The Dance of Death in Medieval Literature: Some Recent Theories of Its Origin Author(s): James M.

Clark Source: The Modern Language Review, Vol. 45, No. 3 (Jul., 1950), pp. 336-345 Published by: Modern Humanities Research Association Stable URL: http://www.jstor.org/stable/3718509 Accessed: 30/11/2010 07:01
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=mhra. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Modern Humanities Research Association is collaborating with JSTOR to digitize, preserve and extend access to The Modern Language Review.

http://www.jstor.org

THE DANCE OF DEATH IN MEDIEVAL LITERATURE

SOME RECENT THEORIES OF ITS ORIGIN Francis Douce and Gabriel Peignot were the pioneers of the systematic investigation of this subject. They were the first to attempt a comprehensivesurvey and to devote a whole book to it. They cleared the ground by disposing of many insidious errors; they collected a large amount of material and paved the way for later inquiry. But they were both concerned in the main with the plastic arts. For them the Dance of Death was a theme for mural paintings, sculptures, engravings, and so forth. They were not particularly interested in the evolution of the genrenor with the question of origins. Curiously enough it was a French artist and engraver named E. H. Langlois who drew attention to the literary aspect of the subject and advanced the theory that the Dance of Death originated in a drama. This hypothesis was skilfully developed by Wackernagel.2 He grouped the Dance together with those dramatic representations which were separated off from the Passion and Resurrection cycles, for instance the Corpus Christi and Annunciation plays. They came into being in the late Middle Ages and remained at the embryonic stage, being merely a series of similar speeches and situations. Wackernagelthought that there was an autochthonous development of the Dance of Death on German soil from the earliest phase (the sermon) until it reached its zenith. One of his reasons for so thinking was that in the Liibeck Totentanz(c. 1463) the word spectel occurs. He took it in the sense of' spectacle', 'play', but it might just as well mean 'spectacle', in the sense of 'sight', and it apparently corresponds to the phrase danse macabrein the text of the Innocents. A substantial contribution to our subject was made by Wilhelm Seelmann.3 He has his small lapses: for example, his French texts are not always free from error. His general method may not be infallible: he acted on the assumption that all examples are ultimately derived from one archetype, the Urtotentanz. But he approached the problem in a truly scientific manner, he took nothing for granted, tested all statements, even those that had been accepted without question for generations. His lists of examples were more complete than any previous ones. It was Seelmann who fully clearedup the mystery of the so-called Minden Totentanz of 1383. He showed by new arguments that the Kleinbasel example had been dated much too early and he traced it to a French source. These two discoveries removed the two strongest arguments for the German origin of the Totentanz. He took over from Wackernagel the theory that the Dance of Death developed out of a drama, but he asserted that the growth did not take place ab ovoin Germany. A French morality play was the starting-point of the Spanish Danza de la muerte, the Paris Danse macabreand the Liibeck Totentanz. All the texts were ultimately derived from this one source. The importation into Germany took place via the Netherlands and Luibeck. Seelmann's views found general acceptance for some fourteen years. Male adopted them in his brilliant survey of French medieval art. Creizenachembodied
1 Essai historique et pittoresque sur les Danses des Morts (Rouen, 1851), p. 138. 2 Zeitschriftfur deutschesAlterthum,ix (1853), 302-65. 3 Die Totentdnzedes Mittelalters (Norden, 1893).

JAMES M. CLARK

337

them in the standard work on the history of the Germandrama. But the theory of a play as the origin of the Dance of Death was subjected to such vigorous and repeated attacks in Germany that it was almost abandoned by experts in that country. Wilhelm Fehse began by attacking the drama theory as it was known to him, that is to say, as it had been expounded by Wackernagel and Seelmann. He was not aware of the evidence for the existence of a Dance of Death drama in Normandyl and Scotland2 long before the pictures at the Innocents were painted or the verses inscribed there were written. Fehse argued that the Dance of Death originated in churchyard superstitions, and was in its early phase a Dance of the Dead. He states his point of view tersely: Form des Totentanzes,sondern der Reigen der Tanz auffordert,ist die urspriingliche Toten, die die Lebendenin ihren Reigen hineinziehen.3 He naturally regarded those texts and pictures as the oldest in which this alleged primitive conception was most clearly exhibited. If the skeletons in the pictures and poems depict individuals, this excludes the dramatic origin, because in a morality play Death would be one, not many. Fehse does not allow sufficiently for the possibility of the adaptation of the original idea to a new medium. In order to follow the second stage of the argument, we must make a slight digression. Under the two series of paintings at Basel were inscribed German stanzas of four lines. The two texts were two very similar versions of the same poem; the Klingenthal text was naturally the oldest, the Grossbaselone was copied from it.4 The pictures at Klingenthal were directly or indirectly inspired by those at Paris, but it is not so easy to account for the accompanying verses, which are not a translation of the Danse macabreof the Innocents. Fehse set himself the task of investigating their origin. He showed that the poem inscribed at Basel is also found in manuscripts at Munich, Heidelberg, and Berlin, and in a Heidelberg block-book printed about 1465. The poem is generally known as the Upper German Quatrain text (oberdeutscher vierzeiliger Totentanztext),in order to distinguish it from the eight-line texts to be found in German block-books, and from the poems recorded at Paris, Liibeck and in Spain. The title is a little clumsy, but may be retained for the sake of convenience. With the aid of textual criticism, Fehse succeeded in proving that all existing versions of the Quatrainpoem are derived from a type representedby the text in the Heidelberg manuscript, which was written between 1443 and 1447. In the same codex there is a Latin poem, which had been hitherto regarded as a translation of the German one. By a masterly piece of research, Fehse demonstrated that the opposite was the case; the Germanversion is translated from the Latin, as far as the speeches of the living are concerned. Those of the dead are a later addition.5 The Latin poem is very short and consists of speeches by the living only, with a prologue and epilogue in the form of a short sermon in verse. One distich is allotted to each person. He names himself, gives a brief designation of his rank or
I agree with Dr Pfister who dates them c. 1440 France(Paris,1908),p. 391. 2 'Lettrede M.C.Leber', pp. 50-1, in Langlois, and c. 1480respectively. 'Der oberdeutsche Totentanztext', vierzeilige op. cit. 3 4 The Klingenthalpoem is in the Alemannic
M.L.R. XLV

Nicht das Drama vom Tod, der die einzelnen Vertreter der menschlichen Stande zum

1Mle, L'Art religieux de la fin du moyen dge en

dialect, that of Grossbaselis in the Kanzleisprache.

'Das Totentanzproblem', Zeitschrift fir deutschePhilologie, xLrr (1910), 261.

Zeitschrift fir 67-92.

deutsche Philologie,

XL

(1908),

22

338

The Dance of Death in Medieval Literature

character, and bewails his fate. Death (or the dead) does not address the living, nor reply to them. The poem is well written, by a writer who has a perfect command of rhyming Latin verse. It is found in the Heidelberg manuscript and nowhere else. Comparingthe Latin and the German texts, Fehse writes:
Es stellte sich heraus... daB die lateinische Fassung die altere ist. Diese braucht nun naturlich nicht auf deutschem Boden entstanden zu sein. Offenbar entstammt dieser lateinische Totentanz der geistlichen Literatur des Mittelalters, die auf internationalem Boden steht.

In another passage, however, he suggests that the dominant idea of the Dance of Death, the churchyard revels of the dead, is taken from Germanic folklore. This is mere speculation. All that can be said to substantiate the idea that the Latin poem
is linked up with these superstitions is the mention of 'companions of Death' (mortis consortes), 'misshapen beings' (distorti), and 'dancers' (corisantes). It is quite possible that the poet had seen such a cycle of pictures as that of the Innocents and that he was recording his impressions. Fehse attempts to prove that this Latin poem goes back to the fourteenth century. The oldest manuscript is dated by Fehse himself between 1443 and 1447, and he is not able to bring forward a scrap of evidence that the poem is half a century older. The question of chronology is important because Fehse would have us believe that the French Danse macabre is derived from the Latin poem in its present form or in an earlier one. But if we assume, propter argumentum, that the Quatrain poem, in its original Latin shape, does actually go back to the fourteenth century, can we accept it as the Urtotentanz, or even as one of the supposititious Urtotentdnze? It is a series of monologues in which the various characters introduce themselves to their audience, as it were. This poem is really nothing but a variant of the well-known Latin Vadomori poem. Fehse himself practically admits this.' He tells us that the lost original of the Latin text must have been illustrated, because in the Heidelberg manuscript there is an inscription referring to a codex albus 2 with pictures. The Latin poem itself contains the phrase 'pictura et exempli figura', and at the end we read 'item alius doctor depictus in opposita parte'. But there is nothing to indicate when the codex albus was written or what kind of pictures it contained. It is essential for Fehse's argument (1) that the Upper German Quatrain text (or the Latin hexameters of which it is a translation, in part at least) should be older than the lines at the Innocents (1424-25); (2) that the Paris text was derived from the Quatrain verses or the hexameters. Fehse takes the first point more or less for granted, and attempts to prove the second. Let us deal with them in turn. When were the Upper German Quatrain poem and its Latin original written ? The best manuscript is, on Fehse's admission, not older than 1443. But he seems convinced that the poem goes back to the fourteenth century, and in support of this contention he quotes Massmann. It is instructive to examine Massmann's remarks.3 He says that the type of rhyme and the tone of the whole poem may be assigned to the fourteenth century ('Das in dem handschriftlichen Texte herrschende Reimgesetz laBit sich sehr natuirlich in das vierzehnte Jahrhundert zuriickverweisen, wohin auch der Grundton des Ganzen deutet'). Massmann wrote in 1847, at a time when modem language scholarship was in its infancy. One has only to examine his editions of the various texts to see that they
1 Loc. cit. p. 278.

3 Loc. cit. pp. 104-5.

Massmann, Die Baseler Todtentdnze(Stuttgart, 1847), p. 123.

JAMES M. CLARK

339

would not pass muster to-day. Fehse's criticism of them is sufficiently scathing. The observation about the tone of the whole poem is not at all convincing. One fails to find anything in it that could not be described as characteristic of the fifteenth century. It is an admirable commentary on the pictures which were painted in that period. The remark about the rhymes sounds more plausible. The examples given are: sin: hin; rich: gewalteclich.But we should not forget that to this very day in Basel, these rhymes are perfect. In those where even the educated speak Schweizerdeutsch, of the Kleinbasel text which Fehse quite correctly pronounces to be interparts polations, the same rhymes occur. The argument is hence invalid. Massmann gives an additional reason for thinking that the poem was archaic in language. He observes that Death addresses all the charactersin the plural (ir, not du), with the exception of the peasant and the child. This is quite irrelevant, because, as Massmannhimself informs us, the Emperor Sigismund, who died in 1437, did the same thing. All this applies to the German Quatrains. Massmann has nothing to say about the age of the Latin hexameters. The reason is that he was of the opinion that the Latin was translated from the German, which Fehse disproved. We now come to the second point in Fehse's thesis, namely the implication (for in its final form he hedges about his statement in various ways) that the verses at the Innocents are subsequent to, and derived from, the German Quatrains or the Latin hexameters. It is not disputed that the Liibeck, Paris and Spanish examples have a certain family resemblanceand are presumably derived from the same source. The opening lines of the Liibeck and Paris texts correspond exactly and are very similar to the Upper German version: Paris: O creatureroysonnable. Liibeck: O redelikecreatuer. Basel: O diser welt weisheit kint, Alle, die noch im leben sint. Latin: O vos viventes, huius mundi sapientes. Seelmann drew the conclusion that the French poem was the archetype, and the Low Germanof Liibeck and the Latin of the Heidelbergmanuscriptwere translations of it. That is a reasonable inference. But Fehse objects that 'huius mundi sapientes' is a quotation from I Corinthiansi. 20, and must therefore be original. Consequently, the French and Low German are translated from the Latin (or the German). This is indeed a slender thread on which to hang an argument! The Biblical passage runs in the Vulgate: 'Nonne stultus fecit Deus sapientiam huius mundi.' The writer, or translator, of the Latin poem may have had this sentence in his mind, but surely it is possible to mix up a quotation with a translation, especially if one is writing leonine hexameters. From viventesone passes easily enough to the rhyme-word sapientes. Fehse's theory can be best summed up in the form of a tree: Vadomoripoem (fourteenthcentury) Latin text (without the phrasevadomori)+ pictures Germanillustratedpoem+ Versesof the Dead I|{~~~ ~~Verses
Basel Totentanz

French didactic poem+ of Death

(c. 1400)

Danse
macabre

Spanish
Danza

Liibeck
Totentanz

+ pictures

+ pictures
22-2

340

The Dance of Death in Medieval Literature

To sum up: the Latin poem in the Heidelberg manuscript is the chief but not necessarily the only source of the Upper GermanQuatraintext, of which two of the later versions were used at Basel. There is no reason to believe that the Latin poem influenced the Danse macabre. The opposite is just as likely. Fehse's theory has been discussed at some length because it constitutes a definite challenge to the priority and supremacy of the French work at the Innocents, and it was therefore desirable to dispose of it in detail.
Wolfgang Stammler's work Die Totentdnze des Mittelalters (1922) is a most

elaborate explanation of the origin of the Dance of Death. Stammler recognizes the weaknesses of the existing theories and concludesthat any solution of the problem must be eclectic. Following Seelmann, he is convinced that the Spanish Danza de la
Muerte, the Paris Danse macabre and the Liibeck Totentanz come from the same

source. Unlike Seelmann, he refuses to accept the drama theory: 'Niemals dagegen ist, nach den vorhandenen Zeugnissen zu urteilen, einer der mittelalterlichen Totentanztexte auf der Biihne verkorpert worden.'1 From Fehse he adopts the idea that the Dance of Death is based on medieval superstitions and represents the midnight revels of the dead in the churchyard. But this does not account for the Spanish poem, in which there is no suggestion of the dead leading away the living, but only of Death leading them off. Stammler therefore assumes that, besides the main current of evolution, there was a second minor channel,in which Death is merely an allegoricalfigure. He postulates the influence or interference of the Legend Des trois mortset des trois vifs and of the Vadomoripoem and draws up a complicated pedigree for the Dance of Death itself, which has the following broad outlines:
Vadomori Verses of the Living Des trois morts et des trois vifs Verses of the Skeletons Latin poem on the Dance of Death Dispute between Death and Life

Latin Dance of the Dead

Upper German Quatrain Poem Danse mcabre High German translation Low German translation

Up~/~r Gm\
z

QSpanish

muerte

Danza de la

Dutch Dance of Death Libeck-Reval Totentanz

1 Reallexikon der deutschen Literaturgeschichte (Berlin, 1928-9), in, 382.

JAMES M. CLARK

341

It will be seen that in some of the details of this scheme, Stammler follows Fehse. What is quite new is the combination of the Legendand the Vadomori.x, y and z are the inevitable missing links. This ingenious theory explains many difficulties, but it leaves as many unsolved and Des troismortsare, and it rests on several unwarrantableassumptions. Vadomori in a certain sense, forerunnersof the Dance of Death, but can we say that they ever actually combined? The Legendmay, or may not, have suggested the idea of making the skeletons speak, but what version of the Legendcontains speeches that developed into those of the Latin poem? The speeches of the skeletons in the Upper German Quatrain poem are later additions and there is no reason for thinking that they were taken from the Legend. The onus of proof is on the inventor of the theory. Stammler traces the Spanish version to a different source from that of the Upper German Quatrain poem because of the different conceptions of death: in one an allegorical figure, in the other a corpse risen from the grave. The first is a Dance of Death, the second a Dance of the Dead. As the Spanish poem is not illustrated, Stammler considersits source to be an allegorical poem. This does not allow for the possibility of originality on the part of the poet, who may have created something quite new out of his material. That a Latin dispute between Life and Death existed is quite possible, but how did the dance motive get into it? This motive is very much in evidence in the Danza de la Muerte;it is present in the title and in almost every verse of the poem. Was it the addition of the poet? If so, cannot we give him credit for a little more inventiveness The division of the characters into classes with the separation of clerics and laymen is common to the Spanish poem, the Lufbeckversion and the French. It was therefore, if we accept Stammler's reasoning, present in the source, the Latin poem on the Dance of Death. How did these conceptions find their way into the an extraordinary coincidence if two Latin poems arose quite independently of each other, both having the division into classes and the twofold hierarchy, and both expressing the equality of all in death under the allegory of a dance. Yet this is precisely what Stammler would have us believe. We may ask how it comes about that in the Spanish and French poems, no less than in the Upper GermanQuatrainversion, there are the introductory words of the preacher. Is this also a mere coincidence? Is the similarity of the persons fortuitous ? In the Danza de la Muertethey begin: pope, emperor, cardinal,king, patriarch, duke, archbishop, constable, bishop, knight, abbot. In the Upper German poem, if we leave out the empress, an evident addition (this is the only female character in the first sixteen persons), we have pope, emperor, king, cardinal, patriarch, archbishop, duke, bishop, count, abbot, knight. In short, the contention that the Danza de la Muerte arose independently of the Danse macabrecannot be upheld. The Spanish poem was inspired by a French original. The evolution of the text of the Dance of Death cannot be entirely separated from that of the mural paintings, although we are only indirectly concerned with it here. He is at great pains to show that the pictures at La Chaise-Dieu are older than those of the Innocents at Paris. He does not appear to have seen the former and only knows them from faulty reproductions. His only authority for the date of La Chaise-Dieu example -is a Halle thesis on Old French
Dispute between Life and Death, or how were they superimposed on it? It would be

342

The Dance of Death in Medieval Literature

costume.1 Stammler forgets that Holbein used for his countess in the Totentanzthe dress of a French countess who had lived a century earlier. To date a series of paintings by the evidence of some of the costumes without corroborative data is a trifle hazardous. The movement of ideas in France in the fifteenth century was not from Auvergne to Paris but in the reverse direction. However much we must admire the skill with which Stammler develops his thesis, we cannot agree with his conclusions. The fundamental fallacy of his method, as of Fehse's, is that of assuming an archetype (or Urtext), from which all later recensions are derived, and further, of assuming that there are clearly marked stages in the evolution. If one of these is not to be found, we are told that it once existed, and has been destroyed. Poetry and art do not develop with the precision of a mathematical proposition. A poet who is instructed to provide a Dance of Death text may give a translation, a very free adaptation, or an original production. It all depends on the man himself. The artist is equally incalculable. There is another pertinent criticism of this type of research, which works so much with abstractions and not sufficiently with facts. It is easy to analyse a work of art, classify the examples, put them into water-tight compartments, as it were. But this method does not sufficiently allow for cross-currents. One instance will illustrate this point. Mantels drew attention to a Low German dialogue between Death and Life in a printed book of the late fifteenth century.2 It is strongly influenced by the verses of the Liibeck Dance of Death. In its turn, this dialogue influences the blockbooks of the late fifteenth century. This happened in other cases too, and if all the data were available the best constructed hypotheses would come toppling down. Gert Buchheit's book followed close on the heels of the work we have just been discussing. The author evidently set out with the object of writing the standard workon the subject. The bookis well illustratedand containsan excellent bibliography. Unfortunately the arrangement leaves much to be desired. There are several inaccuracies in points of detail. For instance Buchheit quotes from what he calls 'der oberdeutsche Urtext', without in any way indicating whose text it is, or how it was arrived at. It is not Fehse's reconstructed text of the Upper GermanQuatrain poem, because it contains readings from the later and inferior manuscript H2. Yet we are told that it is taken from a version which is the source of the Heidelberg manuscript (presumably H1) and of the Kleinbasel and Grossbasel texts. as Further, Buchheit refers to Schroer's Totentanzspriiche 'der vierzeilige oberdeutsche Totentanztext'. This is very misleading, because it would inevitably lead to confusion with the Upper German Quatrains. Moreover, its text, as edited by Schroer, is certainly Middle German, not Upper German.3 Buchheit admits that he wrote the greater part of his book before the appearance of Stammler's work. As the latter brought forwarda new theory of the origins which seemed convincing, Buchheit embodied it in his book, but left older portions unaltered. His account of the subject is therefore not free from a certain confusion. In some portions, evidently written before he had read Stammler, he maintains that the pictures were ultimately derived from a Dance of Death drama, which developed
1 P. Post, Die franzosisch-niederldndische Mannertracht einschliefilich der Ritterrustung im Zeitalter der Spdtgotik, 1350-1478 (Diss., Halle, 1910). 2 'Zwiegesprach zwischen dem Leben und dem Tode', Jahrbuch des Vereins fur niederdeutsche Sprachforschung, Jahrgang 1874, pp. 54-6; Jahrgang 1876, pp. 131-3. 3 This is proved by the use of i and u for ie and uo, the prefix vor for ver, hilt for hdlt, orteil for urteil, etc.

JAMES M. CLARK

343

out of a sermon ('DaB auch die bildlichen Darstellungen letzten Endes auf der Predigtszene beruhen').l The sermon and the poem based on it were in time superseded by mural paintings and printed books. This is, of course, Seelmann's point of view. Buchheit adds that the pictures probably came from the North of France,2 but that the region where the Dance of Death came into being is unknown.3The real problem, the combination of death and the dance, is, he assures us, still unsolved.4 On the other hand, following Stammler, he says that the oldest Dance of Death depicts the nocturnal revels of the dead as depicted by the French clergy from the thirteenth century onwards. It was, he continues, strongly influenced by the poem. There is a second form Legend Des troismortset des trois vifs and the Vadomori of the Dance of Death, which grew out of a dialogue between Death personifiedand man. Spain is the home of the didactic poem and the Paris Danse macabre shows the influence of the Danza de la muerte! The Spanish version was also, he informs us, the ultimate source of the Liibeck Totentanzpictures. Apparently Buchheit thinks that the poem, which is without illustrations, went north from Spain to the Netherlands, and there coalesced with the pictures (without text), which came from the north of France. In the Danse macabreof Paris he thinks that Spanish and northern French influence are also combined. All this is in the highest degree improbable. Buchheit commits the fundamental error of assuming that the Dance of Death went through a gradual evolution of which the different stages can be isolated and defined. Thus, he traces the growth of satirical tendencies, taking as his example the dialogue between Death and the cardinal. He begins with the Upper German Quatrainpoem, in which there is no suggestion of censure, quotes the Liibeck printed book, the Liibeck mural paintings, the Berlin Totentanz, in which the idea of punishment is definitely mentioned, the Ltibeck text of 1489, and finally Schroer's and the Upper German eight-line stanzas (Der Doten dantz), in Totentanzspriiche which the cardinal compares himself with a highwayman. It has already been made abundantly clear that the Spanish Danza de la muerte, which, according to Buchheit, is the ultimate source of the text of the Liibeck Totentanzof c. 1460, and which he places in the first half of the fifteenth century, is strongly anti-clerical. Surely the satirical tendency of the Dance of Death is just as much a matter of the milieu as of the date. The stern ascetic atmosphere of the friary lends itself to satire more readily than the environment of the smugly complacent citizen. The severity of the Danse macabre contrasts strongly with the mildness of tone of the Liibeck mural Totentanzof c. 1460. If one wished to dogmatize at all, one would say that in the course of time the satirical element in the Dance of Death tended to decline, rather than to increase, but all such generalizations are bound to be inaccurate. Italian scholars were late in the field, but their contribution is none the less substantial in quantity and good in quality. They have undertaken the task of bringing to light and relating to each other all traces of the Dance of Death in Italy. They have shown that the alien conception is expressed in a typically national idiom in Italian literature and art. The general history of the subject was related by Vigo. There are also several useful monographs by other writers. The most ambitious effort yet planned in our field is that of a Hungarian priest,
1 P. 34, see also pp. 35-6. 2 P. 49.
3

P. 107. P. 39.

344

The Dance of Death in Medieval Literature

Father Istvan Kozaky.l The Dance of Death did not play a large part in the literature or art of the Magyars, and the investigation of this part of the general theme occupies a correspondingly small portion of the whole work. But Father Kozaky includes in the scope of his labours the literature of death in a more general sense. Only the first part of his magnum opus has appeared as yet. He elaborates a genealogical tree of types of texts that is so complex that Stammler's scheme is in comparison simple. The biographical material is also copious. The Legend of the Three Dead and the Three Living receives detailed treatment. In order to make his researches available to a larger public, the author adds a German translation to the original Hungarian. As the first volume is largely introductory, except so far as Hungary is concerned, it is scarcely possible to judge whether Father Kozaky has new views to propound on the Dance of Death in the strict sense of the term. Some of the soundest research on the subject in recent years was done in America. Miss Eleanor P. Hammond has given us a careful edition of the French Danse macabre and John Lydgate's translation.2 In an earlier paper3 she had done much to advance our knowledge of the Vadomori poem and had edited two recensions. She did more than this: she put forward the idea that the multifarious poetry that comes under the general heading of Dance of Death literature does not necessarily go back to one archetype. If this point had been fully grasped, much error by other writers might have been avoided. Miss Hammond wisely refrained from sweeping generalizations, but she distinguished two types of works: the poem with a hierarchical order and the disputation between Life and Death. Many impulses were at work in the fourteenth and fifteenth centures-so many, that it is not yet possible to disentangle them and say: Here at this point, arose the Dance of Death. One mental habit of the Middle Ages expressed itself in lists and classifications; another and more widespread, the fondness for contrast and argument, expressed itself in debates... .And upon each of these larger tendencies the immediate Death-interest of the fourteenth and fifteenth centuries could find a point of attachment. The final conclusion at which one arrives is that nothing has been written that shakes the claim of the poem at the Innocents, dating from 1424-5, to be the oldest known literary form of the Dance of Death. All attempts to prove that the Dance originated elsewhere have so far failed. The mysterious reference in Le Repit de mort, written by Jean Lefevre about 1376, 'Je fis de Macabre la danse', may or may not mean that he had been seriously ill and near to death. Gaston Paris thought that the reference is to a poem. It is possible, but in the absence of further evidence it remains a mere possibility. We cannot reject the clear statements of the medieval historian Bower in Scotland4 and the Abbe Miette in France that there was a play in which Death summoned various mortals representing different classes of society to the grave. This drama may have been in dumb show; neither of our informants makes any mention of speeches. This may also apply to the later performances at Douai in 14495 and
1 Istvin Kozaky, A HaldtdncokTotenete(Budapest, 1936; Biblioteca Humanitatis Historica a Muso Nationali Hungarico Digesta, I). The sections devoted to the Dance of Death in Hungary are pp. 14-17, 26-35. 2 English Verse between Chaucer and Surrey (London, 1927), pp. 124-42, 418-35.
3 'Latin Texts of the Dance of Death', Modern Philology vmII(1911), 399-410. 4 Joannis de Fordun Scotichronicon, ed. W. Goodall (Edinburgi, 1759). 5 Langlois, op. cit. I, 292.

JAMES M. CLARK

345

Besan9on in 1453.1The Scottish reference clearly points to France as the country of origin, because the play was acted at the marriage of Joleta, daughter of the Count of Dreux, to King Alexander of Scotland. Miss Hammond shows the connexion between the Vadomoripoem and a mime with a single 'recitator'. The exact relationship between the Dance of Death drama and the verses and pictures at the Innocents remains uncertain.2
JAMES M. CLARK GLASGOW 1 Ducange, Glossarium Medice et Infimce Latinitatis, t. iv, p. 158, s.v. MachabseorumChorea. 2 I have dealtmore fully withthewhole question of the Dance of Death both in literature and in the plastic arts in The Dance of Death in the Middle Ages and the Renaissance, Glasgow: Jackson, Son & Co., 1950. (Glasgow University Publications, lxxxvi.)