The Travels of Marco Polo — Volume 1

This second and main portion of the Work is in its oldest forms undivided, the chapters running on consecutively to the end.[1] In some very early Italian or Venetian version, which Friar Pipino translated into Latin, it was divided into three Books, and this convenient division has generally been adhered to. We have adopted M. Pauthier’s suggestion in making the final series of chapters, chiefly historical, into a Fourth.

51. As regards the language in which Marco’s Book was first committed to writing, we have seen that Ramusio assumed, somewhat arbitrarily, that it was Latin; Language of the original Work.Marsden supposed it to have been the Venetian dialect; Baldelli Boni first showed, in his elaborate edition (Florence, 1827), by arguments that have been illustrated and corroborated by learned men since, that it was French.

That the work was originally written in some Italian dialect was a natural presumption, and slight contemporary evidence can be alleged in its favour; for Fra Pipino, in the Latin version of the work, executed whilst Marco still lived, describes his task as a translation de vulgari. And in one MS. copy of the same Friar Pipino’s Chronicle, existing in the library at Modena, he refers to the said version as made “ex vulgari idiomate Lombardico.” But though it may seem improbable that at so early a date a Latin version should have been made at second hand, I believe this to have been the case, and that some internal evidence also is traceable that Pipino translated not from the original but from an Italian version of the original.

The oldest MS. (it is supposed) in any Italian dialect is one in the Magliabecchian Library at Florence, which is known in Italy as L’Ottima, on account of the purity of its Tuscan, and as Della Crusca from its being one of the authorities cited by that body in their Vocabulary.[2] It bears on its face the following note in Italian:—

As far as I can learn, the age which this note implies is considered to be supported by the character of the MS. itself.[3] If it be accepted, the latter is a performance going back to within eleven years at most of the first dictation of the Travels. At first sight, therefore, this would rather argue that the original had been written in pure Tuscan. But when Baldelli came to prepare it for the press he found manifest indications of its being a Translation from the French. Some of these he has noted; others have followed up the same line of comparison. We give some detailed examples in a note.[4]

52. The French Text that we have been quoting, published by the Geographical Society of Paris in 1824, affords on the other hand the strongest correspondingOld French Text published by the Société de Géographie. proof that it is an original and not a Translation. Rude as is the language of the manuscript (Fr. 1116, formerly No. 7367, of Paris Library), it is, in the correctness of the proper names, and the intelligible exhibition of the itineraries, much superior to any form of the Work previously published.

The language is very peculiar. We are obliged to call it French, but it is not “Frenche of Paris.” “Its style,” says Paulin Paris, “is about as like that of good French authors of the age, as in our day the natural accent of a German, an Englishman, or an Italian, is like that of a citizen of Paris or Blois.” The author is at war with all the practices of French grammar; subject and object, numbers, moods, and tenses, are in consummate confusion. Even readers of his own day must at times have been fain to guess his meaning. Italian words are constantly introduced, either quite in the crude or rudely Gallicized.[5] And words also, we may add, sometimes slip in which appear to be purely Oriental, just as is apt to happen with Anglo-Indians in these days.[6] All this is perfectly consistent with the supposition that we have in this MS. a copy at least of the original words as written down by Rusticiano a Tuscan, from the dictation of Marco an Orientalized Venetian, in French, a language foreign to both.

But the character of the language as French is not its only peculiarity. There is in the style, apart from grammar or vocabulary, a rude angularity, a rough dramatism like that of oral narrative; there is a want of proportion in the style of different parts, now over curt, now diffuse and wordy, with at times even a hammering reiteration; a constant recurrence of pet colloquial phrases (in which, however, other literary works of the age partake); a frequent change in the spelling of the same proper names, even when recurring within a few lines, as if caught by ear only; a literal following to and fro of the hesitations of the narrator; a more general use of the third person in speaking of the Traveller, but an occasional lapse into the first. All these characteristics are strikingly indicative of the unrevised product of dictation, and many of them would necessarily disappear either in translation or in a revised copy.

Of changes in representing the same proper name, take as an example that of the Kaan of Persia whom Polo calls Quiacatu (Kaikhátú), but also Acatu, Catu, and the like.

As a specimen of tautology and hammering reiteration the following can scarcely be surpassed. The Traveller is speaking of the Chughi, i.e. the Indian Jogis:—

Such talk as this does not survive the solvent of translation; and we may be certain that we have here the nearest approach to the Traveller’s reminiscences as they were taken down from his lips in the prison of Genoa.

53. Another circumstance, heretofore I believe unnoticed, is in itself enough to demonstrate the Geographic Text to be the source of all other versions of the Work. It is this.

In reviewing the various classes or types of texts of Polo’s Book, which we shall hereafter attempt to discriminate, there are certain proper names which we find in the different texts to take very different forms, each class adhering in the main to one particular form.

Indeed down to nearly 1385, boys in the English grammar-schools were taught to construe their Latin lessons into French.[10] St. Francis of Assisi is said by some of his biographers to have had his original name changed to Francesco because of his early mastery of that language as a qualification for commerce. French had been the prevalent tongue of the Crusaders, and was that of the numerous Frank Courts which they established in the East, including Jerusalem and the states of the Syrian coast, Cyprus, Constantinople during the reign of the Courtenays, and the principalities of the Morea. The Catalan soldier and chronicler Ramon de Muntaner tells us that it was commonly said of the Morean chivalry that they spoke as good French as at Paris.[11] Quasi-French at least was still spoken half a century later by the numerous Christians settled at Aleppo, as John Marignolli testifies;[12] and if we may trust Sir John Maundevile the Soldan of Egypt himself and four of his chief Lords “spak Frensche righte wel!”[13] Gházán Kaan, the accomplished Mongol Sovereign of Persia, to whom our Traveller conveyed a bride from Cambaluc, is said by the historian Rashiduddin to have known something of the Frank tongue, probably French.[14] Nay, if we may trust the author of the Romance of Richard Cœur-de-Lion, French was in his day the language of still higher spheres![15]

Nor was Polo’s case an exceptional one even among writers on the East who were not Frenchmen. Maundevile himself tells us that he put his book first “out of Latyn into Frensche,” and then out of French into English.[16] The History of the East which the Armenian Prince and Monk Hayton dictated to Nicolas Faulcon at Poictiers in 1307 was taken down in French. There are many other instances of the employment of French by foreign, and especially by Italian authors of that age. The Latin chronicle of the Benedictine Amato of Monte Cassino was translated into French early in the 13th century by another monk of the same abbey, at the particular desire of the Count of Militrée (or Malta), “Pour ce qu’il set lire et entendre fransoize et s’en delitte.”[17] Martino da Canale, a countryman and contemporary of Polo’s, during the absence of the latter in the East wrote a Chronicle of Venice in the same language, as a reason for which he alleges its general popularity.[18] The like does the most notable example of all, Brunetto Latini, Dante’s master, who wrote in French his encyclopædic and once highly popular work Li Tresor.[19] Other examples might be given, but in fact such illustration is superfluous when we consider that Rusticiano himself was a compiler of French Romances.

But why the language of the Book as we see it in the Geographic Text should be so much more rude, inaccurate, and Italianized than that of Rusticiano’s other writings, is a question to which I can suggest no reply quite satisfactory to myself. Is it possible that we have in it a literal representation of Polo’s own language in dictating the story,—a rough draft which it was intended afterwards to reduce to better form, and which was so reduced (after a fashion) in French copies of another type, regarding which we shall have to speak presently?[20] And, if this be the true answer, why should Polo have used a French jargon in which to tell his story? Is it possible that his own mother Venetian, such as he had carried to the East with him and brought back again, was so little intelligible to Rusticiano that French of some kind was the handiest medium of communication between the two? I have known an Englishman and a Hollander driven to converse in Malay; Chinese Christians of different provinces are said sometimes to take to English as the readiest means of intercommunication; and the same is said even of Irish-speaking Irishmen from remote parts of the Island.

It is worthy of remark how many notable narratives of the Middle Ages have been dictated instead of being written by their authors, and that in cases where it is impossible to ascribe this to ignorance of writing. The Armenian Hayton, though evidently a well-read man, possibly could not write in Roman characters. But Joinville is an illustrious example. And the narratives of four of the most famous Mediæval Travellers[21] seem to have been drawn from them by a kind of pressure, and committed to paper by other hands. I have elsewhere remarked this as indicating how little diffused was literary ambition or vanity; but it would perhaps be more correct to ascribe it to that intense dislike which is still seen on the shores of the Mediterranean to the use of pen and ink. On certain of those shores at least there is scarcely any inconvenience that the majority of respectable and good-natured people will not tolerate—inconvenience to their neighbours be it understood—rather than put pen to paper for the purpose of preventing it.

[1] 232 chapters in the oldest French which we quote as the Geographic Text (or G. T.), 200 in Pauthier’s Text, 183 in the Crusca Italian.

[2] The MS. has been printed by Baldelli as above, and again by Bartoli in 1863.

[3] This is somewhat peculiar. I traced a few lines of it, which with Del Riccio’s note were given in facsimile in the First Edition.

[4] The Crusca is cited from Bartoli’s edition.

French idioms are frequent, as l’uomo for the French on; quattro-vinti instead of ottanta; etc.

We have at p. 35, “Questo piano è molto cavo,” which is nonsense, but is explained by reference to the French (G. T.) “Voz di qu’il est celle plaingne mout chaue” (chaude).

The bread in Kerman is bitter, says the G. T. “por ce que l’eive hi est amer,” because the water there is bitter. The Crusca mistakes the last word and renders (p. 40) “e questi è per lo mare che vi viene.”

“Sachiés de voir qe endementiers,” know for a truth that whilst——, by some misunderstanding of the last word becomes (p. 129) “Sappiate di vero sanza mentire.”

“Mès de sel font-il monoie”—“They make money of salt,” becomes (p. 168) “ma fannole da loro,” sel being taken for a pronoun, whilst in another place sel is transferred bodily without translation.

“Chevoil,” “hair” of the old French, appears in the Tuscan (p. 20) as cavagli, “horses.”—“La Grant Provence Jereraus,” the great general province, appears (p. 68) as a province whose proper name is Ienaraus. In describing Kúblái’s expedition against Mien or Burma, Polo has a story of his calling on the Jugglers at his court to undertake the job, promising them a Captain and other help, “Cheveitain et aide.” This has fairly puzzled the Tuscan, who converts these (p. 186) into two Tartar tribes, “quegli d’Aide e quegli di Caveità.”

So also we have lievre for hare transferred without change; lait, milk, appearing as laido instead of latte; très, rendered as “three”; bue, “mud,” Italianised as buoi, “oxen,” and so forth. Finally, in various places when Polo is explaining Oriental terms we find in the Tuscan MS. “cioè a dire in Francesco.”

The blunders mentioned are intelligible enough as in a version from the French; but in the description of the Indian pearl-fishery we have a startling one not so easy to account for. The French says, “the divers gather the sea-oysters (hostrige de Mer), and in these the pearls are found.” This appears in the Tuscan in the extraordinary form that the divers catch those fishes called Herrings (Aringhe), and in those Herrings are found the Pearls!

[5] As examples of these Italianisms: “Et ont del olio de la lanpe dou sepolchro de Crist”; “L’Angel ven en vision pour mesajes de Deu à un Veschevo qe mout estoient home de sante vite”; “E certes il estoit bien beizongno”; “ne trop caut ne trop fredo”; “la crense” (credenza); “remort” for noise (rumore); “inverno”; “jorno”; “dementiqué” (dimenticato); “enferme” for sickly; “leign” (legno); “devisce” (dovizia); “ammalaide” (ammalato), etc. etc.

Professor Bianconi points out that there are also traces of Venetian dialect, as Pare for père; Mojer for wife; Zabater, cobbler; cazaor, huntsman, etc.

I have not been able to learn to what extent books in this kind of mixed language are extant. I have observed one, a romance in verse called Macaire (Altfranzösische Gedichte aus Venez. Handschriften, von Adolf Mussafia, Wien, 1864), the language of which is not unlike this jargon of Rustician’s, e.g.:—

“‘Dama,’ fait-il, ‘molto me poso merviler

De ves enfant quant le fi batecer

De un signo qe le vi sor la spal’a droiturer

Qe non ait nul se no filz d’inperer.’”—(p. 41)

[6] As examples of such Orientalisms: Bonus, “ebony,” and calamanz, “pencases,” seem to represent the Persian abnús and ḳalamdàn; the dead are mourned by les mères et les Araines, the Harems; in speaking of the land of the Ismaelites or Assassins, called Mulhete, i.e. the Arabic Muláhidah, “Heretics,” he explains this term as meaning “des Aram” (Ḥarám, “the reprobate”). Speaking of the Viceroys of Chinese Provinces, we are told that they rendered their accounts yearly to the Safators of the Great Kaan. This is certainly an Oriental word. Sir H. Rawlinson has suggested that it stands for dafátir (“registers or public books”), pl. of daftar. This seems probable, and in that case the true reading may have been dafators.

[7] Luces du Gast, one of the first of these, introduces himself thus:— “Je Luces, Chevaliers et Sires du Chastel du Gast, voisins prochain de Salebieres, comme chevaliers amoureus enprens à translater du Latin en François une partie de cette estoire, non mie pour ce que je sache gramment de François, ainz apartient plus ma langue et ma parleure à la manière de l’Engleterre que à celle de France, comme cel qui fu en Engleterre nez, mais tele est ma volentez et mon proposement, que je en langue françoise le translaterai.” (Hist. Litt. de La France, xv. 494.)

[8] Hist. Litt. de la France, xv. 500.

[9] Ibid. 508.

[10] Tyrwhitt’s Essay on Lang., etc., of Chaucer, p. xxii. (Moxon’s Ed. 1852.)

[11] Chroniques Etrangères, p. 502.

[12] “Loquuntur linguam quasi Gallicam, scilicet quasi de Cipro.” (See Cathay, p. 332.)

[13] Page 138.

[14] Hammers Ilchan, II. 148.

[15] After the capture of Acre, Richard orders 60,000 Saracen prisoners to be executed:—

“They wer brought out off the toun,

Save twenty, he heeld to raunsoun.

They wer led into the place ful evene:

Ther they herden Aungeles off Hevene:

They sayde: ‘Seynyors, tuez, tuez!

‘Spares hem nought! Behedith these!’

Kyng Rychard herde the Aungelys voys,

And thankyd God, and the Holy Croys.”

—Weber, II. 144.

Note that, from the rhyme, the Angelic French was apparently pronounced “Too-eese! Too-eese!”

[16] [Refer to the edition of Mr. George F. Warner, 1889, for the Roxburghe Club, and to my own paper in the T’oung Pao, Vol. II., No. 4, regarding the compilation published under the name of Maundeville. Also App. L. 13—H. C.]

[17] L’Ystoire de li Normand, etc., edited by M. Champollion-Figeac, Paris, 1835, p. v.

[18] “Porce que lengue Frenceise cort parmi le monde, et est la plus delitable à lire et à oir que nule autre, me sui-je entremis de translater l’ancien estoire des Veneciens de Latin en Franceis.” (Archiv. Stor. Ital. viii. 268.)

[19] “Et se aucuns demandoit por quoi cist livres est escriz en Romans, selonc le langage des François, puisque nos somes Ytaliens, je diroie que ce est por. ij. raisons: l’une, car nos somes en France; et l’autre porce que la parleure est plus delitable et plus commune à toutes gens.” (Li Livres dou Tresor, p. 3.)

[20] It is, however, not improbable that Rusticiano’s hasty and abbreviated original was extended by a scribe who knew next to nothing of French; otherwise it is hard to account for such forms as perlinage (pelerinage), peseries (espiceries), proque (see vol. ii. p. 370), oisi (G. T. p. 208), thochere (toucher), etc. (See Bianconi, 2nd Mem. pp. 30–32.)

[21] Polo, Friar Odoric, Nicolo Conti, Ibn Batuta.

X. Various Types of Text of Marco Polo’s Book.

55. In treating of the various Texts of Polo’s Book we must Four Principal Types of Text. First, that of the Geographic, or oldest French.necessarily go into some irksome detail.

Those Texts that have come down to us may be classified under Four principal Types.

I. The First Type is that of the Geographic Text of which we have already said so much. This is found nowhere complete except in the unique MS. of the Paris Library, to which it is stated to have come from the old Library of the French Kings at Blois. But the Italian Crusca, and the old Latin version (No. 3195 of the Paris Library) published with the Geographic Text, are evidently derived entirely from it, though both are considerably abridged. It is also demonstrable that neither of these copies has been translated from the other, for each has passages which the other omits, but that both have been taken, the one as a copy more or less loose, the other as a translation, from an intermediate Italian copy.[1] A special difference lies in the fact that the Latin version is divided into three Books, whilst the Crusca has no such division. I shall show in a tabular form the filiation of the texts which these facts seem to demonstrate (see Appendix G).

There are other Italian MSS. of this type, some of which show signs of having been derived independently from the French;[2] but I have not been able to examine any of them with the care needful to make specific deductions regarding them.

56. II. The next Type is that of the French MSS. on which M. Pauthier’s Text is based, and for which he claims the highest authority, Second; the remodelled French Text, followed by Pauthier.as having had the mature revision and sanction of the Traveller. There are, as far as I know, five MSS. which may be classed together under this type, three in the Great Paris Library, one at Bern, and one in the Bodleian.

The high claims made by Pauthier on behalf of this class of MSS. (on the first three of which his Text is formed) rest mainly upon the kind of certificate which two of them bear regarding the presentation of a copy by Marco Polo to Thibault de Cepoy, which we have already quoted (supra, p. 69). This certificate is held by Pauthier to imply that the original of the copies which bear it, and of those having a general correspondence with them, had the special seal of Marco’s revision and approval. To some considerable extent their character is corroborative of such a claim, but they are far from having the perfection which Pauthier attributes to them, and which leads him into many paradoxes.

It is not possible to interpret rigidly the bearing of this so-called certificate, as if no copies had previously been taken of any form of the Book; nor can we allow it to impugn the authenticity of the Geographic Text, which demonstratively represents an older original, and has been (as we have seen) the parent of all other versions, including some very old ones, Italian and Latin, which certainly owe nothing to this revision.

The first idea apparently entertained by d’Avezac and Paulin Paris was that the Geographic Text was itself the copy given to the Sieur de Cepoy, and that the differences in the copies of the class which we describe as Type II. merely resulted from the modifications which would naturally arise in the process of transcription into purer French. But closer examination showed the differences to be too great and too marked to admit of this explanation. These differences consist not only in the conversion of the rude, obscure, and half Italian language of the original into good French of the period. There is also very considerable curtailment, generally of tautology, but also extending often to circumstances of substantial interest; whilst we observe the omission of a few notably erroneous statements or expressions; and a few insertions of small importance. None of the MSS. of this class contain more than a few of the historical chapters which we have formed into Book IV.

The only addition of any magnitude is that chapter which in our translation forms chapter xxi. of Book II. It will be seen that it contains no new facts, but is only a tedious recapitulation of circumstances already stated, though scattered over several chapters. There are a few minor additions. I have not thought it worth while to collect them systematically here, but two or three examples are given in a note.[3]

There are also one or two corrections of erroneous statements in the G. T. which seem not to be accidental and to indicate some attempt at revision. Thus a notable error in the account of Aden, which seems to conceive of the Red Sea as a river, disappears in Pauthier’s MSS. A and B.[4] And we find in these MSS. one or two interesting names preserved which are not found in the older Text.[5]

But on the other hand this class of MSS. contains many erroneous readings of names, either adopting the worse of two forms occurring in the G. T. or originating blunders of its own.[6]

M. Pauthier lays great stress on the character of these MSS. as the sole authentic form of the work, from their claim to have been specially revised by Marco Polo. It is evident, however, from what has been said, that this revision can have been only a very careless and superficial one, and must have been done in great measure by deputy, being almost entirely confined to curtailment and to the improvement of the expression, and that it is by no means such as to allow an editor to dispense with a careful study of the Older Text.

57. There is another curious circumstance about the MSS. of this type, viz., that they clearly divide into two distinct recensions, The Bern MS. and two others form a sub-class of this Type.of which both have so many peculiarities and errors in common that they must necessarily have been both derived from one modification of the original text, whilst at the same time there are such differences between the two as cannot be set down to the accidents of transcription. Pauthier’s MSS. A and B (Nos. 16 and 15 of the List in App. F) form one of these subdivisions: his C (No. 17 of List), Bern (No. 56), and Oxford (No. 6), the other. Between A and B the differences are only such as seem constantly to have arisen from the whims of transcribers or their dialectic peculiarities. But between A and B on the one side, and C on the other, the differences are much greater. The readings of proper names in C are often superior, sometimes worse; but in the latter half of the work especially it contains a number of substantial passages[7] which are to be found in the G. T., but are altogether absent from the MSS. A and B; whilst in one case at least (the history of the Siege of Saianfu, vol. ii. p. 159) it diverges considerably from the G. T. as well as from A and B.[8]

I gather from the facts that the MS. C represents an older form of the work than A and B. I should judge that the latter had been derived from that older form, but intentionally modified from it. And as it is the MS. C, with its copy at Bern, that alone presents the certificate of derivation from the Book given to the Sieur de Cepoy, there can be no doubt that it is the true representative of that recension.

58. III. The next Type of Text is that found in Friar Pipino’s Latin version. It is the type of which MSS. are by far the most numerous. Third; Friar Pipino’s Latin.In it condensation and curtailment are carried a good deal further than in Type II. The work is also divided into three Books. But this division does not seem to have originated with Pipino, as we find it in the ruder and perhaps older Latin version of which we have already spoken under Type I. And we have demonstrated that this ruder Latin is a translation from an Italian copy. It is probable therefore that an Italian version similarly divided was the common source of what we call the Geographic Latin and of Pipino’s more condensed version.[9]

Pipino’s version appears to have been executed in the later years of Polo’s life.[10] But I can see no ground for the idea entertained by Baldelli-Boni and Professor Bianconi that it was executed with Polo’s cognizance and retouched by him.

59. The absence of effective publication in the Middle Ages led to a curious complicationThe Latin of Grynæus a translation at fifth hand. of translation and retranslation. Thus the Latin version published by Grynæus in the Novus Orbis (Basle, 1532) is different from Pipino’s, and yet clearly traceable to it as a base. In fact it is a retranslation into Latin from some version (Marsden thinks the printed Portuguese one) of Pipino. It introduces many minor modifications, omitting specific statements of numbers and values, generalizing the names and descriptions of specific animals, exhibiting frequent sciolism and self-sufficiency in modifying statements which the Editor disbelieved.[11] It is therefore utterly worthless as a Text, and it is curious that Andreas Müller, who in the 17th century devoted himself to the careful editing of Polo, should have made so unfortunate a choice as to reproduce this fifth-hand Translation. I may add that the French editions published in the middle of the 16th century are translations from Grynæus. Hence they complete this curious and vicious circle of translation: French—Italian—Pipino’s Latin—Portuguese?—Grynæus’s Latin—French![12]

60. IV. We now come to a Type of Text which deviates largely from any of those hitherto spoken of, and the history Fourth; Ramusio’s Italian.and true character of which are involved in a cloud of difficulty. We mean that Italian version prepared for the press by G. B. Ramusio, with most interesting, though, as we have seen, not always accurate preliminary dissertations, and published at Venice two years after his death, in the second volume of the Navigationi e Viaggi.[13]

The peculiarities of this version are very remarkable. Ramusio seems to imply that he used as one basis at least the Latin of Pipino; and many circumstances, such as the division into Books, the absence of the terminal historical chapters and of those about the Magi, and the form of many proper names, confirm this. But also many additional circumstances and anecdotes are introduced, many of the names assume a new shape, and the whole style is more copious and literary in character than in any other form of the work.

Whilst some of the changes or interpolations seem to carry us further from the truth, others contain facts of Asiatic nature or history, as well as of Polo’s own experiences, which it is extremely difficult to ascribe to any hand but the Traveller’s own. This was the view taken by Baldelli, Klaproth, and Neumann;[14] but Hugh Murray, Lazari, and Bartoli regard the changes as interpolations by another hand; and Lazari is rash enough to ascribe the whole to a rifacimento of Ramusio’s own age, asserting it to contain interpolations not merely from Polo’s own contemporary Hayton, but also from travellers of later centuries, such as Conti, Barbosa, and Pigafetta. The grounds for these last assertions have not been cited, nor can I trace them. But I admit to a certain extent indications of modern tampering with the text, especially in cases where proper names seem to have been identified and more modern forms substituted. In days, however, where an Editor’s duties were ill understood, this was natural.

61. Thus we find substituted for the Bastra (or Bascra) of the older texts the more modern and incorrect Balsora, dear to memories of the Arabian Nights;Injudicious tamperings in Ramusio. among the provinces of Persia we have Spaan (Ispahan) where older texts read Istanit; for Cormos we have Ormus; for Herminia and Laias, Armenia and Giazza; Coulam for the older Coilum; Socotera for Scotra. With these changes may be classed the chapter-headings, which are undisguisedly modern, and probably Ramusio’s own. In some other cases this editorial spirit has been over-meddlesome and has gone astray. Thus Malabar is substituted wrongly for Maabar in one place, and by a grosser error for Dalivar in another. The age of young Marco, at the time of his father’s first return to Venice, has been arbitrarily altered from 15 to 19, in order to correspond with a date which is itself erroneous. Thus also Polo is made to describe Ormus as on an Island, contrary to the old texts and to the fact; for the city of Hormuz was not transferred to the island, afterwards so famous, till some years after Polo’s return from the East. It is probably also the editor who in the notice of the oil-springs of Caucasus (i. p. 46) has substituted camel-loads for ship-loads, in ignorance that the site of those alluded to was probably Baku on the Caspian.

Other erroneous statements, such as the introduction of window-glass as one of the embellishments of the palace at Cambaluc, are probably due only to accidental misunderstanding.

62. Of circumstances certainly genuine, which are peculiar to this edition of Polo’s work, and which it is difficult to assign to any one but himself, Genuine statements peculiar to Ramusio.we may note the specification of the woods east of Yezd as composed of date trees (vol. i pp. 88–89); the unmistakable allusion to the subterranean irrigation channels of Persia (p. 123); the accurate explanation of the term Mulehet applied to the sect of Assassins (pp. 139–142); the mention of the Lake (Sirikul?) on the plateau of Pamer, of the wolves that prey on the wild sheep, and of the piles of wild rams’ horns used as landmarks in the snow (pp. 171–177). To the description of the Tibetan Yak, which is in all the texts, Ramusio’s version alone adds a fact probably not recorded again till the present century, viz., that it is the practice to cross the Yak with the common cow (p. 274). Ramusio alone notices the prevalence of goître at Yarkand, confirmed by recent travellers (i. p. 187); the vermilion seal of the Great Kaan imprinted on the paper-currency, which may be seen in our plate of a Chinese note (p. 426); the variation in Chinese dialects (ii. p. 236); the division of the hulls of junks into water-tight compartments (ii. p. 249); the introduction into China from Egypt of the art of refining sugar (ii. p. 226). Ramusio’s account of the position of the city of Sindafu (Ch’êng-tu fu) encompassed and intersected by many branches of a great river (ii. p. 40), is much more just than that in the old text, which speaks of but one river through the middle of the city. The intelligent notices of the Kaan’s charities as originated by his adoption of “idolatry” or Buddhism; of the astrological superstitions of the Chinese, and of the manners and character of the latter nation, are found in Ramusio alone. To whom but Marco himself, or one of his party, can we refer the brief but vivid picture of the delicious atmosphere and scenery of the Badakhshan plateaux (i. p. 158), and of the benefit that Messer Marco’s health derived from a visit to them? In this version alone again we have an account of the oppressions exercised by Kúblái’s Mahomedan Minister Ahmad, telling how the Cathayans rose against him and murdered him, with the addition that Messer Marco was on the spot when all this happened. Now not only is the whole story in substantial accordance with the Chinese Annals, even to the name of the chief conspirator,[15] but those annals also tell of the courageous frankness of “Polo, assessor of the Privy Council,” in opening the Kaan’s eyes to the truth.

Many more such examples might be adduced, but these will suffice. It is true that many of the passages peculiar to the Ramusian version, and indeed the whole version, show a freer utterance and more of a literary faculty than we should attribute to Polo, judging from the earlier texts. It is possible, however, that this may be almost, if not entirely, due to the fact that the version is the result of a double translation, and probably of an editorial fusion of several documents; processes in which angularities of expression would be dissolved.[16]

63. Though difficulties will certainly remain,[17] the most probable explanation of the origin of this text seems to me to be some suchHypothesis of the sources of the Ramusian Version. hypothesis as the following:—I suppose that Polo in his latter years added with his own hand supplementary notes and reminiscences, marginally or otherwise, to a copy of his book; that these, perhaps in his lifetime, more probably after his death, were digested and translated into Latin;[18] and that Ramusio, or some friend of his, in retranslating and fusing them with Pipino’s version for the Navigationi, made those minor modifications in names and other matters which we have already noticed. The mere facts of digestion from memoranda and double translation would account for a good deal of unintentional corruption.

That more than one version was employed in the composition of Ramusio’s edition we have curious proof in at least one passage of the latter. We have pointed out at p. 410 of this volume a curious example of misunderstanding of the old French Text, a passage in which the term Roi des Pelaines, or “King of Furs,” is applied to the Sable, and which in the Crusca has been converted into an imaginary Tartar phrase Leroide pelame, or as Pipino makes it Rondes (another indication that Pipino’s Version and the Crusca passed through a common medium). But Ramusio exhibits both the true reading and the perversion: “E li Tartari la chiamano Regina delle pelli” (there is the true reading), “E gli animali si chiamano Rondes” (and there the perverted one).

We may further remark that Ramusio’s version betrays indications that one of its bases either was in the Venetian dialect, or had passed through that dialect; for a good many of the names appear in Venetian forms, e.g., substituting the z for the sound of ch, j, or soft g, as in Goza, Zorzania, Zagatay, Gonza (for Giogiu), Quenzanfu, Coiganzu, Tapinzu, Zipangu, Ziamba.

64. To sum up. It is, I think, beyond reasonable dispute that we have, in what we call the Geographic Text, as nearly as may be an exact transcript of theSummary in regard to Text of Polo. Traveller’s words as originally taken down in the prison of Genoa. We have again in the MSS. of the second type an edition pruned and refined, probably under instructions from Marco Polo, but not with any critical exactness. And lastly, I believe, that we have, imbedded in the Ramusian edition, the supplementary recollections of the Traveller, noted down at a later period of his life, but perplexed by repeated translation, compilation, and editorial mishandling.

And the most important remaining problem in regard to the text of Polo’s work is the discovery of the supplemental manuscript from which Ramusio derived those passages which are found only in his edition. It is possible that it may still exist, but no trace of it in anything like completeness has yet been found; though when my task was all but done I discovered a small part of the Ramusian peculiarities in a MS. at Venice.[19]

65. Whilst upon this subject of manuscripts of our Author, I will give some particulars regarding a very curious one, containing a version in the Irish language.

This remarkable document is found in the Book of Lismore, belonging to the Duke of Devonshire. That magnificent book,Notice of a curious Irish Version of Polo. finely written on vellum of the largest size, was discovered in 1814, enclosed in a wooden box, along with a superb crozier, on opening a closed doorway in the castle of Lismore. It contained Lives of the Saints, the (Romance) History of Charlemagne, the History of the Lombards, histories and tales of Irish wars, etc., etc., and among the other matter this version of Marco Polo. A full account of the Book and its mutilations will be found in O’Curry’s Lectures on the MS. Materials of Ancient Irish History, p. 196 seqq., Dublin, 1861. The Book of Lismore was written about 1460 for Finghin MacCarthy and his wife Catharine Fitzgerald, daughter of Gerald, Eighth Earl of Desmond.

The date of the Translation of Polo is not known, but it may be supposed to have been executed about the above date, probably in the Monastery of Lismore (county of Waterford).

From the extracts that have been translated for me, it is obvious that the version was made, with an astounding freedom certainly, from Friar Francesco Pipino’s Latin.

Both beginning and end are missing. But what remains opens thus; compare it with Friar Pipino’s real prologue as we give it in the Appendix![20]