The Translations of Nebrija
FIGURE 4. The Roman Temple of Diana at Mérida, Spain. Photo by Byron Hamann.
FIGURE 5. Woodcut portrait of Antonio de Nebrija, from the title page to a 1585 edition of the Diccionario.
FIGURE 6. Genealogies of vocabularies: the translations of Nebrija.
FIGURE 7. Contemporary uses of Mesoamerican chia seeds.
FIGURE 8. The puma, mascot of the Universidad Nacional Autónoma de Mexico, Mexico City.
FIGURE 9. Front cover of Richard Percyvall’s 1591 Bibliothecae Hispanicae.
Printed in New Spain in the middle of the sixteenth century, Alonso de Molina’s Vocabulario en la lengua Castellana y Mexicana is a relic from the meeting of three worlds. Its pages bring together prehispanic Mesoamerica, late medieval and early modern Europe, and the antique Mediterranean. The lexical entries of the Vocabulario thus join Franciscan friars, Aztec priests, and pagan Romans. To understand the _Vocabulario_—and indeed to understand all indigenous language vocabularies created in sixteenth-century New Spain—we have to consider a complex knot of different times and places.
Some 1500 years before Columbus arrived in the Americas, the armies of ancient Rome were conquering and colonizing much of what is now Europe, North Africa, and the Middle East. Roman colonists left their mark on this landscape in many ways. They built forums and temples and aqueducts. They erected statues to their gods and emperors. They buried their dead in the ground. And they broke and lost portable objects brought from Rome: ceramics, glassware, jewelry, coins. As a result, long after the fragmentation of the Roman Empire, material reminders of Rome’s ancient presence could be encountered in the cities, and soil, of former provinces—such as in Iberia (Figure 4). Already during the Middle Ages, people throughout Europe had been intrigued by the omnipresent ruins of Roman antiquity.16 This interest was transformed, and intensified, from the fourteenth century onwards: the period of revival and recreation that became known as the Renaissance.17
The study of language played an important role in this Renaissance culture. The Catholic Church had continued to use Latin, of course, but as humanist scholars began to focus on ancient Latin texts and inscriptions, they realized that the Latin used by the Church was very different from the Latin used by the ancient Romans. A number of writers wanted to recover and revive the Latin language as it had been spoken in the time of the Caesars—the Latin of the great orator Cicero, for example.18 In Iberia, the most important scholar of ancient Latin was the humanist Antonio de Nebrija (1441-1522; Figure 5).19
Nebrija attended the University of Bologna in the 1560s as a young student. During his stay, he became inspired by the new approaches to ancient Latin then being developed by Italian scholars. He brought these ideas back with him to Iberia when he returned in 1570. A decade later, in 1481, Nebrija published an influential book for learning Latin grammar—a grammar that quickly went through a number of editions. Nebrija followed this with other works of scholarship on ancient Latin. At the same time, he became interested in the Castilian language as well. Castilian, as a Romance language, had its origins in the Latin spoken by the ancient Roman colonizers who had settled in Iberia. In 1492 Nebrija published a Latin-to-Castilian Lexicon (word list or vocabulary), as well as a book of Castilian grammar. This was the first grammar of a vernacular language to be printed in Europe, and it was modeled on the grammar of Latin that Nebrija had published a decade before. In 1495, Nebrija “reversed” his 1492 Lexicon_—that is, he published a Castilian-to-Latin word list, which he titled a _Vocabulario.20
In the early modern world, “translation” had two meanings, both involving movement. One meaning is quite familiar to us today: the translation, or movement, from one language to another. But translation could also indicate physical movement as well, such as in the translation of holy relics from one shrine to another. Both of these senses of translation can be applied to the influence of Antonio de Nebrija’s Castilian-Latin dictionary throughout the early modern world. This text was translated—moved—throughout Europe and across the Atlantic. As it traveled, it was used as a model for translating both European and Native American languages to Castilian.
Nebrija’s Castilian-to-Latin dictionary was republished over two dozen times between 1495 and 1600 (usually in bi-directional volumes that combined both Castilian-to-Latin and Latin-to-Castilian formats).21 Copies of this work traveled to the New World, and were used as models for the creation of Castilian-to-indigenous language dictionaries.22 Alonso de Molina even makes reference to Nebrija’s Vocabulario in the introductory pages of his Vocabulario en la lengua Castellana y Mexicana: “…as Antonio de Nebrija puts it in his Vocabulario…”23
Significantly, the two dozen editions of Nebrija’s Castilian-to-Latin vocabulary or dictionary (in the 1500s the work was often titled a Dictionarium) are not all alike.24 The core Vocabulario published by Nebrija in 1495 went through constant revisions throughout the course of the sixteenth century. Some of these revisions were authored by and printed by Antonio de Nebrija and his heirs (his sons Sancho and Sebastian and his grandson Antonio).25 Other revised editions were pirated copies. Some of these pirated editions mimicked “official” publications by Nebrija and his heirs almost exactly, but others added new Castilian-Latin entries. By studying all of these different editions together, we can construct a genealogy chart tracing how these two- dozen-plus editions of Nebrija’s Castilian-to-Latin dictionary are related to one another, and how their lists of entries expanded over time (Figure 6). Overall, the original 1495 dictionary went through five major periods of transformation: in the Salamanca edition of 1513, in the Granada edition of 1536, in the Antwerp edition of 1545, the Granada edition of 1585, and the Antequera edition of 1595.
But in addition to these five major periods of change (most involved expansions in thenumber of entries included), every individual edition of Nebrija’s Castilian-Latin dictionary contained minor variations in spelling (and occasionally in word order). Sixteenth-century books were printed by hand, and every letter in them had to be set individually. Not surprisingly, this painstaking process meant that errors—or simply variations—emerged from one printing to another. These minor variations make each edition unique. Because of these variations, we can identify with certainty which specific edition of Nebrija’s Castilian-Latin dictionary was used by Molina as a model for his 1555 Castilian-Nahuatl dictionary: the Granada edition of 1545. A copy of this edition is available online at the Biblioteca Virtual Andalucía.26 Identifying this source volume is extremely important. Knowing precisely which edition of Antonio de Nebrija’s Castilian-Latin Dictionarium was used as a model for Molina’s Castilian-Nahuatl Vocabulario allows us to understand a great deal about the Nahuatl terms preserved in Molina’s text—and, by extension, about the vision of Central Mexican life that the Vocabulario’s entries preserve.
First, many of the Castilian-language entries in Molina’s Vocabulario, entries for which Nahuatl translations are provided, were originally terms gathered in order to translate ancient Latin categories. (Indeed, Nebrija’s Latin-Castilian vocabulary cited the specific ancient authors who had used particular Latin terms, including Pliny, Seneca, Virgil, Galen, and Plautus). 
Thus, for example, Molina’s Vocabulario includes entries for “Senador romano” (Roman senator) and “Nariz romano” (Roman nose)—entries for which Nahuatl “translations” are provided. Molina’s Vocabulario also copies entries from Nebrija that refer to specific features of the late medieval and early modern Mediterranean world, such as “mezquita” (mosque). 28 The Nahuatl translations offered for these terms, then, are neologisms, invented on the spot.
Two Nahuatl phrases are given for “mosque”: Mahoma tlatlatlauhtilizcalli (meaning “Muhammed’s house of oratory”) and mahoma calli (“house of Muhammed”).29 Such complex (and slightly awkard) Nahuatl neologisms tell us much about the colonial situation in which Molina’s Vocabulario was created—but they do not reflect Nahua categories with prehispanic histories.
However, even though Molina used Nebrija’s 1545 Castilian-Latin Dictionarium as a model, he realized that some Castilian terms had no equivalent in Nahuatl—and he also realized that some Nahuatl words had no equivalent in Castilian. Molina used Nebrija’s Dictionarium as a flexible model—not as a procrustean bed. In his introduction, Molina points out that new words and things were being encountered by both Europeans and Nahuas in sixteenth-century New Spain: “we have many things that they did not previously know or attain: and for these things they did not have, and do not have, proper words; and, on the other hand, there are things which they had that we lack in our language…” 
Molina used two strategies to deal with all of these innovations. Sometimes, words were defined in an admittedly awkward way, with “long circumlocutions and approximations.” 
In other cases—especially when translating Nahuatl words to Castilian—Molina invented new Castilian words, based on already-existing terms. One of these neologisms is “Abaxador,” used to translate the Nahuatl tlatemouiani.32 “Abaxador”, Molina explains—in his apology for making up new Castilian words to translate Nahualt ones—means “he who takes something down.” The Franciscan friar created this word from the already-existing verb abaxar (to take down—today’s abajar) combined with the common use of the -dor ending in Castilian terms for people who do things (portador, a person who carries; gobernador, a governor).33
Molina’s entries do not simply differ from Nebrija’s by including neologisms. Molina’s entries also include words from Native American languages that had already been adopted by Castilian-speakers living in the New World. In 1555, Europeans had been coming to the Americas for over sixty years—and had lived in Mesoamerica for almost forty years. A number of terms from indigenous languages—Arawakan, Nahuatl—had already been incorporated into the Castilian lexicon by 1555. As a result, some of the “Castilian” entries in Molina’s _Vocabulario_—entries not found in Nebrija—are actually Castilianized Native American terms.
The Arawakan term for maize, for example—incorporated into Castilian as maiz or mayz_—appears on the “Castilian” side of a number of entries, as in “Mayz seco en maçorcas. centli. cintli” (Dried maize on the cob. centli, cintli). Another Castilianized Native American word was _cacao (chocolate, or cocoa), probably borrowed from the Nahuatl cacahuatl: “Beuida de cacao solo. atlanelollo cacaua atl.” (Beverage with only cacao. atlanelollo cacaua atl.)  And another was chía, from the Nahuatl chiyan—a plant from the mint family (_Salvia hispanica) whose seeds were toasted and used as a flavoring as well as pressed to release their varnish-like oil: “Azeyte de chia. chiamatl” (Chia oil. chiamatl; Figure 7).35
In other cases Molina—and his indigenous collaborators—defined Mesoamerican things according to European terms that were almost, but not quite, the same. The term cauallero, knight, is defined with the Nahuatl term pilli_—which means noble, a category different from that of prehispanic warrior (“knight”) per se. The word for lion (“Leon animal”) is translated as _miztli, the Nahuatl term for the American puma: a related but distinct feline species (Figure 8).36 The word for Hell (infierno), a Catholic place of post-mortem torment, is defined by the far more neutral mictlan_—a Nahua realm of the dead. This is another challenge when using the _Vocabulario today: to understand when apparently straightforward European-Mesoamerican equivalents are not, in fact, as precise as they first appear.
The complicated issues involved in translating sixteenth-century Nahuatl into sixteenth-century Castilian are, perhaps, fairly obvious. As a basic rule of thumb, one should always check to see if a Castilian entry in Molina is also present in Nebrija’s 1545 Dictionarium. If the same entry is present in both texts, then you should ask why Nebrija’s entry was used by Molina and given a Nahuatl translation. Does the category copied from Nebrija refer to something that was obviously part of the Mesoamerican world (such as terms for water, agua, or rain, lluuia)? Does the category copied from Nebrija approximate—but with important differences—an aspect of Mesoamerican life (and, if so, what are the differences separating, say, a Leon animal from a miztli)? Or is the copied category a linguistic fossil—as with the translation for “mezquita”—that tells us much about European interests but very little about Mesoamerican ways of life? Alternatively, when a Castilian category used by Molina is not found in Nebrija, chances are that Molina added that category in order to capture some aspect of life in the New World that had no precedent in Europe.
Comparing Molina’s 1555 Vocabulario with the 1545 Granada edition of Nebirja’s Dictionarium, then, allows us to ask a whole new set of questions about how the Nahuatl terms in Molina’s text were generated, and more broadly about the connections linking sixteenth- century Mesoamerica, sixteenth-century Europe, and the ancient Mediterreanan past. In the past, scholars have sometimes used the categories in Nebrija-inspired indigenous-language dictionaries at face value, and thus have interpreted prehispanic Mesoamerican practices according to categories actually referring to ancient Rome! But understanding Molina’s Vocabulario also involves another, less obvious issue surrounding translation: the translation of words across time.
Words can change their meaning over the course of centuries, even decades. The meaning of a word used in Spanish today may be quite different from the meaning of that “same” word in sixteenth century-Castilian. When consulting Molina’s Vocabulario, then, we can’t just think about complicated questions of translation from Castilian to Nahuatl (and to Latin) in the sixteenth century. When using Molina’s Vocabulario to study Nahua language and culture, we also have to think about how Castilian itself has changed over the past five centuries. For example, when Molina justifies in his introduction why he focused on Nahuatl as it was spoken in Texcoco and Mexico, he claims this is because it is in these towns “donde mejor y mas curiosamente se habla la lengua.” Today, the word curiosamente means curioulsly or strangely.
But in the sixteenth century, curiosamente meant “carefully”—not strangely. Thus Molina’s statement is best translated as “where the language is best and most carefully spoken.”37 Fortunately, a number of early modern dictionaries exist that can help us understand the meanings of Castilian words five hundred years ago.38 And this brings us back to the theme of “The Translations of Nebrija.” Many of these early modern dictionaries—such as the two Spanish-English dictionaries published in London in the 1590s—were themselves modeled on editions of Nebirja’s Spanish-Latin Dicitionarium. As a result, many of the Nebrija-borrowed Castilian entries translated into Nahuatl in Molina’s 1555 Vocabulario are translated into sixteenth-century English in Richard Percyvall’s 1591 Bibliothecae Hispanicae (Figure 9) and in John Minsheuï¿½s 1599 A Dictionarie in Spanish and English.39 Thus the archaic Castilian term Abahar (to give off steam), a term (as discussed above) included in Nebrija’s 1545 Dictionarium and Molina’s 1555 Vocabulario, is translated in Percyvall’s 1591 Bibliothecae Hispanicae as “to smoke or to fume.” A few pages later, Percyvall defines the word baho (root of abahar) as “vapor.” Nebrija’s Castilian-Latin Dictionarium also served as a model for creating late sixteenth- and early-seventeenth century Castilian-Tuscan (Italian) and Castilian-French dictionaries as well.40 These also provide important records of the meanings of early modern Castilian words.
In conclusion, when using Molina’s Vocabulario, you should always think about thecomplex process involved its creation, and about the many different cultures—from Iberia to Mesoamerica to ancient Rome—that its pages link together. And you should also think about the travels and translations of Nebrija within Europe. Those translations of Nebrija created Molina’s siblings, and their contents (Castilian-English, Castilian-French) can help us interpret Molina’s often archaic Castilian terms—and, in turn, the complex translation of those terms into Nahuatl.
Text by Byron Hamann
16 Camille 1989.
17 Sixteenth-century artist and historian Giorgio Vasari (1511-1574) coined the term rinascita_ (“rebirth”) to speak of the art of his era; however, the terms Renaissance (in English) and _renacimiento (in Spanish) did not emerge until the ninteenth century, used to indicate a re-naissance, rebirth, of classical knowledge and art. On the nineteenth-century origins of the term renaissance in English, see the Oxford English Dictionary: www.oed.com.proxy.uchicago.edu/viewdictionaryentry/Entry/162352
18 Burke 2004, 57-58.
19 For a brief biography of Nebrija and his interest in recovering Latin as it existed in the Classical past, see Guzmán Betancourt 1997; on his stay in Italy, see Caro Bellido and Tomassetti Guerra 1997, 78-87.
20 On Nebrija’s various innovations in the creation and structuring of dictionaries, see Lope Blanch 1997; on the relationship between Nebrija’s 1492 and 1495 dictionaries, see Guerrero Ramos 1995.
21 For catalogs of the various editions of Nebrija’s works, see Esparza Torres and Niederhe 2001; Wilkinson 2010, 30-38; and Abad 2001, 97-112 (for publications up to 1520).
22 Leonard 1992, 222, 268; López Bernasocchi and Galeote 2010, 22-23.
23 “comolo pone Antonio de librixa en su vocabulario” (Molina 1555, v).
24 To my knowledge, the only previous investigations into the internal variations of Nebrija’s Castilian-Latin vocabulary as it developed over time are by María Lourdes García-Macho (1992, 1993, 1995, 2005, 2008), who compares and contrasts the 1495 and 1516 editions.
25 Gallego Morell 1970; Cátedra García 1996; Leiva Soto 2000.
26 Detailed arguments about the sixteenth-century translations of Nebrija, and the connection of specific editions of Nebrija to Spanish dictionaries for both European and New World langauges, see Hamann n.d. Prior arguments about the influence of Nebrija on subsequent dictionaries in both the New World (Galeote 2003, 140, 142; Galeote 2004, 544; Hernández 2000; Hernández 2005, 1791; Hernández 2006, 68; Hernández 2008, 198; Karttunen 1995; Laughlin 1988, 11-17; Lope Blanch 1999, 555-558; Romero Rangel 2006, 141-142; Smith-Stark 2002, 534; Smith-Stark 2009) and the Old (Guerrero Ramos 1992; Nieto Jiménez 1994) have focused their comparisons on the 1495 and 1516 editions—both of which were reprinted in the twentieth century. Unfortunately, these two editions were seldom the models for spinoff dictionary production in either the Americas or in Europe (but see Clayton 1989 and 2003 and Tellez Nieto 2010, 161). The 1545 Granada edition can be consulted online at www.juntadeandalucia.es/cultura/bibliotecavirtualandalucia/catalogo/catalogo_imagenes/grupo.cmd?path=10236
27 García Macho 2008, 1322.
28 Molina 1555, 223v, 171r.
29 Molina, 1555, 171r; see also 45v, 184r, 185r.
30 Molina 1555, iv v.
31 “que ellos tenian de que nosotros careciamos en nuestra lengua, no se pueden bien dar a entender, por vocablos precisos y particulares: y por esto assi para entender sus vocablos como para declarar los nuestros, son menester algunas vezes largos circunloquios y rodeos” (Molina 1555, iv v).
32 “En este vocabulario se ponen algunos romances, que en nuestro Castellano no quadran, ni se vsan mucho: y esto se haze por dar a entender mejor la prorpriedad dela lengua de los indios, y assi dezimos. Abaxador aunque no se vsa en nuestro romance: por declarar loque quiere dezir esta palabra. tlatemouiani, la qual en buen romance quiere dezir, el que abaxa algo” (Molina 1555, v r).
33 The idea that Nebrija’s publications were used to blindly straightjacket indigenous languages and categories into Latin models has been critiqued by Hernández de León-Portilla 1993; Nansen Díaz 1997; Manrique Castañeda 1997; and Galeote 2001, xxvi.
34 Molina 1555, 33r; the Mayan kakaw apparently derives from Nahuatl, and not the other way round; Macri and Looper 2003, 286.
35 Molina 1555, 30r. For a full list of Native American words already incorporated into Castilian in the 1555 and 1571 editions of Molina’s Vocabulario, see Galeote 2001: LIX-LXIII.
36 Molina 1555, 46v, 154r.
37 “CVRIOSO, el que trata alguna cosa con particuar cuydado y diligencia, y de alli se dixo curiosidad…” (Covarrubias 1611, 260r).
38 The first monolingual Spanish dictionary is Sebastián de Covarrubias y Orozco’s 1611 Tesoro de la lengua castellana o español; .pdf versions are available free on Google Books.
39 Both of these works are available via the Early English Books Online website (which, alas, requires a subscription); Percyualle is also available free on Google Books.
40 For Castilian-Tuscan, see Las Casas 1570 (and subsequent editions); for Castilian-French, see Pallet (1604 and 1606) and Ovdin (1607); all of these are available free on Google Books.