“Postreros acentos de la lira indiana”: The Discovery of the Cantares mexicanos in the Nineteenth Century

On a crisp October morning in 1895, the participants of the 11th International Congress of Americanists congregated in the ex-convent of San Agustín in Mexico City, the new site of the Biblioteca Nacional de México. José María Vigil, the library's director, welcomed the international delegates by presenting a freshly printed catalog, as well as some bibliographic treasures. These included works by early Franciscan writers in New Spain and an old leather-bound manuscript containing “antiguos cantares mexicanos.”¹

Later that day, Vigil made a formal presentation of the manuscript: “These songs appear in an ancient codex that used to sit at the university library but later went missing. . . . While organizing the Biblioteca Nacional, I had the fortune to find this codex among a pile of old books.”² Vigil then introduced Mariano Sánchez Santos, whom the congress's organizing committee had commissioned to translate 27 of the songs. Sánchez Santos read the second song of the collection, in Nahuatl and in his own Spanish translation:

It's impossible to fully convey the significance that this announcement would have during the next century. The resurfacing of Cantares mexicanos y otros opúsculos, MS 1628 bis of the Biblioteca Nacional, had momentous effects for the study of ancient Mexico. Ángel María Garibay singled it out as “the most important [document] about Mexican antiquity in existence,” while Miguel León-Portilla found in the icnocuicatl (songs of bereavement) included in the collection direct access to a Nahua philosophy.⁴ The over 90 songs included in the collection formed the cornerstone of what decades after their 1895 rediscovery came to be regarded as Nahuatl literature, not only by showcasing the most elaborate and refined form of Nahuatl expression but also by providing, under the notion of xochicuicatl (flowery song), the promise of an emic concept of Nahua poetics. Since then, the adaptation of Nahua verbal art to the paradigm of Western lyrical poetry has been closely tied to the fortunes of this manuscript.⁵ And yet, like the G. K. Chesterton philosophical allegory in which an English yachtsman sets sail for the South Seas but winds up discovering England by accident, the exhumation of the Cantares mexicanos amounted to the discovery of what had been there all along.⁶ Not only because the manuscript never left the shelves of the university and the national libraries, but also because Nahua cuicatl (song) could have been found elsewhere, embedded, for example, in Nahua historical accounts such as the Anales de Tlatelolco (ca. 1545–65) and the Anales de Cuauhtitlan (ca. 1565). Nahua song was also profusely excerpted in the most important linguistic work written in New Spain, Horacio Carochi's Arte de la lengua mexicana (1645), and, albeit thoroughly Christianized, in Bernardino de Sahagún's Psalmodia Christiana (1583). Indeed, Nahua verbal art had been hiding in plain sight for three centuries.

What kinds of prejudices, epistemic biases, and intellectual blind spots made Nahua song invisible for so long? And what kinds of transformations in the status of knowledge and its institutions made the Biblioteca Nacional manuscript—and with it Nahua verbal art—finally amenable to literary discovery? In this article, I explore the entanglement between a bibliographic quest and the configuration of interpretative devices that made Nahua song finally legible and translatable at the end of the nineteenth century.

I identify two previous forays into Nahua verbal art in the nineteenth century. The first one revolves around the publication in 1854 of José Joaquín Pesado's Las aztecas, a poetry collection that claimed to be based on Nahua songs. Pesado relied on the aid of a fascinating figure, the Nahuatl scholar and intellectual don Faustino Galicia Chimalpopoca, whose role in the recuperation of Aztec song has been largely overlooked in Mexican literary history. This first attempt to incorporate Nahua song into an emerging national literature ultimately failed. However, it disseminated bibliographic information and a number of transcriptions that proved crucial, as the Cantares manuscript went missing shortly thereafter.

The second foray revolves around the translation and publication in 1887 of a selection of Aztec songs by Daniel Garrison Brinton in his pioneering Library of Aboriginal American Literature, a work that Brinton undertook without access to the Biblioteca Nacional manuscript. This second and largely successful literary discovery was made possible by a context of disciplinary realignment that, following Mauricio Tenorio Trillo, we can describe as a “stereophonic scientific modernism,” a period of great methodological experimentation and disciplinary overlap that predated the professionalization of modern social sciences between Mexico and the United States.⁷ This modernist moment, in which the emerging field of Americanism took shape, was eclectic in its methodologies and cosmopolitan in its scale, and it involved intellectual networks spanning from Porfirian Mexico to Europe and the United States. I contend that it was the convergence of linguistic, ethnographic, bibliographic, and historical knowledge around the 1880s—and not the quest for a literatura nacional—that finally allowed the torrent of Aztec verbal art to flow once again.

The hermeneutical conditions that obtained at the cusp of scientific modernism and Porfirian erudition resulted not only in the exhumation of the Biblioteca Nacional manuscript but also in the establishment of a model for interpreting Nahua song that has remained largely unaltered until the present day. This paradigm, which I have described elsewhere as lyrical-historicist, revolves around the assimilation of Nahua cuicatl to European lyrical enunciation.⁸

In her study of Amerindian verbal arts in translation, Morgane Alida Avery refers to such adaptations as “taken words” (palavras levadas), or “that into which [Amerindian words] are turned once they are taken—that is, transcribed, translated, described, and interpreted.”⁹ By reframing translation within a wider set of textual, critical, and entextual or typographic interventions, Avery underscores the myriad interpretative strategies that can be deployed for Amerindian verbal arts, each resulting in hermeneutical decisions regarding their nature.

In the case of modern Mexico, the dominant interpretation of sixteenth-century Nahua song as lyrical poetry has tended to silence alternative views—whether shamanic (posited by John Bierhorst), ethnopoetic (offered by Munro S. Edmonson or Dennis Tedlock), or aural-material (argued by Gary Tomlinson or Gabriel Pareyón)—and erase its own tracks, presenting its hermeneutical decisions as the only conceivable ones. And yet, as Aby Warburg noted in a famous 1926 lecture on Rembrandt, “every age has the renaissance of antiquity that it deserves.”¹⁰ For Warburg, Johann Winckelmann's “classical serenity” and Friedrich Nietzsche's “frenetic demonism” tell us less about the Greco-Roman world than about the affective disposition of their age.¹¹ Similarly, the assimilation of Nahua verbal art to the most prestigious of European literary genres—and its implicit notions of individuality and exalted subjective expression—tells us less about pre-Columbian civilization than it does about the twentieth century's desideratum for a Mexican antiquity. The seeming self-evidence of this assimilation signals the model's success, but the lyrical-historicist paradigm is but one among many possibilities of interpretation.

In this essay, the chronicle of an international quest for a fabled manuscript should also be read as the history of the configuration of a successful hermeneutical device that at present seems to be reaching its end.

Texcocan Bards, Letranistas, and Other Pseudo-Aztecs

As a scholar of viceregal Mexico, I was shocked to realize that until the last decade of the nineteenth century, authentic examples of Nahua cuicatl, so widely mentioned and extolled by colonial writers, were practically unknown. Before then, all but a few lines of cuicatl in Nahuatl were known, and what passed for Aztec lyric were three or four apocryphal Spanish compositions attributed to the Texcocan king Nezahualcoyotl: the prose paraphrase popularly known as the “Lamentaciones,” the “Liras,” and “Romances” in Spanish Golden Age prosody that don Fernando de Alva Ixtlilxochitl attributed to his royal forebear, and a pseudo-Otomi poem collected by José Joaquín Granados y Gálvez in his Tardes americanas (1778).¹² When Carlos María de Bustamante searched in the mid-1820s for the compositions attributed to the Texcocan king, he was only able to find one—the “Lamentaciones.”¹³ And in the book that established the international fame of the ancient Mexican poet-king, William Prescott's History of the Conquest of Mexico (1843), the US historian relied exclusively on Ixtlilxochitl's “Liras,” which he popularized in English rhyme.¹⁴ As late as 1878, in a book that contributed to the Nezahualcoyotl fever that swept through fin de siècle Mexico, Pedro Mascaró y Sosa passionately advocated that the Texcocan king be “regarded as an elegiac poet,” but the Uruguayan bibliographer relied exclusively on the apocryphal compositions.¹⁵ Indeed, as Gordon Brotherston noted, the “Lamentaciones” of Nezahualcoyotl and Fernando de Alva Ixtlilxochitl's dabbling in Golden Age poetry were “practically the only ‘samples’ of Nahuatl lyric poetry known about for over three centuries.”¹⁶ By all accounts, the budding patriotic fervor surrounding the Texcocan poet-king failed to procure authentic Nahuatl texts.

Hence, it is hardly surprising that, in the short remarks that he devoted to the question of Indigenous poetry in Mexico, Marcelino Menéndez y Pelayo regarded such fragments as “innocent literary forger[ies].”¹⁷ The Spanish critic merely echoed the opinion of most Mexican bibliographers and historians, including, as he noted, those who knew Indigenous languages. Indeed, ever since Alexander von Humboldt searched in vain for the collections of Nahuatl songs still known to Lorenzo Boturini, most critics categorically denied the survival of any form of authentic Nahua song.¹⁸

Among these authorities, none carried more weight than Alfredo Chavero. In a chapter on Nahua song, dance, and music in the first volume of México a través de los siglos (1884)—the grand synthesis of Porfirian Americanist scholarship—Chavero subscribed to the view that the “sweet and mellifluous language of the Mexica was particularly apt for poetry,” but he also asserted that “the truth of the matter is that we have no samples of that poetry.”¹⁹ He extended this judgment even to popular Nezahualcoyotl poems: in his 1892 edition of Ixtlilxochitl's Historia Chichimeca, the locus classicus of the Texcocan king's hagiography, he asserted that “no authentic song by Nezahualcoyotl is known” and that those circulating were either modern compositions or “adulterated ancient” ones.²⁰ Chavero expresses the consensus on the matter, and yet his confidence is perplexing, considering that he himself had quoted a short Nahuatl ditty in his tragedy Quetzalcóatl (1877), drawn from a copy of the Anales de Cuauhtitlan in his possession.²¹

The paucity of authentic Nahuatl songs made the presentation of the Cantares mexicanos in 1895 all the more significant. For the first time in over three centuries, sixteenth-century Nahua cuicatl drawn from an original source were made available to an international scholarly audience. The public staging of its exhumation was one of the crowning events of the lavish 11th International Congress of Americanists, the first organized by an American republic and for which the Mexican government had sent invitations to over 3,000 scientific societies across Europe, the Americas, and even Japan.²² The recovery of Aztec song consolidated Porfirian Mexico's commitment to the study and valorization of American antiquity. This overlap of scientific advance and national consolidation was underscored by the fact that, at the Secretaría de Fomento's behest, in 1904 Antonio Peñafiel prepared a lavish facsimile edition of the manuscript—the first collotype printed in the country.²³

As the vice president of the organizing committee and one of the congress's leading voices, Vigil must have informed the other members—including Chavero—about his intentions in early 1895, for the presentation took careful planning.²⁴ It was not enough to showcase the manuscript among other library treasures: Peñafiel transcribed and printed the Fábulas de Esopo en idioma mexicano included in MS 1628 bis as a token for the delegates. And the committee paid Mariano Sánchez Santos, a renowned Tlaxcalan playwright and protégé of Ignacio Manuel Altamirano, 600 pesos to translate 27 of the Cantares mexicanos.²⁵ At the congress Sánchez Santos read two of his translations together with the original Nahuatl—to great acclaim, according to Enrique de Olavarría y Ferrari's chronicle of the congress—and both were included in the published congress proceedings two years later.²⁶

The importance of Cantares mexicanos y otros opúsculos' exhumation for the study of ancient Mexico cannot be overemphasized. And yet the staging at the Biblioteca Nacional, perhaps unwittingly, mystified the nature of the finding. Indeed, cursory accounts of the Cantares mexicanos' textual history have perpetuated the view that Vigil's presentation was unprecedented.²⁷ And yet the sociology of knowledge teaches us that discoveries rarely arise ex nihilo; they are but the last visible stage of longer processes and stem from the gradual establishment of a social context of inquiry and epistemic conditions amenable to discovery.²⁸ By rending the veil of the Porfirian presentation, we encounter a series of false starts, partial inroads, and previous unsuccessful attempts.

That Chavero used a fragment of Aztec song before ostensibly denying its modern survival already suggests that Nahuatl verbal art was there all along for those willing to find it. So, more than a problem of accessibility or linguistic intelligibility, the paucity of authentic examples of Nahua verbal art bespeaks an unwillingness to tease them out from the colonial texts where postrevolutionary Mexican scholars would later find them. It's important to note that even the Cantares manuscript had been described—and portions of it transcribed—at least twice during the nineteenth century. In 1866, however, Joaquín García Icazbalceta reported his unsuccessful search for it in the national university library.²⁹ The manuscript must have been misplaced during the consolidation of Mexican libraries that followed the desamortización, when many collections were transferred to the Biblioteca Nacional.³⁰ In light of this, the Porfirian exhumation was not a foundational moment but rather the long-overdue culmination of earlier forays into Aztec verbal art. These earlier forays, and the paper trail that they left in their wake, prepared the ground for the 1895 finding.

In all fairness, José María Vigil himself candidly framed his discovery within a cosmopolitan context of scientific exchanges, and in an important article published six years before the 11th congress, he provided the main clues as to how he obtained the Cantares manuscript, via a history of correspondence spanning from Philadelphia to London and from Mexico City to Venice. By reconstructing the history of the Cantares mexicanos' partial recuperations, we will be able to chart the social context of knowledge, the reaccommodation of disciplines, and the establishment of new epistemic conditions that paved the way for the literary discovery of Aztec verbal arts at the end of the nineteenth century.

In 1854, José Joaquín Pesado published a small poetry volume entitled Las aztecas. The book claimed to present a series of poems—eight “Cantares varios” and seven “Cantos de Netzahualcoyotl”—that were “drawn from ancient Mexican songs [cantares].” The 1886 posthumous edition of Pesado's collected works added two further sections of ballads on Aztec themes, titled “Poesías eróticas” and “Leyendas mexicanas.”³¹

While unanimously considered the earliest attempt to incorporate ancient Aztec poetry into Mexican literature, Las aztecas’ critical fortunes were more varied. Some, following José María Roa Bárcena—Pesado's disciple and biographer—commended his efforts “to add an entirely new string to the Castilian lyre,” while most others, critical of Pesado's ignorance of Indigenous languages, mocked the blatantly neoclassical diction of the poems; Alfonso Reyes, for instance, noted that Pesado's Nezahualcoyotl showed the uncanny influence of Horace.³² More nuanced was the opinion of Menéndez y Pelayo, who included some of the poems in his famous Antología de poetas hispano-americanos; for him, the artifice of what he regarded “an innocent literary jest” comparable to Prosper Mérimée's famous hoax did not detract from the book's value and originality.³³

In all fairness, unlike his dozens of other creative adaptations—or blatant plagiarisms—of Horace, Torquato Tasso, or Alphonse de Lamartine, in this particular case Pesado acknowledged that his poems were but “versions and imitations.”³⁴ Menéndez y Pelayo was right in regarding Las aztecas as an “effort to create an Indigenous poetry” and concurred with the general view that Pesado had introduced the “Indigenous genre” in Mexican poetry.³⁵

Indeed, Pesado's book is part of a general trend in the second half of the nineteenth century to “Mexicanize” literature, which meant adapting European models to Mexican themes.³⁶ This literary procedure became the trademark of the Academia de Letrán. The letranistas were the first writers in Mexico to explore the possibilities of a literary aztequismo. Examples of this genre, some of the earliest indigenist narratives published in Hispanic America, appeared right from El Año Nuevo's first issue in 1837, including José María Lacunza's “Netzula” (signed December 1832) and Eulalio Ortega's “La batalla de Otumba.”³⁷ A poetic subgenre that Daniel Wogan called “pseudo-Aztec” developed in tandem with and outlived the letranistas. This poetry not only treated Indigenous themes and characters but sought to evoke actual Aztec poetry, as can be seen in Manuel Ramírez Aparicio's “Amor azteca” (1861), José Sebastián Segura's “Cantarcillo azteca” (1872), and Roa Bárcena's own Leyendas mexicanas (1862).³⁸

The first aztequista poetry collection, however, was probably Recuerdos de Anáhuac, published in 1852 by 24-year-old Emilio Rey. A Cantabrian émigré, Rey fashioned himself as a member of the Academia de Letrán, and it's likely that Pesado's collection tried to outdo Rey's original idea, which was eclipsed by Las aztecas for posterity.³⁹ And while it's true, as Wogan points out, that Pesado's Aztecs “are not Indians but rather graceful nineteenth-century Christian gentlemen (and sometimes ladies),” he himself toyed with the idea of passing them off as actual examples of Aztec poetry.⁴⁰ Thus, in El parnaso mexicano, Pesado's 1855 anthology of Mexican poetry “from the ancient Aztecs” to the early nineteenth century, he reproduced in full the poems published in Las aztecas a year earlier, this time describing them as “ancient Mexican poems . . . translated into Castilian verse.”⁴¹

Such deceitful claims, however, had a modicum of truth: unlike later iterations of the genre, Pesado did try to base Las aztecas on authentic sixteenth-century sources. For example, while Menéndez y Pelayo heard in the “Consejos” poems included in the first part of Las aztecas more biblical echoes than anything gleaned from the “hieroglyphs of Anáhuac,” Marco Antonio Campos notes that they were in fact based on Sahagún's Historia general de las cosas de Nueva España (ca. 1578), which had been published by Bustamante in 1829.⁴²

Pesado's 1854 collection claimed in its title that the poems were based on “ancient Mexican cantares.” What were those cantares? In his biography of the poet, Roa Bárcena noted that Pesado used a prose translation of some cantares provided by don Faustino Galicia Chimalpopoca.⁴³ It is almost certain that the prose translation mentioned by Roa Bárcena was based on the Biblioteca Nacional's Cantares mexicanos manuscript, for we know that José Fernando Ramírez commissioned a copy of the work from the same individual. In the advertencia to the manuscript, signed May 1859, Ramírez wrote that he had entrusted Chimalpopoca with such a task “in order to warrant the faithfulness of the copy.”⁴⁴

This means that during the early 1850s there was at least one real effort to translate the Cantares mexicanos into Spanish. Unfortunately, this translation—the first translation of Nahua cuicatl to a European language—is not known to have survived. However, these transcriptions and translations are crucial moments in the literary discovery of the Cantares mexicanos, and they have largely escaped the attention of scholars. For example, Fernando Tola de Habich, the foremost scholar on the letranistas and the modern editor of Pesado's works, shrugged off Pesado's reliance on Chimalpopoca's translations as “an accusation too trivial to be taken seriously.”⁴⁵ This kind of heavy-handed dismissal is characteristic of the way in which, until very recently, historiography has dealt with one of the most complex and multifaceted figures of nineteenth-century Mexico. In order to understand the impact of this manuscript on the literary discovery of Nahua song, we shall now focus our attention on this largely forgotten figure.

Don Faustino Galicia Chimalpopoca was born in 1805 in San Pedro Tláhuac, south of Mexico City, to a noble Indigenous family that claimed to descend from the third Tenochca tlatoani. Lawyer, administrator of lands, and legal representative of the Indigenous parcialidades of Mexico City; professor of jurisprudence at the Colegio de San Gregorio, of Nahuatl at the University of Mexico, and of Otomi at the Seminario Conciliar del Arzobispado; scholar of colonial Nahuatl documents; and member of the Sociedad Mexicana de Geografía y Estadística (SMGE)—Chimalpopoca was the most important Nahua intellectual of nineteenth-century Mexico.⁴⁶ And yet his reputation has been tarnished, since the second half of the nineteenth century, by a double disgrace: the first due to his intimate collaboration with the Second Empire (1863–67), the second due to his reputation as an unreliable translator.⁴⁷

While Chimalpopoca's translations were generally viewed positively during the last third of the nineteenth century, subsequent opinion was framed by the critical remarks of José Fernando Ramírez and Primo Feliciano Velázquez—a rival translator—that prefaced later editions of the Anales de Cuauhtitlan.⁴⁸ (Despite his reservations, Ramírez continued to commission Chimalpopoca to transcribe and translate Nahua documents for over a decade.) His twentieth-century character assassination was perpetrated primarily by Garibay, whose unremitting disdain for this Indigenous scholar was supported by neither argument nor evidence.⁴⁹ Chimalpopoca's textual scholarship and translation practices remain to be studied, but it is clear that his unique poetics of translation, accomplished almost without precedent and before the development of modern philological criteria, did not sit well with nineteenth-century positivism.⁵⁰ And yet during his lifetime he remained the undisputed cultural mediator between Indigenous Mexico and its urban elites and between Indigenous communities and the state. It was as part of these mediating roles that Chimalpopoca undertook the study, transcription, and translation of Nahuatl documents.

Chimalpopoca attended the Colegio de San Gregorio, an institution founded by the Jesuits in 1586 dedicated to educating Indigenous people. San Gregorio's extensive library included hundreds of printed works and manuscripts in Indigenous languages. It was amid this library that Chimalpopoca began his study of ancient Nahuatl documents and his fruitful collaboration with José Fernando Ramírez.⁵¹ José Joaquín Pesado turned to Chimalpopoca, well established by the 1850s as a Nahuatl expert, in his quest for Nahuatl sources. It's likely that Chimalpopoca would have pointed Pesado in the direction of the Cantares mexicanos, a work that he would transcribe for Ramírez a few years later.⁵²

Further evidence of Chimalpopoca's lost translation of the Cantares was furnished by Ignacio Montes de Oca y Obregón in 1886. In the prologue to the third edition of Pesado's works, Montes de Oca y Obregón confirmed Pesado's use of a prose version of the poems but added an intriguing clue about the collaboration: “The distinguished and noble D. Faustino Chimalpopoca, who was the person who presented Pesado with the literal prose translation of most of these poems, used to tell his students of Mexican language that the verse of our poet did not resemble in the least either the Aztec original or his own translation.”⁵³

This indirect testimony not only confirms that Chimalpopoca translated the Cantares but also gives us an inkling of how he viewed Pesado's aztequista poetry: that Las aztecas failed to grasp the poetics of Aztec song as well as the merit of his own prose rendition. It's likely that Pesado would have found the songs' dense diction simply meaningless or unpalatable, and indeed he made little use of Chimalpopoca's versions and did not even credit him.

For this reason it is all the more surprising that Mexican literary history effected what can only be described as a critical inversion. Las aztecas has been persistently commended as a pioneer or foundational work whose shortcomings are blamed on Chimalpopoca. And while Pesado's trite sentimentality has been hard to appreciate even for his most ardent vindicators, the figure of Chimalpopoca has proven, to this day, all too receptive to the incessant hurling of blame. Marco Antonio Campos, for example, extolled Pesado's merit “as the first poet of independent Mexico who had true inklings of the flower and song of the Nahua poet and who, thanks to his talent and ability to versify, turned deficient translations into good poetry.”⁵⁴ More recently, Christopher Domínguez Michael has voiced a similar view, describing Chimalpopoca's “deficient versions” and his work in general as “very uncouth,” that of an “improvised nahuatlato.”⁵⁵ This isn't simply a question of bad faith. The critical scapegoating of a beleaguered Nahua intellectual seems to have stifled the most crucial question: How can one explain the midcentury failure to grasp one of the richest corpuses of Amerindian verbal art? This failure is not devoid of irony, as it was precisely the letranistas who embraced the project of building a national literature by writing about Mexican themes. This episode expresses the limits of integrating pre-Hispanic literature—and, for that matter, colonial literature—into a literatura nacional.

Two landmark contributions to Mexican literary historiography show to what extent the inclusion of pre-Hispanic literatures remained fraught throughout the nineteenth century. In an influential 1866 lecture on Mexican poetry, Joaquín Baranda vindicated the literary expression of Anáhuac. Following Boturini, he extolled the poetic qualities of Nahuatl; following Francisco Javier Clavijero, he praised the “meter and cadence” of Aztec “lyrical poetry.” However, Baranda was less interested in literary historiography than in the problem of attaining “literary independence,” which by definition could only accompany political independence. For this reason, not even the literature of New Spain could be “properly considered Mexican poetry”—much less the songs and hymns of Anáhuac.⁵⁶

A decade later, José María Vigil devoted several pieces to the history of Mexican literature. In two important articles from 1876 and 1878, he voiced the common concern with literary independence that had so occupied his generation. With Baranda and Altamirano, he held that Mexican literature could only begin after independence.⁵⁷ However, Vigil endowed pre-Hispanic and colonial literatures with a new, propaedeutic function. For him, knowledge of the literary past would act as a corrective or, to use his own term, as an instance of “regeneration”: by means of grasping “the historical precedents of the races that inhabit our territory . . . [we] could undertake the lofty endeavor of our regeneration.”⁵⁸ This propaedeutic view informs Vigil's Reseña histórica de la literatura mexicana, written by the end of 1891 but not published until 1909.⁵⁹ In this sweeping but unfinished history, Vigil devoted a chapter to pre-Hispanic literature. In a similar vein to Baranda, his “quick glance” at the “literature of the civilization that preceded us” is meant to be but a preamble to a truly national literature.⁶⁰

Returning to the midcentury, why did the nationally inclined letranistas fail to grasp Nahua cuicatl? To be sure, the linguistic challenges of the notoriously difficult Cantares are partially responsible. But we can venture further hypotheses. In one of the few pieces to tackle the question directly, Montserrat Galí Boadella asserted that Indigenous texts simply failed to appeal to the Romantic sensibility that swept through independent Mexico.⁶¹ The opacity of Indigenous texts would have precluded the “affective projection” and identification inherent to Romanticism, particularly through the feelings of love and piety.⁶² As Galí Boadella noted, the paucity of indigenist literature contrasts with the surprising abundance of medieval revivalist literature in Mexico, particularly in the 1840s.⁶³ In this regard, it is telling that “Netzula,” the first indigenist work printed in Mexico, owed more—including its bears and lions—to François René de Chateaubriand's Atala (1801) than to anything remotely reminiscent of pre-Columbian Mexico. The letranistas, with their belated Romantic imagination and narrow aesthetics, could only resort to previously adapted material, which allowed only the most shallow and formulaic engagement with the Indigenous past. Faced with the exuberant, almost Pindaric diction of Nahua song, the letranistas recoiled into the complacency of a vaguely pseudo-Aztec poetry.

While Chimalpopoca's prose translation has not surfaced, it is worth noting his unique approach to sixteenth-century Nahuatl texts. In stark contrast to postrevolutionary nationalist historiography, which set out to use sixteenth-century Nahuatl texts as sources on pre-Hispanic Mexico, Chimalpopoca saw these texts as unambiguous expressions of the vernacular Nahua Catholicism that had flourished for three centuries in New Spain. As a devout guadalupano, Chimalpopoca lavished much care and attention to the study of Nahua Christian texts, particularly liturgical drama and works of Guadalupan devotion.⁶⁴ The Cantares mexicanos, with its profuse Christian and Marian imagery, would have been particularly appealing to this devout Nahua scholar. A century later, Mariano Cuevas identified one of the songs as the earliest testimony of the Tepeyac apparition.⁶⁵ As we shall see, the Christian reading of the Cantares mexicanos was largely rejected during the second discovery of Nahua song at the end of the nineteenth century, which led to a concerted effort to “pre-Columbianize” sixteenth-century Nahua cuicatl in postrevolutionary Mexico.⁶⁶ The pre-Columbian interpretation gained traction with Ángel María Garibay's translations in the 1930s and 1940s and was consolidated in the work of Miguel León-Portilla; it is largely accepted today.⁶⁷ Chimalpopoca would have been appalled by such a reading, which downplayed the songs' Christian, and particularly Marian, motifs. Echoes of the Christian reading continued to be heard in Sánchez Santos's 1895 translation and in the translations of another devout Indigenous nahuatlato, Mariano Jacobo Rojas.

In the end, Chimalpopoca's prose translation of the Cantares mexicanos went unnoticed. However, his transcription of the song collection would have lasting, if unintended, consequences. The dispersal of his papers not only called attention to the subject but also triggered an international desire for Aztec verbal art at a time when, beyond the borders of Mexico and its parochial literary academies, scholars began an intense exploration of non-European literatures.

In July 1880, part of Ramírez's fabled library went up for public auction at Puttick & Simpson in London.⁶⁸ Among many treasures, Ramírez's Bibliotheca Mexicana included Chimalpopoca's transcription of the Cantares, the original of which, as we have seen, was no longer accessible by this date.⁶⁹ The vendue brought together Americanists and bibliophiles from across the globe, so the news about the resurfacing of a faithful copy of the Cantares—described in detail in the catalog—did not go unnoticed. Writing from the Bureau of American Ethnology, the punctilious James Constantine Pilling noted that Chimalpopoca's transcription was, “in all probability, the only complete copy existing of these remnants of ancient Mexican poesy.”⁷⁰ Indeed, the vendue catalog instigated an international search for the Cantares that in the last quarter of the nineteenth century would lead directly to the second discovery of Aztec verbal art.

Selective Blind Spots, Cosmopolitan Injunctions

If the news about the existence of a copy of the Cantares mexicanos—appearing in a sales catalog of which only 50 copies were printed—had any consequence, this is because by the 1880s there existed an international network, spread between Europe, the United States, and Mexico, dedicated to the study of Amerindian languages. Unlike the parochial literary pursuits of fin de siècle Mexico, the linguistic research that flourished in tandem within this international network was from the start a cosmopolitan venture. Since its inception, for instance, the SMGE had an international scope, from the composition of its membership to its relationship with learned societies across the globe, including the Smithsonian and the American Philosophical Society in the United States.⁷¹ The linguistic works published by members of the SMGE, such as Manuel Orozco y Berra's Geografía de las lenguas y carta etnográfica de México (1864) and Francisco Pimentel's Cuadro descriptivo y comparativo de las lenguas indígenas de México (1862–65), were received with the greatest acclaim in the United States, France, and Germany. This international network constituted the echo chamber in which the surfacing of Chimalpopoca's transcription of the Cantares resonated. The cosmopolitan intersection of Amerindian linguistics and the study of the literature of non-Western peoples finally created the epistemic conditions for the discovery of Aztec verbal art.

The news of a surviving copy of the Cantares resonated most strongly among those who took an interest in the study of Amerindian literary expression. This emerging subfield of Americanist research was most clearly associated with the figure of Daniel Garrison Brinton, the first chair of ethnology in North America. After his early studies of the US Southwest and inspired by Humboldt's essay on the Amerindian verb—which he translated—Brinton took an interest in Amerindian linguistics. Brinton's investigations around the 1880s focused on the literary appreciation of the songs and stories compiled by ethnographers throughout the continent. This led to the pioneering Library of Aboriginal American Literature (1882–90), a collection of Amerindian works that Brinton edited whose purpose was to make available pieces of Amerindian expression that were unpublished, in private hands, or printed “in the ponderous transactions of learned societies.”⁷²

A sense of Brinton's views can be found in an important piece titled Aboriginal American Authors. First delivered in August 1883 at the Fifth International Congress of Americanists in Copenhagen and then expanded for publication in Philadelphia, this is Brinton's most explicitly programmatic piece.⁷³ He begins it by asserting categorically, and against common prejudice, that the “literary faculty in the native mind” across the continent is “indicated by a vivid imagination, a love of narration, and an ample, appropriate, and logically developed vocabulary.” He proposes a sweeping overview of “the literary efforts of the aborigines of this continent,” which, he notes, is “a chapter in the general History of Literature hitherto wholly neglected.”⁷⁴ Brinton's Library of Aboriginal American Literature sought to redress this by publishing works bound to become classics of Amerindian literature, such as The Annals of the Cakchiquels (in 1885) and The Güegüence (in 1883).

In 1887, Brinton published for the very first time a selection of the Cantares mexicanos in translation, years before the actual Biblioteca Nacional manuscript resurfaced. The publication was of immense significance, and among its direct effects was the exhumation of the original manuscript by José María Vigil. We shall now reconstruct how this exhumation unfolded in the wake of Brinton's endeavors.

Brinton's Copenhagen lecture is a programmatic text, not only because it provides an état de la question but also because Brinton used it, I believe, to cast a net for new bibliographic findings.⁷⁵ The lecture's discussion of Mexican materials, for instance, mentioned the long-lost collection of Nahua songs by Ixtlilxochitl, which Brinton had “reason to suspect” was “probably in the recesses of some private library.” Brinton did not elaborate on his reason for suspecting this, but it's clear that his overview of historical documents was meant to draw the attention of the potential collector “unaware of the nature of his possessions.”⁷⁶

At this point, a crucial difference between the Copenhagen lecture and the Philadelphia essay emerges. In the essay, Brinton mentioned the Cantares for the very first time: “Under the title Cantares de los Mexicanos [sic], there was long preserved in the library of the University of Mexico a manuscript of the sixteenth or seventeenth century, with a large number of supposed ancient Aztec songs; but what has become of it now, nobody knows.” For this statement Brinton drew from the standard source—García Icazbalceta's Apuntes para un catálogo de escritores en lenguas indígenas de América of 1866—but added an intriguing remark: “There are, however, two copies of it extant, somewhere.”⁷⁷ It's very likely that Brinton was referring here to the partial Chimalpopoca transcription found today at the Biblioteca Nacional de México.⁷⁸

Brinton must have learned about the lost Biblioteca Nacional manuscript between August 1883, when he delivered the Copenhagen lecture, and late October 1883, when he expanded and published the piece. I suspect that Brinton might have learned about the manuscript during his stay in Europe, and from the man who had last reported on it. A detail makes this likely: the sole Mexican delegate to the Americanist congress in Copenhagen was none other than García Icazbalceta, director of the Mexican Academy of Language and, as we have seen, the person who reported on the missing manuscript in 1866.⁷⁹ If this hypothesis is correct, Brinton would have learned from the Mexican bibliographer about Chimalpopoca's transcription—the whereabouts of which, after the London sale of 1880, seem to have been lost—and about the existence of a second copy. Around the time that the Philadelphia essay was published, Brinton must have tapped into his extensive network of correspondents in Europe and Mexico in search of this second copy.

An unexpected turn of events led to Brinton finding it shortly thereafter, most likely in early 1884. That year, he acquired a cache of documents from Alphonse Louis Pinart's library sale in Paris, most of which came from the fabled manuscript collection of the abbé Brasseur de Bourbourg. Among the documents that Brinton acquired was the original “mémorial de Tecpan-Atitlán.”⁸⁰ It must have come as a surprise to find, among documents pertaining to Guatemala and to Mayan languages, a copy of 27 Nahuatl songs that the abbé himself had made from the original manuscript in February 1865. This incomplete and faulty transcription furnished the basis for Brinton's publication of the songs and sits today in the Daniel Garrison Brinton Collection at the University of Pennsylvania.⁸¹

Brinton's book including the songs, Ancient Nahuatl Poetry, Containing the Nahuatl Text of XXVII Ancient Mexican Poems (1887), made available for the first time a selection of Nahua song in the original and in English. The text, volume 7 of the Library of Aboriginal American Literature, also included notes, a vocabulary, and the first modern treatise on Nahua poetics since Lord Kingsborough briefly discussed the topic in his Antiquities.⁸² Later scholars, such as Eduard Seler and Garibay, took issue with Brinton's translations. He was aware of the limitations of his versions, deeming them “the best rendering that I could give them at present,” but he was convinced that the importance of “these monuments of ancient native literature” outweighed all reservations. At the same time, he understood the trailblazing nature of his enterprise and claimed to have tackled the translation of the notoriously difficult Cantares without any assistance: “The text of the ancient songs which it contains offers extreme and peculiar difficulties to the translator, and I have been obliged to pursue the task without assistance of any kind. Not a line of them has ever before been rendered into an European tongue [sic], and my endeavors to obtain aid from some of the Nahuatl scholars of Mexico have, for various reasons, proved ineffectual.”⁸³

Brinton studied Nahuatl via Agustín de la Rosa's manuals and cultivated an extensive network of correspondents in Mexico that included Chavero, Nicolás de León, and Pimentel. So why did his efforts to get help from contemporary Mexican scholars and nahuatlatos prove ineffectual? To be sure, the difficulty of the Cantares must have played a role.⁸⁴ However, a clearer answer can be found via an important article in which José María Vigil recounts the immediate events leading to his finding of the manuscript. By following this Ariadne's thread laid out by Vigil as it unraveled in Porfirian Mexico, it's possible to reconstruct the events that led directly to the manuscript's rediscovery but also, more importantly, to illuminate the kinds of selective blind spots and epistemic gridlocks that rendered Aztec verbal art largely invisible to Mexican intellectuals before Brinton.

José María Vigil's account suggests that the original manuscript was discovered in response to an inquiry by a little-known figure: the Venetian philologist and poet Marco Antonio Canini. Among the ambitious literary projects undertaken by this larger-than-life figure was the preparation of an extraordinary anthology of love poetry from across the world, entitled Il libro dell'amore. In its final edition, published in Venice from 1885 to 1890, Canini's anthology comprised five volumes and boasted translations of erotic and love poetry from over 160 languages and dialects.

In the third volume of Il libro dell'amore, Canini attempted to include examples of Amerindian love lyric from across the continent. For Mexico and Mesoamerica, he was interested in publishing poems in Nahuatl and Yucatec Maya. The quest for the erotic poetry of Indigenous Mexico, which Canini described at length in a section of the third volume's introduction (signed December 1887), opens an unexplored vantage point on the winding paths of Americanism, philology, and literary culture in Porfirian Mexico.⁸⁵ Although related and coeval, these paths would rarely cross. And one of those rare occasions on which they did brought about the second discovery of Nahuatl verbal art and the recovery of MS 1628 bis.

Canini's quest for Mesoamerican lyric led him to two natural sources. For Yucatán he perused the works of Brasseur de Bourbourg, and for central Mexico he resorted to the work of Brinton. However, the fact that he used Brinton's published Aboriginal American Authors as a guide and had no knowledge about Ancient Nahuatl Poetry suggests that the bulk of his Americanist research must have taken place between 1884 and early 1887. In other words, he learned about the putatively lost Cantares in Brinton's writing but did not get to see the 27 songs in translation.

His bibliographic research soon reached a dead end, so he did what he often did when that happened: he wrote to local authorities on the matter. As he related in the introduction to Il libro dell'amore's third volume, “I have written to several Mexicans in order to acquire popular poetry in Nahuatl, Maya, and other languages spoken in that vast country, about which has written with erudition Mr. Pimentel.”⁸⁶ In addition to Pimentel, Canini reported inquiring with Altamirano, Francisco Sosa, and Vigil about ancient or modern love poetry in Indigenous languages. Right from the start, however, Canini acknowledged that despite his best efforts the search failed to produce a single piece: “At present, I have not been able to acquire authentic popular songs in Aztec or Nahuatl, neither ancient nor modern.”⁸⁷

Why, may we ask, were Canini's efforts such a resounding failure? After all, he wrote to some of the foremost Mexican intellectuals of the age. Pimentel was the most accomplished linguist of Porfirian Mexico, the author of the grand Cuadro descriptivo y comparativo de las lenguas indígenas de México, which had been expanded and reissued in 1874. In a report on the situation of Indigenous peoples that Pimentel wrote for Maximilian in 1864, waxing poetic about the lost grandeur of Anáhuac, he had referred to the songs of Nezahualcoyotl as the “last notes of the Indian lyre.”⁸⁸ And as Canini himself noted, Altamirano was not only an “illustrious poet” but, just like President Benito Juárez, a “pure Indian,” which in Canini's view made him “an even more competent” authority on the matter.⁸⁹

What Canini did not realize was that he had unknowingly stepped into the greatest blind spot of Mexican intellectual culture, one shared by liberals and conservatives, intellectuals of European, mestizo, and even Indigenous stock: a contempt for Indigenous Mexico, or what Vigil, a rara avis in this regard, denounced as a “disdain for the vanquished races.”⁹⁰ This contempt was shared not only across political lines but also by those who had made a career of studying Mexican antiquities and Indigenous languages.

Unbeknownst to him, in his search for Nahuatl poetry the Venetian had resorted to two of the staunchest anti-Indigenous intellectuals in Mexico. Altamirano was undoubtedly the harshest of these intellectuals; his vitriol against Indigenous languages and his support of exhaustive Hispanization is well documented. Altamirano, whose parents were Indigenous people from Guerrero and who had learned Spanish as a schoolboy, would write things like this apropos of the failed Hispanization of Indigenous peoples during viceregal times: “What would have been lost? Only a bramble of tongues and dialects from which today's archaeology and philology barely draw but a paltry benefit.”⁹¹ Sure enough, Altamirano answered Canini's inquiries with characteristic contempt: “He replied that among Indigenous peoples in Mexico there exists today no poetry, neither popular nor cultured, and that from the ancient one there are but a few insignificant fragments, quoting one that he claims is the only one that he knows about.”⁹²

Francisco Sosa, a liberal historian, replied along the same lines. Sosa, however, did not miss the opportunity to promote his own poetry and sent Canini a volume of his verse. The Venetian scholar reciprocated by including one of his sonnets in Il libro dell'amore. Canini's perfunctory praise of Sosa as a “noteworthy poet and critic” might have led the Mexican critic, as one of the editors of the Revista Nacional de Letras y Ciencias, to publish Canini's piece, but the overall picture that the Venetian scholar draws of the Porfirian intelligentsia is not particularly flattering.⁹³

The fruits of Canini's endeavor could not have been more disappointing. In the end, he was only able to include two pseudo-Aztec poems in the anthology—one taken from an English-language periodical, the other from Pesado's Las aztecas—well aware that they were inauthentic.⁹⁴

But this is more than a tale of prejudice and ignorance compounded by opportunism. Canini's dire portrayal of leading Mexican intellectuals' attitudes toward Indigenous expression underscores the epistemic conditions that kept Nahua song and other forms of verbal art hidden in plain sight during the better part of the century. Nowhere is this more evident than in the case of the person who, at least in principle, was in the ideal position to publicize Nahua song: Francisco Pimentel. Canini was familiar with the work of this great erudite Porfirian, and he claimed to have written to him. Pimentel was not only the foremost authority on Mexico's Indigenous languages but also its foremost literary critic and historian, as can be gathered from his monumental Historia crítica de la poesía en México (1892). The preface to the second edition of the Cuadro descriptivo had a breathtaking summary list of contributions to Amerindian linguistics, including grammatical descriptions of many languages and the establishment of new linguistic families. This is why reading Pimentel's opinion in the Historia crítica on the existence of Nahua literature is, to say the least, disconcerting.⁹⁵

Contrary to Brinton—whose work he knew well and with whom he corresponded—Pimentel denied the status of literature to any form of Indigenous expression. In his view, “Mexican literature in a strict sense . . . is that which follows European [poetic] art and is in Castilian.” As for the survival of ancient Aztec song, he shared Chavero's negative view, even though he conceded the existence of a “mixed Indo-Hispanic” literature.⁹⁶ Pimentel embodies the epistemic limits and the selective blindness that ensue from absolute fidelity to a set of narrow prescriptive aesthetic ideals. In his poetics, steeped in an eclectic philosophy inspired by Georg Wilhelm Friedrich Hegel and Hippolyte Taine, there were only two valid universal poetic principles: classicism, or the primacy of form, and Romanticism, or the primacy of subjective expression.⁹⁷ Under such constraints, Amerindian verbal arts were simply ungraspable. Like Chavero, Pimentel displayed not only a selective blindness but an actual disavowal: he lived long enough to see Brinton's translation of the Aztec cantares without revising his views on the matter.

In one regard, however, Canini's Americanist investigation was consequential. As we have seen, the Venetian perused Brinton's Philadelphia essay, with its musings about the whereabouts of the Biblioteca Nacional manuscript. And so he turned to the fourth Mexican intellectual mentioned in his introduction to volume 3 of Il libro dell'amore, José María Vigil: “Brinton asks what happened to a precious manuscript that used to be housed at the Biblioteca de México, under the title ‘Cantares de los Mexicanos,’ and I raise the same question to Mr. Vigil, ardent patriot and distinguished Mexican writer, author of an excellent translation of Persius and currently in charge of the Biblioteca de México.”⁹⁸ More a polite injunction than a true inquiry, Canini's missive to Vigil would, at last, yield palpable results.

Vigil's own account of the events finally leading to the manuscript's finding appears in an article published in 1889. In the same volume of the Revista Nacional de Letras y Ciencias as Canini's piece, the director of the Biblioteca Nacional conveyed the good news to the Venetian scholar: “It is with satisfaction that I reply to him that this precious manuscript exists in the Biblioteca Nacional.” He goes on to describe what led to the discovery:

The text makes clear that Vigil had stumbled on the manuscript in the process of reorganizing the Biblioteca Nacional, which began around 1881 at its new site in the ex-convent of San Agustín. What the vignette does not explain is why Vigil's frustration when contemplating the unintelligible pages—the identity of which must have been perfectly clear to him, thanks to the manuscript's Spanish title page—did not cause him to reach out immediately to those “learned nahuatlatos” who could have come to his aid.

As was all too frequent in nineteenth-century Mexico, Vigil was probably spurred to action only when his patriotism was piqued by the lavish publication of an outsider, in this case the “eminent Americanist” Daniel Garrison Brinton: “The reader will understand, in view of all this, my delight when a book, beautifully printed in Philadelphia (1887), and with the title Ancient Nahuatl Poetry . . . landed in my hands.” Vigil's article included two Spanish translations of Brinton's as well as a presentation of many of Brinton's ideas, and it must be read as an invitation to Mexican scholars such as Francisco del Paso y Troncoso, recently appointed to a new Nahuatl professorship at the Escuela Preparatoria, to jump-start the study of precious manuscripts like the Cantares mexicanos that had been collecting dust on the shelves of libraries and archives.¹⁰⁰

León-Portilla noted that Vigil's exhumation of the original manuscript had scant tangible results.¹⁰¹ Indeed, besides the important paleographic transcription and facsimile published by Peñafiel in 1899 and 1904, respectively, the Porfirian rediscovery of the Cantares yielded nothing comparable to even Brinton's partial and imperfect volume. Sánchez Santos's 1895 translations were probably found lacking by Peñafiel and Vigil, who had commissioned them, and with the exception of the two poems printed in the 11th International Congress of Americanists' proceedings, they remain unpublished to date. Besides these, only one other song translated directly from Nahuatl made it into print by the turn of the century.¹⁰² It would take four decades before a new generation of nahuatlatos undertook again the translation of the most important collection of ancient Amerindian song, and much longer before serious inroads were made into the study of Nahua poetics.

And yet the Porfirian recuperation had, to my mind, lasting if intangible consequences, as it established the hermeneutical coordinates under which the final discovery of Nahua song would follow in the twentieth century. While no significant translation or study was published by Porfirian scholars, their articles and paratexts, particularly those of Peñafiel, were successful in disseminating a series of crucial ideas that, originating in Brinton's introductory study to Ancient Nahuatl Poetry, would shape the predominant understanding of Aztec song to this day.

The interpretation of Nahua song that consolidated in the second discovery was twofold. On the one hand, Brinton extended full literary naturalization to Aztec song. Against the prejudices of his age, which believed that “a taste for poetry is a mark of high culture,” Brinton strove to show that Indigenous peoples throughout the Americas are “passionate lovers of verse and measure, of music and song.”¹⁰³ And against those who, like Pimentel in Mexico, held that true poetry could only be written in European languages and prosody, Brinton evoked the opinion of the first European commentator on Amerindian song, Michel de Montaigne, who, after having heard a French translation of Tupi songs from Brazil, wrote, “I have had enough to do with poetry . . . to say about this that not only is there nothing barbarous in this fancy, but that it is altogether worthy of Anacreon.”¹⁰⁴ In his works, Brinton adopted Johann Georg Hamann's dictum, which had inspired Johann Gottfried von Herder, Humboldt, and the young Johann Wolfgang von Goethe—“poetry is the common mother-tongue of the human race”—and applied it to actual content culled from ethnographic literature and historical documents.¹⁰⁵ Whether it was the great literary “monuments” of American antiquity, the Inuit folk songs recorded by Heinrich Rink, or a modern Nahuatl song recorded by Hermann Berendt in Tamaulipas, Brinton held these expressions to form an important chapter of “the general history of the growth of the poetic faculty.”¹⁰⁶ This approach could not strike a stronger contrast with the attitudes of Porfirian intellectuals reported by Canini.

In the preface to Ancient Nahuatl Poetry, Brinton sketched a veritable program of study and exegesis for what would be known thereafter as “Nahuatl literature”: “In conclusion, I would mention that there is a large body of Nahuatl literature yet unpublished, both prose and poetry, modern and ancient, and as the Nahuatl tongue is one of the most highly developed on the American continent, it is greatly to be desired that all this material should be at the command of students.”¹⁰⁷ Brinton would go on to publish some fragments of this Nahuatl corpus, such as the ritual songs recorded in the Florentine Codex, but the program that he envisioned would only be realized half a century later. When in the 1930s Mexican scholars began once again to recuperate Aztec song, they did so under the aegis of “literature” and “poetry.”¹⁰⁸ In many ways, postrevolutionary Mexican scholars sanctioned Brinton's original program, which amounted to adapting Nahua verbal art to the model of European lyrical poetry.

But whereas Brinton's notion of poetry, grounded in German idealism, conceived of it as a human faculty, the twentieth-century Mexican school appealed to a neoclassical system of genres to which Nahua verbal arts were recursively assimilated and thus vindicated as literature. And if nineteenth-century critics had at least tried to inflect neoclassical genres in order to reflect non-European content, whether referring to a “poesía méjico-gentílica” (as did Mascaró y Sosa) or a “literatura indo-hispana” (as did Pimentel), in the hands of Garibay and León-Portilla—both classically trained scholars—Nahua xochicuicatl would be reframed as lyrical poetry tout court.

The second idea that consolidated in fin de siècle Mexico was the notion that the songs had a pre-Columbian origin—a thesis accepted nowadays by scholars of the Cantares mexicanos. This thesis framed the official presentation of the manuscript in 1895, when Vigil, paraphrasing Brinton, explained that even if “Christian ideas” appeared in the songs, it was easy to show that these “had been interpolated by the friars to adapt them to the new religious beliefs.”¹⁰⁹ Despite initial doubts, this notion quickly took hold, particularly after Peñafiel emphatically endorsed it in the paratexts of his editions.¹¹⁰ Peñafiel rebuked those who considered the Cantares as mendicant compositions meant for the conversion of Indigenous peoples because “nothing of the sort ever existed in Spain”; he claimed that the Cantares represented “the historical-mythological tradition as taught in the religious-military schools.”¹¹¹

Peñafiel overlooked not only the fact that sixteenth-century Franciscans did write and publish Nahua cuicatl for religious use but also that half the songs in the Cantares mexicanos mention postconquest dates and events. The thesis of a pre-Columbian origin for the songs was formulated by Brinton on the basis of his partial and indirect access to the original manuscript. In what is likely an early description by Brinton of Brasseur de Bourbourg's partial copy, he noted that the “hymns contained in it date back for the most part to a time anterior to the conquest.”¹¹²

When a few years later Brinton printed his translation of the Cantares, this was the general view that he would adopt. In the introduction, he refuted those who denied the survival of Aztec song by emphasizing the antiquity of the songs in his collection, many of which he “consider[ed] to have been composed previous to the Conquest,” even if he was also aware that “the songs are evidently from different sources and of different epochs.” Brinton expressed some reservations, noting that the definitive dating of the songs should proceed “from a careful scrutiny of the internal evidence.” But, perhaps unwittingly, he succeeded in formulating an imagined scenario that, adopted and expanded by Vigil and Peñafiel, would have lasting consequences: “In applying these tests, it should be remembered that a song may be almost wholly ancient, that is, composed anterior to the Conquest, and yet display a few later allusions introduced by the person who preserved it in writing, so as to remove from it the flavor of heathenism.”¹¹³

Without using the term, Brinton formulated the basis for what we can call the theory of interpolation, which, together with the lyrical model, has defined the Mexican school's orthodox reading of the Cantares ever since. More than a consistent critical or philological hypothesis, the theory of interpolation has operated as a kind of foundational myth, a “historical deduction” (historische Ableitung) predicated on the modern desideratum for a pristine Mexican antiquity.¹¹⁴

Conclusion

The Cantares mexicanos has led a paradoxical existence in modern Mexico. On the one hand, it has been the cornerstone of numerous reconstructions of ancient Mesoamerica; on the other, it has had a tendency to be constantly lost or misplaced, to fall out of sight, or to recede into the horizon. And every attempted incursion into it exhibits an uncannily similar pattern, including the staging of yet another rediscovery followed by endless, Kafkaesque postponements.

This paradoxical status of the Cantares as a deferred foundation is most clearly seen in the numerous failed attempts at producing a national edition, which would grant Aztec poetry full citizenship in the canon of Mexican literature. As we have seen, this was the ultimate goal of Vigil and Peñafiel. And it would be the goal of Luis Castillo Ledón in 1917 and Rubén M. Campos in 1936, both of whom enlisted the aid of another Indigenous nahuatlato and intellectual, Mariano Jacobo Rojas. And once again, both proved in the end unsuccessful.¹¹⁵

In the 1930s and 1940s, Garibay began to publish his translations of Aztec song in literary journals such as Ábside; decades later, his translations appeared in a series of anthologies and the monumental Historia de la literatura nahuatl (1953–54). But even Garibay did not succeed in publishing the whole Cantares mexicanos, even if taken together his dispersed translations amount to two-thirds of the collection.

We would have to wait until the 1990s, spurred by John Bierhorst's controversial English translation, for the project of a national edition to be envisioned once again.¹¹⁶ This time around, the projected edition was spearheaded by the Universidad Nacional Autónoma de México (UNAM) and a host of other national and international institutions. An interdisciplinary Seminario Cantares mexicanos, under the academic direction of León-Portilla, was formalized in 1992 and, a few years later, produced a new facsimile of the volume.

In 1995, a forthcoming edition of the entire MS 1628 bis was announced, but in the end, the ambitious project would take nearly two more decades to complete.¹¹⁷ In 2011, the first full edition and translation of the songs was finally published in Mexico by UNAM, 152 years after Chimalpopoca's transcription and 116 years after Vigil announced the exhumation of the original manuscript.¹¹⁸ In many ways, the UNAM edition completed the Porfirian recovery of Aztec verbal art and lent authority to the poetic-lyrical model.

Have we finally discovered the Cantares mexicanos? Or do we merely have the Cantares mexicanos we deserve?

I would like to thank the readers who at various stages offered me insightful comments: Adela Pineda Franco, Pedro Ángel Palou, Alejandro Quintero Mächler, and the anonymous reviewers for HAHR. Special thanks go to Deborah Brown Stewart at the Penn Museum Library for helping me locate Charles-Étienne Brasseur de Bourbourg's transcription of the Cantares mexicanos in the Daniel Garrison Brinton Collection and resolve one of many bibliographic riddles that this story is rife with; to Alberto Partida Gómez at the Biblioteca Nacional de México for his help locating and digitizing Mariano Sánchez Santos's manuscript; to José Luis Nogales, who aided me with its transcription; and to Fredy Aguilar Reyes, who at the final stage of this project granted me access to the Biblioteca Andrés Henestrosa in Oaxaca. I would also like to express my gratitude to Ruth Moorman and Sheldon Simon for generously supporting my research at Boston University. Finally, this article would not have been written without the support of Nicole T. Hughes.

Notes

References