity was led round through the benches (and, probably, killed afterwards).
^Montanus, p. 17; Sebastian Franck, Weltbuch, I., p. 131; Ulrich Jahn, Deutsche Opfergebrduche, p. 266.
^ At the end of the sixteenth century the occupations of the time of October to January are described in the following way :
October. " Frigoribus coelum magis intractabile reddit, October, stabula hinc cogit adire pecus. Arboribus fructus adimit, spoliatque decore, Atque etiam cupide turbida musta bibit.
Aliter. October mustum calcatis exprimit uvis Et serit hoc anno quae redeunte metat.
November. Ligna vehit mactatque boves, et laetus ad ignem Ebria Martini festa November agit. Ad pastum in silvam porcos compellit, et ipse Pinguibus interea vescitur anseribus.
Aliter. Autumnus quaecunque dedit, consume November, Et pinguem hybema glande trucido suem.
70 YULE AND CHRISTMAS
of the winter. This movement was, on legal ground, accompanied by a shifting of a great number of duties and taxes to the beginning of the new year, reckoned either as at Christmas or on January i, so that Michael- mas, Christmas, Easter, and St. John's Day came to be the four great terms all over Europe, wherever the Roman Calendar and the Roman Church succeeded in uprooting the ancient Germanic tri-partition of the year.^ The economic evolution, more especially the prevalence of agriculture over cattle-keeping, thus tended to destroy the ancient Germanic mid-November celebration, whilst favouring both a harvest festival held earlier in the year and the development of a festival about the middle of the German winter.
December. ^ -
In nive persequitur vestigia pressa ferarum, Abluit et calida membra December aqua. Affert Solslitium, celebrat cunabula Christi, Et iugulat porcos, tribula dura ferit.
A liter. Haud avis, baud fera venanti deest ulla Decembri, Quamvis ningat atrox et gelet usque vadum.
Januarius. lanus vina bibit, crepitantique assidet igni. Et pingues carnes torret, editque suem. Annum praeteritum claudit, reseratque futurum, Sed venam ferro tangere, iure vetat"
{Ranzovii Exempla, Quibus Aetrologicae Scientiae Certitudo Contprobatur, Coloniae, 1585, pp. 304, 306, 307, which latter two are there wrongly numbered 400 and 303). Another piece from the same time says of December :
" Prassen will ich und leben wol, Eine Sau ich itzunder stechen sol." (Grasse, Des deutschen Landmanns Practica, Dresden, 1858, p. 28).
^ Michaelmas appears as a term for paying duties very frequently from the sixteenth century. Landesordnung des Herzogtums Preussen von 1525, Pfannenschmid, Gernianische Erntefeste, p. 118; Richter, Kirchenordnungen, I., 32; Ibid., II., 355; Hoyaische Kirchenordnung von 1573, where Michaelmas is called the vierte Hochfesttag, and put into parallel with Christmas, Pasch, and Whitsunday, thus clearly standing in relation to the Roman quartering of the year.
CHAPTER VII.
SOLSTICES AND EQUINOXES.
Jacob Grimm, as has been shown already, had a perfect grasp of the six-fold division of the year found among the Germanics at the dawn of history. And though he did not know that it had been borrowed from the Orient, and probably ultimately from Egypt, and that it was by no means genuine and common Aryan property, yet he did not fail to see how deep-rooted it was in the legal, cultural, and economic conditions of our ancestors. It is a very strange fact that he should have thought a knowledge of two solstices and two equinoxes, together with a quartering of the year, reconcilable with the conclusions as to the Aryan year in general, which of necessity must be drawn from a six-fold division. Had he been aware that these two ways of looking at the course of the year were mutually exclusive, he would have been led to a further examination of each, and then would have found that a cognizance of solstices and equinoxes must be denied to the early Aryans, as well as to the Germanics before their acquaintance with the Roman calendar; that the quartering of the year is of purely Roman origin, and is not found elsewhere ; that there is no historical evidence whatever for a celebration of solstices and equinoxes among the Germanics in their pre- Roman time; that philology and folklore, the history of the Christian Church, and the history of agri- culture all point to a three-fold partition of the year with the beginning about the middle of November. It was Jacob Grimm's way to regard our ancient ancestors as speculative philosophers who stood aloof from the
72
YULE AND CHRISTMAS
Struggle for existence, and who shaped their yearly course according to their own fancies and their belief in gods; and he failed to see that it is the economic conditions which, in primeval times, as they do in our own, fixed all the more important features of daily and yearly life, leaving only a very limited realm to a manifestation of personal likes and dislikes; nay, that that realm gets smaller and smaller every step we go further back into the past.
Jacob Grimm was a king in his kingdom of Germanic philology, and even where he stumbled on his royal road, he could not help indicating the way to walk safely. But what about those who followed his route? Was it not strange that they should think his stumbles worthy, above all, of imitation, that they should altogether neglect his useful hints and the material gathered by him which pointed in the right direction? It looks like a joke in the history of Germanic antiquarian studies, that the man who after Grimm made this subject his special study, and devoted years to it, should have wasted all his energy in the attempt to prove that the Germanics in pre-Roman times had exactly the same year as the Romans; that they, therefore, had nothing to get from them, and rejoiced in quartering their year and celebrating imaginary solstices and equinoxes.^
The observation of the change of cold in winter and heat in summer is one thing, that of the movement of the rising-point of the sun on the horizon is another. If some peoples of antiquity sought to find a causal connection between the two things, that connection was hopelessly wrong, the proper relation of the two lines of observation having been known only since Copernicus, i.e., since the sixteenth century. That primitive people were bound to connect them is by no means true, and it is more than doubtful whether the next thing would be to observe so-called solstices and equinoxes. The fixing of the date at which day and night are exactly equal lacks entirely in economic interest and significance, and certainly never affected the minds of primitive peoples. The observation of so-called
^ So even Heino Pfannenschmid seems to think when he explains his theory of the Germanic year in his otherwise excellent book on Germanische Erntefeste im heidnischen und christlichen Culttis (Hannover, 1878, pp. 16 ss. and 326 ss).
SOLSTICES AND EQUINOXES 73
solstices, on the other hand, is extremely difficult. Whilst in autumn and spring the rising-point of the sun visibly shifts from day to day, it scarcely shifts at all from the beginning of December to the middle of January, and from the beginning of June to the middle of July. Even the astronomers whom Caesar had at his disposal were not able to fix the solstices and equinox^^; actually ; and although he ordered the winter sol- stice to take place icember 25, and the summer solstice on June 24, they persistently and obstinately disobeyed — the winter solstice making a point of taking place on December 23, a.d. i ; on December 22, a.d. ioi; on December 21, a.d. 201; on December 20, a.d. 301; on December 19, A.D. 401; on December 18, a.d. 601; on December 17, a.d. 801; and so on; so that in 1501 it was wicked enough to take place on December 12. The spring equinox and autumnal equinox apparently shared the delight in moving backward by eighteen hours a century, and shifted in the same degree away from March 21 and September 22.
It was not earlier than at their close contact with the Romans that the Germanics became acquainted (as with other Roman institutions) with sol- stices and equinoxes,^ although not with their true astronomical dates, but with the pseudo-equinoxes and solstices of the Julian calendar, to which their wise men faithfully stuck for a millenium, whilst popular tradition knew nothing of these foreign-made goods. Nevertheless these innovations brought the ancient Germanics face to face with a task which may be called philological. They were compelled to create new words for the new conceptions with which they were made familiar, and they chose the simplest way that offered, by merely translating the Latin terms. But not all tribes fulfilling that task in the same way, there arose considerable divergence in the expressions used, fror'i which divergence even now we can see that the Germanics had no ancient common word for them, such as they had for winter and summer. Nay, they even took over the particular limitations with which the Latin expression solsticium was used. Solsticium is in Latin, with a few late exceptions, exclusively used for the
^ Kuhn's article in Zacher's Zeitschrift fur deutsche Philologie, 1868, I., 118, is not to be taken seriously, at least so far as the Germanics are concerned. He fails to give any proof for their knowledge of a solar year with solstices and equinoxes.
74 YULE AND CHRISTMAS
summer solstice. How well aware the Roman mind was of that appears from the fact that the adjective solsticialis refers, even in late Latin, exclusively to summer and the middle of it, and is used as the contrary of brumalis. The name for the shortest day of the year was simply bruma (supposed to be a contraction of brevissima \dies\)^ which also meant the whole of winter. Germany herself has evolved four words for solstice, all four of which apply to the summer solstice alone : sunwende, sungiht^ sunstede, and sommertag. Grotefend,^ who maintains that by sohticium without an additional qualification the summer solstice is, in most cases, meant, is unable to give even a single instance from a medieval document in which a winter solstice occurs, and (although he heads his paragraph Sohticium estivale, brumale) gives examples solely of sohticium estivale. In another paragraph, however,^ he quite properly remarks that the whole of the following expressions : sonnwenden, sonnabenden, sonnenbenttag, sunn- benden, sunnewenttag, sunibentag, sunwende, sunnstede, sungichten, sungicht, suniich, apply to the summer solstice alone. This amounts to the fact, that no medieval instance is known of December 25, or any of the days about it, having ever been called solstice in the German language : nay, that there is no medieval word wintersonnwende or the like, the corre- sponding term in New-High-German being of quite modern growth. Wherever the word sunnewende occurs in the Middle-High-German poetry of the twelfth and thirteenth centuries there is no doubt that it can apply to the summer solstice solely. No poet or writer of prose thinks of adding any adjective to make that clear.^ Sonnenwende is turning of the sun ; sungicht is walk of the sun ; and sunstede is standing of the sun — three quite different things. These terms do not occur all over Germany, but are restricted to several dialects. So sunstede is exclusively Frisian, at least
^ Zeitrechnung des deutschen Mittelalters, 1891, I., p. 178.
"^ Zeitrechnung des deutschen Mittelalters, I., p. 181.
^"Hiute ist der ahte tac nach sunewenden," Iwein, 114; "ze einen sunewenden, " Nibelungenlied, 32, 4, and Lachmann's Nibelungen Not, 2023, i ; " vor disen sunwenden," Ibid., 678, 3; 694, 3; "zen nsehsten sunwenden," Ibid., 1352, 4; 1424, 4; IVigalois, 1717; "an sunwenden abend," Nibelungen Not, 1754, 3; "ze sunewenden," Tristan, 5987 ; "Santjohans sunewenden tac," Ls., 2, 708.
SOLSTICES AND EQUINOXES 75
as far as the continent is concerned, for it also occurs in Anglo-Saxon, thus appearing to be an Anglo-Frisian term.^ From this we may conclude that the Germanics became acquainted with the Roman summer solstice at a time when the Western Germanics had already separated into German* and Anglo-Frisians, but beforeMhe;* community of speech bgtw^^ Angles' and Frisians was broken uplpprtTiiis^ is ♦the more likely, as-^FrisianTTaiid English have in common, another, word > for the same notion: Fnsiaii sumerdey, English summer night-f' whilst sommertag for solstice is sporadic in German. This being tfee .sj9.te of matters among the Western Germanics, nobody will wondeni^jfe^^^he -several Northern Germanic tribes evolved almost each a nam&^|>^'eir own. How could it have been otherwise, since they were not^ acquainted^ wwith, the. summer solstice^ juntil after they had settled in the several parts of the north! Thus Danish hjj^s Solhverv, or throwing of the sun^ from' which" in" modern" "times -is derived Vintersolhvers- festen. Norwegian, likewise," has So^kve^ a.nd So/kver;j^ jyvith. the modern derivatives Sommersolkve^t and Vinter solhverv. But Swedish has Solsta^tl^ with the modern derivative Vintersolstand, and Icelandic' has 61^/y/[)i/«r witii the modern derivative Vetrarsblstodur. ■ ^J;. >,/'.., .
In German the word sonnwende (solstice), thoughVflever us§d for winter solstice, is sometimes usfec^- for equinox^ so that Germany can boast of having three solstices, which s,she certainly deserves on account of her ancient three seasons.^ In FlarMjBrs the equinoxes are called summer day (March 17) and winter day (September 21),^ whilst in Frisian and English
^ It is found, e.g. , in an Anglo-Saxon treatise on astjonomy based entirely on Beda's work, De Ratione Temporum : Thomas Wright, Popular Treatises on Scieiue written during the Middle Ages in Anglo-Saxon, Anglo-Norman, and English, London, mdcccxli.. Historical Society of Science Publication, pp. 8, 9: " Aestas is suincM^se haefdh sunn-stede; hiems is winter, se haefdh otherne sunn-stede." '"^^ ,-,
^Grotefend, ZeUrechnung des deutschen Mittelalters, 189 1, I., p. 178.
^"Sonnwende der ander in der vasten," Groiekn^, Zeitrechnung, 1891, I., 181. Grotefend maintains the same usage to have existed among the Frisians, Ibid., I., 189: "A sunna ewenda bifara sente Liudgeris dei" (Richthofen, Friesische Rechtsquellen, 169); but I should be rather inclined to think that we have there to do with Saturday instead of equinox.
* Grotefend, Zeitrechnung, 1 891, I., p. 178.
76 YULE AND CHRISTMAS
summer day means summer solstice. In Middle-High-German the literal translation of equinox {ebennaht) is very rare and very late, so that it almost seems to have been borrowed from Frisian. ^ At any rate it never was a popular date or an early term. It is not before the fifteenth century that the equinox is used for dating documents, and even then it is supplemented by other things.^ The complicated expression Tag und Nachtgletche, which bears the stamp of artificial manufacture, is of quite modern origin. Frisian and English evolve a little earlier than German their common term for equinox, A.S. evenniht or emnihie,^ Frisian evennaht.^ Among the Northern Germanics the term is exceedingly rare and very late. Modern Icelandic has Jafn-dcegur and jafn-dcegri, equal days ; Modern Danish has ^cEVudogn; Modern Norwegian has jaftidoegri, jevndogn, and jafnncetti.
If we knew nothing about the actual division of the Germanic year, it would, on the authority of these philological facts, be safe to assume that the ancient Germanics did not base their seasons and tides on solstices and equinoxes. There was once a theory current according to which everything — myth, cult, custom — was traced back to an alleged sun worship or observations of the events visible in the sky, such as the rising of the sun, the hiding of the sun behind clouds, and the shifting of the rising- point of the sun on the horizon. But, as regards Germanic tribes, that theory is so little applicable as to make it quite certain that among our
^ ' ' Der onager, in dem merzen an dem funf und zweinzigisten tage so luot er zwelfstunt unde sam ofte in der naht, davon bekennet man sint, daz ebennaht belouhtet ir sunne unde waet ir wint," Karajan, 82, 26 ; Mliller und Zarncke, Mittelhochdeutsches Worterbuch, Leipzig, 1863, L, 301. Ibid., " Ebennahtec, equinoxialis obent-nehtig," Diefenbach, Glossen., 109; " Equinoxium, ebennachtig," lbid.\ " aequinoctium ewennachtig," Mone, VI IL, 249.
^"1402 als equenoxium was umme sunte Gregorius dge nten" (Magdeburger Sc happen- chronik, 304), Grotefend, Zeitrechnung, 1898, II., 2, 194.
^Thomas Wright, Popular Treatises on Science, London, 1841, pp. 8, 9, "Ver is lencten-tid, seo hsefdh emnihte ; autumnus is hserfest, the hsefdh odhre emnihte," Saxon Chronicle, Laud MS., E, 1048; "to hserfestes emnihte," John Earle, Two of the Saxon Chronicles Parallel, Oxford, 1865, pp. 179, 180.
* Richthofen, Friesische Kechtsquellen, 390-392. "Letera evennaht" is the September equinox, Grotefend, Zeitrechnung, I., 54.
SOLSTICES AND EQUINOXES 77
ancestors the sun was no deity. We have not only absolutely no traces of sun worship among the Germanic nations, but even in historical times the sun has been of different gender in different Germanic languages. Nay, different Germanic tribes even had different words for sun, which, though coming from the same root, were formed with different suffixes (Gothic sunno, fem., and sunna, masc. ; German Sonne, fem. ; English sun, masc. ; Gothic sauil, neut.; Anglo-Saxon sol; Old Scandinavian sol). As to deities, the Germanics seem to have originally had one god only, his name being *Tiwaz (Greek Zevs, Latin Dies-piter), to whom in common Germanic times another was added, named *Thonaraz, whilst North Germany still later pro- duced a third, * Wodanaz, who in the Middle Ages immigrated to Scandinavia, but never won the adoration of the High-German tribes. Besides, there was one goddess, called Frija. At any rate we may affirm that at the time when, probably in the first century of our era, the Germans took over from the Romans the Phoenician week of seven days, and replaced their names by German terms which corresponded exactly to the Roman terms, there was not even a god to take the place of Saturnus.
Whilst the summer solstice was probably taken over directly from popular Roman tradition, the equinoxes seem to have become familiar to the clerical Germanic mind through the bearing the spring equinox had on the fixing of Easter, the more so because violent discussions about the proper time for holding Easter were going on for several centuries, and most seriously affected the Anglican Church. It is in connection with this controversy that we get our first information about our ancestors' ability to find the term of the spring equinox. Ceolfrid's Letter on Easter and the Tonsure, written circa a.d. 710,^ shows that about that time the capacity to
^ Haddan and Stubbs, Councils and Ecclesiastical Documents, Oxford, 1871, III., 289: ' * Aequinoctium autem, juxta sententiam omnium Orientalium, et maxima Aegyptiorum, qui prae ceteris doctoribus calculandi palmam tenent, duodecimo kalendarium Aprilium die pro venire consuevit, ut etiam ipsi horologica inspectione probamus." What stress was laid by the Middle Ages on the coincidence of Christmas and the winter solstice is evident from the fact that the keeping up of that coincidence is given as the reason for the institution of the leap year. Bracton's Note Book, ed. by F. W. Maitland, Vol. III., London, i887, p. 301 (fol. 196): "Sed hoc fit propter quandam necessitatem ad evitandam illud inconveniens, quod esset intemperies hiemalis in signis aestivalibus, et quia si possit
•jS YULE AND CHRISTMAS
fix the date of the equinoxes by observation had been attained in Great Britain, though several centuries elapsed after their first acquaintance with them before equinoxes and solstices were accepted as terms quartering a solar year of 365 days and a quarter. It was probably not before the eleventh century that this took place. The Anglo-Saxon treatise of astronomy, which is entirely based on Beda's Z>e Temporum Ratione'^ and on Roman views, calls the four Roman quarters of the year, which are halved by solstices and equinoxes, lencten-tid, sumor, hcerfest, and winter, of which lencten-tid, by its very name a compound with tide, is shown to be not an old term.
Notwithstanding all these facts. Professor Weinhold goes on talking about Germanic solstices and equinoxes as if nothing in the world were a fact better established. After having wrongly fixed the terms of the dual division of the year at the end of September and of March, and two of the terms of the three-fold partition on almost the same days, he proceeds ^ to declare that the Germanics halved their two seasons, summer and winter, and thus arrived, absolutely like the Romans, at four seasons (which, however, were no longer seasons, but broke entirely through the system of actual seasons). In his fanciful way he sets down the following bold guesses : ^ " Midwinter and midsummer, Christmas and the feast of John Baptist, according to ecclesiastical denomination, stand out in the German year as very ancient high tides. Through the standing still of the sun, which, according to the opinion of that time, stopped in turning round to a new
contingere quod Natale Domini celebraretur in aestate et Nativitas B. Johannis Bapt. in hieme." This passage was in all probability written before A.D. 1256 ; Ilenrici de Bracton, De Legibus et Consuetudinibus Angliae, Londini, 1640, Lib. V., 2, De Essoniis, fol. SSQ*" : ' ' Ille vero dies excrescens qui non est computabilis, ea ratione propter necessitatem ad vitandum illud inconveniens ne festum Natalis Domini celebretur in aestate et Nativitas Sancti Johannis Baptistae in hieme, quod contingere posset infra quingentos vel sexcentos annos, et etiam ita contingeret intemperies hiemalis in signis aestivalibus. "
^ Historical Society of Science, Popular Treatises on Science written during the Middle Ages, ed. by Thomas Wright, London, 1841, pp. 8, 9: " Feower tida synd ge-tealde on anum geare, that synd, ver, sestas, autumnus, hiems. Ver is lencten-tid, seo hsefdh emnihte ; sestas is sumor, se hsefdh sunn-stede ; autumnus is hserfest, the hsefdh odhre emnihte ; hiems is winter, se hsefdh otherne sunnstede."
^ Deutsche Jahrteilung, p. 9. ^ Ibid,, p. 9.
SOLSTICES AND EQUINOXES 79
journey, the people felt themselves driven to solemn rest and the service of the deity of the sky which led the sun. Divination and prophecy pre- vailed during those tides, and with their mysterious thrill interrupted the noisy joy which wreathed round heathen sacrifices." Yet there is not a shadow of historical evidence for these fancies. The Germanics neither had a festival about Christmas nor about the day of John Baptist. The Twelve-nights, of which he talks a little further on, are simply the Dode- kahemeron of the old Church, which existed there for centuries before they appeared among any Germanic tribe.^ Nay, all through the Middle Ages the term Sonnenwende, or solstice, has not a single time been shown to have been applied to December 25 : its use is absolutely restricted to June 24, just as the word solsticium was among the Romans. If Wein- hold^ places the Anglo-Saxon word lidha for June and July alongside the Dutch lauwe, louwmcent for January, explains them as lind and lau, trans- forms these meanings to "resting," and refers that adjective to the "rest of the sun," which, according to popular belief, i.e., according to his belief, took place about midwinter and midsummer, one may well be doubtful whether that serves to strengthen the position of his own hypothesis. The goddesses Ostara and Hreda, on whom he^ lays much stress, he has later given up himself. But he still deduces from the facts that the Scandinavians divided their year by October 9 to 14 and April 9 to 14 {vetrnatt and sumarndti), and that the Germans are shown to have had the Roman seasons, one of which began about October i, the conclusion that equi- noxes, of which the Germanics knew absolutely nothing, "divided the most ancient German year."*
In his Weihnacht-Spiele und Lieder aus Siiddeutschland und Schlesien ^ there is even a chapter headed in the index as "The Germanico-heathen celebration of the winter solstice," in which he gives a still more enraptur- ing delineation of that alleged Germanic festival, without being in the least disturbed by the fact that such a thing never existed. There even the
^Compare my own book Geschichte der deutschen Weihnacht, Leipzig, 1893, P- 282.
^ Deutsche Jahrteilung, p. 14.
^Weinhold, Deutsche Jahrteilung, 1862, p. 15. ^ Ibid., p. 6.
*Wien, 1875, pp. 4, ss.
8o YULE AND CHRISTMAS
error occurs, that the solstice had been called Jul, accompanied by another, that the, winter solstice was the beginning of the Germanic year. We learn that that time was devoted to Wodan, and Fricke, or Holda, or Berchta or Hera, or Gode; that the boar {bar) led about through the village was not a boar at all, but a bear; that it was not the central figure of the pro- cession, but probably merely accidental : and we have a hundred other products of unscientific imagination. The description given of the holy Twelve-nights of the Germanics ^ is almost touching. That the Christmas fires have a close relation to the sun ; that yule has etymologically to do with wheel', that the Christmas tree is to be derived from Wodan; that a great number of the customs in use from Martinmas to Easter should properly be held on Christmas eve, or, at least, on the Twelve-nights; these and an extensive list of other most surprising fancies can be learned from that book. So the whole of the thirty-six pages which Professor Weinhold's disciple. Dr. Ulrich Jahn, in his book Die deutschen Opfergebrduche bet Ackerbau und Viehzucht,'^ devotes to the offerings about the time of the winter solstice, contain, in so far as they are meant to apply to pre- Christian times, nothing but unhistorical speculations, and would have been better omitted from that book, which, in various respects, may be called useful, and certainly represents a much more critical attitude on the part of its author than any of the attempts of Professor Weinhold to deal with the problems of German popular tradition.
^ Given Wien, 1875, P- ^^' ^ Breslau, 1884, pp. 253-289.
CHAPTER VIII.
THE CALENDS OF JANUARY.
In the first century before Christ a number of Germanic tribes, through ^ commerce and war, came into close contact with the Romans, taking over from them in rapid succession the Roman capital alphabet of Egyptian origin to turn it into their runes, the Phoenicio-Roman week, the pre-Julian calendar with its beginning of the year on March i, some astronomical wisdom, and a variety of other things. They took over the institution of the ancient Roman leap year with its intercalary month, ^ although they did not add this mensis Mercidonius every second year between February 23 and 24, but about the middle of summer and at intervals which we do not know. This intercalary period of apparently about thirty days was the first thing to interfere with the congruity of the German year, which, so far, had known only tides of sixty days, but had not taken account of lunar periods for the purpose of dividing time, however conscientiously they might observe them as bringing good or bad luck to the affairs of daily life. Tacitus ^ keeps that usage quite distinct from the Germanic division of the year. So it continued for at least three-quarters of a
^We know this from Beda, De Temporum Ratione, chap, xv., who expressly testifies to the existence of an intercalary month among the ancient Angles.
"^ Germania, chap. xi. : "Coeunt nisi quid fortuitum et subitum incidit, certis diebus, cum aut inchoatur luna aut impletur ; nam agendis rebus hoc auspicatissimum initium credunt."
F
82 YULE AND CHRISTMAS
millennium.^ The introduction of Roman months instead of tides of about sixty days necessarily led to a breaking-up of the latter into two parts of equal length, for which new names were required. The fact that the six Germanic tides, which formed the course of a year, began about the middle of the Roman months made things a little difificult. Yet the Roman month- names were taken over and bye and bye replaced by new German terms, which were formed by means of a word probably identical with Germanic moon, mdnodh (Gothic menoths, Old Saxon monadh). It is, however, doubtful whether Latin mensis is exactly of the same derivation. Similarly, the relation of the root of moon to Sanskrit ma, to measure (Greek /xerpov), is disputed with good reason. The moon was, among the Aryans and among the Germanics in particular, anything but the medium for dividing the course of the year, for which they received, at a very early date, a ready-made theory of probably Egyptian origin. The word manodh was added to each of the sections in order to mark them clearly out from the old three-score-day tides, the names of which were used for forming the new compounds. So the old liuleis tide of sixty days at the beginning of winter was divided into a first liuleis month and a second liuleis month, the Lida tide in summer into a first Lida month and a second Lida month, to which, in intercalary years, a third Lida month was appended, whilst the words Liuleis, Lida, etc., without any addition, continued to mean a tide of about sixty days, two of which formed a Germanic season of a long hundred of days. Only gradually self-dependent names were developed for these half- tides which were denoted months, most of these names being taken from economic life, which naturally varied considerably between the coasts of the Baltic and the south of France, and from the British Isles down to the coasts of the Adriatic,^ and remained an ever new source of name- giving, especially during the time of transition from prevalent pasture to
'^ Acta Conciliorum, Parisiis, 1 714, Vol. III., col. 1686, Concilium Quinisextum sive in Trullo, A.D. 706, Ixv. : "Qui in noviluniis a quibusdam ante suas officinas et demos accenduntur rogos, supra quos etiam antiqua quadam consuetudine salire inepte ac delire Solent, iubemus deinceps cessare. Quisquis ergo tale quid fecerit, si sit quidem clericus, deponatur : sin autem laicus, segregetur. "
'^ Weinhold, Die deutscken Monatnamen, Halle, 1869, pp. 24-28.
THE CALENDS OF JANUARY 83
prevalent agriculture. The inexhaustible variety of circumstances which led to names for these new half-tides made impossible the development of one and the same series of home-made month-names for all Germanic tribes, or even for each of the principal groups of them ; nay, for individual tribes. The several hundreds of Germanic month-names found on Germanic territory from the sixth century down to the present time, with their innumerable variations of meaning, make impossible of attainment a system which would embrace them all.
In the year of Rome 707, />., forty-five years before the date from^^ which later the Christian era was counted, the Julian calendar began to reign in Italy and Gaul, and on the coasts of the Mediterranean in general. Within the hundred and fifty years which followed, Roman legions and Roman administration carried it over the Rhine into Germany, and beyond the channel into the British Isles. As long as Gaul remained a Roman province entirely Romanized; as long as down the banks of the Rhine there flourished large Roman towns; as long as hundreds of thousands of Germans served in Roman armies, visiting Italy, Greece, Asia Minor, Egypt, fighting all over the world then known, and more than once disposing of the Roman imperial throne ; as long as invading Goths went down to Italy and Vandilian invaders to the north of Africa without losing contact with those Germanic tribes which remained northward of the Alps — there was practically no limit to the entry of Roman knowledge into Germanic terri- tory; and in the suite of every-day experience there came Roman learning with its poets, historians, philosophers, rhetoricians, scientists, and physicians. All along the Rhine there flourished Roman rhetoric schools in consider- able number, in which the noble science of grammar and the trivium as well as the quadrivium were taught thoroughly. Among the Germanics there was no self-dependent scholarship that could successfully compete with those finest products of a higher civilization, and so it became, for about a thousand years, the task of the Germanics to receive and ever receive mental gifts from the civilization of the empire they destroyed.
In the suite of the new calendar which, after Julius Caesar, began the year with the Calends of January (at which date, subsequently to 153 B.C.,
84 YULE AND CHRISTMAS
the Roman consuls had entered their offices), the whole annual course of Roman festivals passed by degrees to Gaul, Germany, and the south of Great Britain; above all, the Saturnalia, Brumalia, and Kalendae Januartae, which, during the first three centuries of our era, were with great regularity observed in all the great towns along the Rhine, and thence spread to the inner parts of Germany, as far as Bohemia; nay, even to the Slave tribes east of the Germans, and to the Lithuanians north of them.^
Since even Professor Weinhold admits that the Roman calendar was one of the three forces which shaped the medieval German calendar,^ it will be worth while to see of what kind the Roman customs were which could be transferred to Germany along with the institution of the Calends of January and the neighbouring festivals. There was first of all the custom of New-Year's gifts or Strenae.^ In imperial Rome the people and the Senate were expected to present New- Year's gifts to the emperors,* it being related that Augustus had had a nocturnal vision requiring that people should annually, on a certain day, present money to him, which he received with a hollow hand.^ During his reign they were given on the Capitol; but CaHgula was so lost to a sense of shame,
^ The Lithuanians, according to the old significance of their winter festival, called many centuries later their Christmas Kalledos, a name which has wrongly been brought into connection with Lithuanian Kalada, log (Grimm, Deutsche Mythologie, 4th ed., p. 522), but is certain to have sprung from Calendae, since the Czechic word for Christmas is up till to-day Koleda (Polish Kolenda, Russian Koljada) ; a fifteenth century source calling Bohemian Christmas processions calendisatiotus (Usener, Christlicher Fest branch, Bonn, 1889 ; Johannes von Holleschau's Treatise on Christmas), and a verb colendisare appear- ing in old sources of Bohemian law (Rossler, Prager Recht, p. 95, No. 140). Compare my own Geschichte der deutschen Weihnacht, Leipzig, 1893, p. 287, note to p. 14,^, where the quotations from Holleschau's treatise are given.
"^ Uber deutsche Jahrteilung, Kiel, 1862, p. 3.
^ The habit of New-Year's presents boni ominis causa is first mentioned by Plautus (+ 184 B.C.) in his Stichus, iii. 2, 6 ; v. 2, 24. Their purpose is explained by Ovid, Fasti, i. 187. Cakes and fruits were the principal gifts (Martialis, viii. 33; xiii. 37; Seneca, Epistulae, Ixxxvii.). It seems to have been under Augustus that money took the place of these eatables. The custom still prevailed about A.D. 400 under the Emperors Arcadius and Honorius.
* Suetonius, in Augustus, chap. Ivii. ; in Tiberius, chap, xxxiv. ; in Caligula, chap. xlii. Compare Preller, Roniische Mythologie, p. 161.
^"Cavam manum asses porrigentibus praebens," Ibid., in Augustus, chap. xci.
THE CALENDS OF JANUARY 85
as to publish an edict expressly requiring such gifts, and to stand in the porch of the palace, on the Calends of January, in order to receive those which people of all descriptions brought to him.^ It was reckoned a handsome enough way of receiving gifts, when the bosom- fold of the cloak was expanded; but when they were received with both hands hollow,^ or in "goupins," to use the Scotch word, it was accounted objectionable. Hence rapine was proverbially expressed in that manner.^
But the celebration of the Calends of January was by no means the only festivity of that time of the year in ancient Rome ; there was a whole series of festivals, so that Seneca ( + a.d. 39) could write to his friend Lucilius : " It is now the month of December, when the greatest part of the city is in a bustle. Loose reins are given to public dissipation; every- where may you hear the sound of great preparations, as if there were some real difference between the days dedicated to Saturn and those for transact- ing business. Thus, I am disposed to think, that he was not far from the truth who said that anciently it was the month of December, but now the year. Were you here, I would willingly confer with you as to the plan of our conduct ; whether we should live in our usual way, or, to avoid singu- larity, both take a better supper and throw off the toga. For what was not wont to be done, except in a tumult or during some public calamity to the city, is now done for the sake of pleasure, and from regard to the festival Men change their dress. It were certainly far better to be thrifty and sober amidst a drunken crowd, disgorging what they had recently swallowed." *
These festivals were the Saturnalia, ^dth their equality between rich and poor, freemen and slaves, and their presents of all descriptions,^ lasting from December 17 to December 19; or seven days, from December 17 to
^ Suetonius, in Caligula, chap. xlii.
^ " Utraque manu cavata."
^ Ammianus Marcellinus, Lib. XVI. ; Rosin, AntiquUates, p. 29.
* Seneca, Epistulae, xviii. ; Jamieson, Etym. Diet, of Scot. Lang., "Yule," IV.
^ " Cereos Satumalibus muneri dabant humiliores potentioribus, quia candelis pauperes, locupletes cereis utebantur," Festus Pompeius, Lib. III. The new year's gift was called Kaletidaticum. Ducange, Glossarium, explains " Kalendaticum praestatio quae Januarii
86 YULE AND CHRISTMAS
^ December 23. All labour rested, and, under the call lo Saturnalia! lo Saturnalia I people gave themselves to a wild joy. Then followed the Brumalia, fixed by Caesar erroneously on December 25, the alleged shortest \ day of the year, called since that time occasionally Dies Invicti Solis, day j of the unconquered sun. The character of Saturnalia, Brumalia, and | Kalendae Januariae was very wild and lascivious, so wild that, together with the Matronalia of the first of March (and sometimes with the Septi- montium, the feast of incorporation of the seven hills with the city of Rome, also celebrated in December), they were, by the fathers of the Christian Church, regarded as a perfect essence of heathendom, which was by no means meant to be a compHment. So Tertullian ( + a.d. 220) could say: "By us who are strangers to sabbaths and new moons, once acceptable to God, the Saturnalia, and the feasts of January, and Brumalia, and Matronalia, are frequented; gifts are sent hither and thither; there is i the noise of the Strenae, and of games and feasting. O ! better faith ; of the nations in their own religions, which adopts no solemnity of the Christians."^
Kalendis iiebat." Charta Rogerii Siciliae Regis an. I137, apud Falconem Beneventantim, p. 315: "Angarias, terraticum, herbaticum, carnaticum, kalendaticum, vinum, olivas, relevum, etc. KaXaj'5t(c6»' in Justiniani edicto xiii." In England, as late as the thirteenth century, the custom of benevolences exacted by kings was connected with the Roman Calends custom : " Rex autem Regalis magnificentiae terminos impudenter transgrediens, a civibus Londinensibus, quos novit ditiores, die Circumcisionis Dominicae, a quolibet exegit singulatim primitiva, quae vulgares nova dona novi anni superstitiose solent appellare." The king is Henry III., A.D. 1249 (Matthaei Paris, Monachi Albanensis Angli, Historia Major, London, 1640, p. 757, under the year 1249). Matthew Paris goes on to say: "Veruntamen festo beati Aeduwardi, quod est in vigilia Epiphaniae, appropinquante, vocavit dominus Rex per literas suas copiosam Magnatum multitudinem : ut simul cum eo, qui in vigilia sancti, videlicet die Lunae in pane et aqua et in veslibus laneis jejunaverant, prout de more solet, ipsum festum magnifice celebrarent in ecclesia S. Petri apud West- monasterium. "
^ Tertullian, De Idolatria, chap. xiv. These sweetmeats, called by the name of Strenae, were therefore prohibited by the early church (V. Rosin, Antiquitates, p. 29). The Strenae are traced as far back as to King Tatius, who at this season used to receive branches of a happy or fortunate tree from the grove of Streniae as favourable omens with respect to the new year. In another passage {De Idolatria, chap. x. ) Tertullian says: "Saturn- alia, strenae captandae, et septimontium, et brumae, et carae cognationis honoraria exigenda omnia." Compare also Tertullian's De Fuga in Persecutione, chap. xiii.
f
THE CALENDS OF JANUARY 87
With the introduction of the Julian calendar all kinds of southern Calends rights found entrance to the Germanics : the making of processions j through the streets and singing of songs, the lighting of candles and lamps, the adorning of their houses with laurel and green trees, the giving of presents, men dressing up in women's garments, masquerades in the hides of animals, and the erection of a table of fortune for the good luck of the new year.^
Half a century before the beginning of our era the first Roman legion had entered Great Britain, and not much later British soil was in constant occupation by Roman legions. The great mass of Roman inscriptions found in Britain gives ample evidence as to their sojourn there. It is not astonishing that among these we find one devoted to the "God The Unconquered Sun " {Deus Invictus Sol), ^ which further supports the general assumption that these legions did not only celebrate the Calends of January, but the Brumalia as well, and a fortiori the Saturnalia. The exact date when the Roman legions were withdrawn from Great Britain is not known, but there is no doubt that Roman civilization and Roman religious tradition survived them there, so that when Augustine and his Roman fellow-missionaries of Christianity landed in Britain (a.d. 592) they found there December 25 as a day marked in the festive calendar, at least
^ Acta Concilioruvi, Parisiis, 1714, Vol. III., col. 365, Conciliuin Turonense, II., A.D. 567, xxii. : " Enim vero quoniam cognovimus nonnuUos inveniri sequipedas erroris antiqui, qui Kalendas lanuarii colunt cum lanus homo gentilis fuerit : rex quidem, sed Deus esse non potuit." And eight years later: " Non liceat iniquas observationes agere Kalendarum, et otiis vacare gentilibus, neque lauro aut viriditate arbonim cingere domos. Omnis haec obsevatior paganismi est." (Caput Ixxiii. of the Capitula Martini Episcopi Bracensis, circa A.D. 575.) Ibid., Vol. III., col. 399.
^ Alonutnenta Historica Britannica in the chapter, "Ex Inscriptionibus Excerpta de Britannia," p. 116, No. 103: "Deo Invicto Soli Soc Sacrum Pro Salute Et Incolumitate Imp. Caes. M. Aureli Antonini Pii Felic. Aug. L. Caecilius Optatus Trib. Coh. I. Vardul Cum Con* braneis Votum. Deo * * A Solo Extruct * * *" (Riechester or Rochester, Northumberland). The Inscriptions, No. 102, " Deo Invicto Herculi Sacr. L. Aemil. Salvianus Trib. Coh. I. Vangi V.S. P.M." (Risingham, Northumberland), and No. 75, " Silvano Invicto Sacrum C. Tertius Veturius Micianus," etc., show, however, that Invictus was a rather general divine predicate, which excludes the possibility of interpreting Inscription No. 103 as dedicated to the sole unconquered God, taking Wz as the dative singular of solus.
88 VULE AND CHRISTMAS
of the south of England, which naturally paved the way for a celebration of Christ's Nativity on the same day.
As regards the continental Germanics, as late as the middle of the eighth century there existed a perfect unity of popular sacramental usage as to the Calends of January between them and the Romans, the German Calends rites not only resembling the Roman ones absolutely, but even being felt to be identical with them by the people celebrating them. We know this from an incident of a.d. 742. In that year, Winfrid (Bonifacius), the " apostle of the Germans," in a letter written to Pope Zacharias, com- plained of a strange fact which hindered his getting on better in sowing the gospel in the souls of the Alamanni, Boioarii, and Franci. For when he interdicted them from certain heathen customs, they justified themselves by the excuse that they had seen similar things at Rome, close to St. Peter's Church, where these things were regarded as perfectly permissible. And they told Boniface they had seen that every year on the eve of the Calends of January, after heathen custom, processions went through the streets singing unchristian songs and using heathen exclamations, that people erected tables of fortune and kept a Calends fire from which they would not give anything away, just as they refused to lend anything else to their neighbours during that time, and that women went publicly about