The Early History of the Boxer Codex

Abstract

In 1950 this journal published Charles Boxer's description of the Codex that bears his name and is now in the Lilly Library in Bloomington, Indiana. In this article I shed light on a number of questions about the volume that he left unanswered. The plan of the essay is as follows. First, I recapitulate the nature, composition and contents of the Codex, and its associations with Chinese artists. Next, I turn to evidence that the explorer Quíros saw the manuscript in the court in Madrid no later than 1610; in particular I compare the text of his Summary relation with part of the Boxer Codex. Finally, an examination of the binding shows the Codex was only bound in Madrid, strongly suggesting how and when it got to Spain.

Charles Boxer acquired the Boxer Codex at an auction in 1947 and described it in an article in this Journal in 1950. ¹ The volume comprises a little over 300 leaves of rice paper to which Boxer added four folios of other manuscript: an account signed by Luis Pérez Dasmariñas principally comprising extracts from letters. The greater part of the work, which I shall refer to as the ‘Culture items’ and on which most interest has focused hitherto (but which I shall not concentrate on here) comprises descriptions of the culture and inhabitants of the South East and East regions of Asia, together with drawings and descriptions of Chinese deities and fabulous animals. ² The remainder, which I shall refer to as the ‘Exploration items’, comprises works from three sources. Leaves 101r–139r contain two texts, both apparently coming via Bishop João Ribeiro de Gaio of Malacca, concerning Aceh and Malay states. ³ Leaves 139r–149v contain an account of a voyage made by Roxo de Brito, a Portuguese, in 1581–1582 south and west of New Guinea, which will be a major focus. ⁴ Leaves 213r–239v have a narrative due to Fr Martín de Rada, OSA, on his visit to China in 1575. ⁵

The latest date that can be inferred internally from the Codex is 1590–1591. ⁶ Boxer believed that either Gómez Pérez Dasmariñas or his son, Luis Peréz Dasmariñas, who were successive governors of the Philippines at the end of the sixteenth century, was the original commissioner of the Codex. ⁷ The new governor of the Philippines and his son arrived in Manila in 1590, only 19 years after the Spanish settlement of Manila. From the advent of the Spanish until 1617, Gómez Pérez Dasmariñas was praised as the best, perhaps the only good, governor of the Philippines. ⁸ He was very trusting of the Chinese:

When he left Manila . . . he engaged a galley manned by Chinese, good rowers, which they had given him, and he paid those from the Parian (the Parian is like the alcaycería where the Chinese had all kinds of shops and offices), and those he had unshackled, and retaining their weapons, more as soldiers than oarsmen, very much trusting them. ⁹

This was his undoing, for he was murdered by the Chinese crew on 25 October 1593. ¹⁰ It is tempting to suggest that Gómez Pérez Dasmariñas's acquaintance with and trust in the Chinese corresponded with his commissioning of the Chinese artist to provide the illustrations in the Culture items of the Codex. There is, however, additional supporting though circumstantial evidence. Regalado Trota Jose wrote: ¹¹ [The] achievements [of the Chinese] in ivory carving were so developed by 1590 that the first bishop of Manila, Domingo de Salazar – a Dominican who arrived together with Sedeño and the first Jesuits – enthusiastically wrote the King:

In the Parian are found all the workers with all the skills and mechanical arts needed for a Republic, and in such great quantity. . . They have so perfected themselves in this art, that they have wrought marvelous works both with the chisel and with the brush. Having seen some ivory images of the Child Jesus it seems to me that nothing more exquisite than these could be produced; and such is the opinion of those who have seen them. The churches are now being provided with these images, which they sorely lacked before; with the Sangleys’ ability to replicate those images from Spain, it should not be long when even those made in Flanders will not be missed. ¹²

It would therefore appear that both father and son were familiar with the work of the Chinese artists during the lifetime of the former. ¹³

After he succeeded his father as governor in 1593, or at the latest in 1594, Luis Peréz Dasmariñas commissioned a Chinese artist to make a very large sculpture of the Virgin Mary as Nuestra Señora del Rosario, with ivory face, hands and child. The statue stands 139 cm high and is now known as La Naval. The person who oversaw the carving was Hernando de los Ríos Coronel. ¹⁴ On the fatal 1593 expedition the latter was sailing with Luis Pérez Dasmariñas and asked permission to go ahead of Gómez Pérez Dasmariñas since an opposing wind was slowing down the galley in which the governor was travelling. The son may well have felt guilt over the death of his father, and that may have been a major factor in the commissioning of the statue. It seems unlikely that the younger Dasmariñas, who was only 25 when he became governor, ¹⁵ would undertake such a large enterprise as that of the statue, and so soon after the death of his father, if he had not seen previous work by Chinese artists. I therefore am inclined to the view that the elder Dasmariñas commissioned the Codex, the pages subsequently coming to his son. ¹⁶

Now I turn to the Exploration items. On leaves 139r–149v of the Codex there is an account of a voyage by Miguel Roxo de Brito in Maluku. Boxer and Pierre-Yves Manguin noted that the journey was mentioned in the Summary relation written by Pedro Fernández de Quirós (probably in 1610, see below) and printed by Zaragoza in 1880. ¹⁷ (Quirós is well known as a pilot and for his attempt to discover the great south land he called Austrialia [sic].) ¹⁸ Quirós wrote that this Summary relation was made from “information supplied to Quirós at the Court of Spain by Hernando de los Ríos”. ¹⁹ I suggest that Quirós saw either the Codex itself or an accurate copy of the account of Roxo de Brito's voyage in the hands of Hernando de los Ríos Coronel who met Quirós in Spain sometime between 1607 and 1610.

Hernando de los Ríos Coronel had arrived in the Philippines from Spain in 1588. ²⁰ He is much less well known than father and son Dasmariñas, but, like Quíros, he had trained as a pilot and, besides serving under Gómez Pérez Dasmariñas and accompanying him on his last expedition, he accompanied Luis Peréz Dasmariñas on two later expeditions in 1596 and 1598, being second-in-command (almirante) on the latter. ²¹ His role as a pilot would have made him privy to secret matters about navigation and maps. Throughout his life he maintained these interests and wrote a number of rutters or sea-logs, in particular of his voyages to Spain and back in 1605–1606 and 1610–1611. ²² As Boxer says: “It is probable that only a senior government official or a high ranking ecclesiastic would have had access to such confidential documents as Loarca's report of 1580, and [Bishop] Ribeiro Gaio's of 1589, to say nothing of de Brito's account of his New Guinea voyage.” ²³ De los Ríos also knew of the work of another author of a text in the Codex, Fr Martín de Rada. ²⁴ These various factors suggest that it is not surprising that de los Ríos should have been in the small circle of people who knew about the voyage of Roxo de Brito.

Fr Celsus Kelly, OFM, dates the Summary relation of Quirós as “c. Oct.” of 1610, but this cannot be the date when he met de los Ríos, because the latter left Cadiz on 29 June 1610 to return to the Philippines, although it could be the date when Quirós wrote the summary. As I shall show, however, Quirós seems to have seen the Spanish text of de Brito's voyage that is in the Codex. ²⁵ Certainly de los Ríos and Quirós were both in Spain between 1607 and 1610 because de los Ríos had been sent to Spain as Procurator General – the sole advocate at the Spanish court for the inhabitants of the Philippines – and he was there from 1606 to 1610. ²⁶

In his role as Procurator General de los Ríos put to the king many requests on behalf of the Philippines, while Quirós “dedicated himself from [1607] onwards to sending multiple petitions [indeed more than 50] to the king, Philip III, seeking sponsorship for a further voyage to the South Pacific”. ²⁷ It was not until 1608 that Quirós printed his Eighth Memorial, ²⁸ which may have been the first time that de los Ríos became aware of the voyage of Quirós – though that seems unlikely – but which may explain why it may have been as late as 1610 that they discussed the voyage of Roxo de Brito. ²⁹ Given that Quirós wrote so many memorials, Kelly's date of 1610, which is based on the dating of many writings of Quirós and others, seems quite plausible. ³⁰

There is remarkable similarity between the texts of Roxo de Brito and Quirós. Below is the full text of the report by Quirós as recorded by Zaragoza in the left column with corresponding excerpts from the Boxer Codex version on the right. ³¹

Relacion sumaria que saqué de la que me dió en esta corte el licenciado hernando de los rios, procurador de filipinas.

The similarity of the content is obvious; more than 80 per cent of the Quirós text has the same semantics, though the syntax is often somewhat different. ⁴⁷ More dramatic is the sequencing of the items and their location in the Codex version. With the exception of the mention of weapons and a very short one of pearls, the order of topics is exactly the same as in the text of Roxo de Brito in the Boxer Codex. Further, the Summary relation only treats of material on folios 142r–147v: the central 12 of the 22 pages of the original and, in Chapter XXXIX of his famous account of his search for Australia, Quirós said that he would include “a chapter of [Roxo de Brito's] relation at the end of this discourse”, though he did not in fact include the ‘chapter’ there but put it separately as the Summary relation. ⁴⁸ It would be hard to keep in mind all that is in Roxo de Brito's account, and since de los Ríos had many other matters of concern in Madrid, in particular presenting his list of many items requiring attention in the Philippines, it is unlikely that de los Ríos simply told Quirós the details. ⁴⁹ These factors strongly suggest that de los Ríos had, at the very least, a version of Roxo de Brito's text with him in Madrid, and also that he had (or had had) access to the original Codex. ⁵⁰

Now I would like to turn to another aspect of the Codex, concerning its binding. Boxer felt that the Codex could have been taken by the British when they invaded the Philippines in 1762, but he also says “it might have been sent to Spain at any date after 1590”. ⁵¹ The physical condition of the Codex is remarkably good, in contrast with that of the more than 400 books of the same vintage in the library of the University of Santo Tomas in the Philippines, many of which arrived in the Philippines before 1620. ⁵² Boxer also said that he could not tell whether the book had been rebound. ⁵³ The leather used to bind books is highly subject to deterioration in the tropical climate and I have seen nothing in the Philippines from that period to match the elaborately tooled quality of the Boxer Codex binding. Further, the Codex has a sheep binding. Since there were no sheep in the Philippines, it is most unlikely to have been made there.

As Boxer noted, on most pages three sides of a rectangle have a blue ruling surrounding the text. Curiously the rectangle is often not aligned between recto and verso. However, one can determine that the margins have remained unchanged, showing that the pages have not been recut. The binding is therefore the original. Boxer also noted that the binding “is of a familiar late sixteenth/early seventeenth century Iberian type”. ⁵⁴ The filler under the pastedown inside the front cover could not have come from the Philippines because it is set in movable type, and the books printed in Manila at that time were all xylographed. In fact, the cover is from a sheet of paper that contains parts of pages 226 and 231 of a book by Pablo de Mera published in 1614. ⁵⁵ Similarly, the filler inside the back cover is of an even-numbered page from the same book. ⁵⁶ Since the pages under the front pastedown are not separated, this indicates it was from a sheet that was not bound, and therefore surely came from the printer. This also means that the Codex was bound in the printing house, la Compañía de Impresores y Libreros del Reino [the Royal Company of Printers and Booksellers], that produced Pablo de Mera's book. All of the above seem to provide conclusive evidence that the book was bound in Spain, in 1614 at the earliest.

The pages too are in remarkably good condition and much better than those of virtually all the books from that period that I have seen in the Philippines. The latter tend to have many wormholes, but there are only a few holes in the Boxer Codex, and these are in the leaves at the end of the book. Moreover, they do not penetrate the binding. This again suggests that the Codex left the Philippines early; I therefore next turn to the question of when it arrived in Spain.

As I noted above, the elder Dasmariñas was murdered by his Chinese crew in 1593, and his son died in the Sangley uprising in 1603, ⁵⁷ both deaths taking place while de los Ríos was still in the Philippines. De los Ríos was a pilot, indeed the only pilot I know by name in the Philippines at that time, ⁵⁸ and he had been close to both father and son. Consequently, he was in the perfect position to have access to the Codex. In 1605 he was sent to Spain by the citizens of Manila as Procurator General. The timing suggests that the Codex was taken by de los Ríos to Spain, two years after the death of the son. Moreover, since it seems clear that the book was bound by the Royal Company of Printers and Booksellers, it appears that the book remained in the court. This also suggests that de los Ríos did not own the book but seemingly was the person who took it to Spain with the express aim of showing, or giving, it to the king. The book's presence in the court would certainly explain why Quirós, who was in the court, was able to make such an accurate and order-preserving précis of part of the voyage of Roxo de Brito. I therefore conclude that de los Ríos took the unbound pages to Spain when he went there in 1605 and that they were subsequently bound in Spain no earlier than 1614, four years after de los Ríos returned to the Philippines. ⁵⁹ There remains the question of the subsequent history of the Codex, in particular why the manuscript left the court, but this appears to be unknown until it appears in the library of Lord Ilchester in Holland House.