Mistaking an African language for an Amazonian one: The case of Arda

A few days ago I was thumbing through some issues of the Journal de la Société des Américanistes available online when I came across an 1910 article by Alexander Chamberlain entitled Sur quelques familles linguisitiques peu connues ou presques inconnues de l’Amérique du Sud. The article was in the main fairly uninteresting: it consisted of a list of language names with some basic geographical, classificatory, and bibliographic information for each language — a common enough genre of linguistics article at the time. However, one of the languages mentioned caught my eye: Arda.

According to the article, Arda was already extinct at the time the article was written, but was supposedly spoken on the Amazon between the Nanay and Marañon Rivers. The article also mentioned a source for data on this language: P. José de Nájera’s 1658 work, Doctrina cristiana, y explicación de sus misterios, en nuestro idioma español, y en lengua arda. I was, to say the least, stunned: I had never before come across mention of either this language name or this source — which is saying something, since this very part of the Amazon Basin has been a strong interest of mine for several years.

I was positively aquiver: either, I thought, this represented a source on a forgotten Amazonian language, or a forgotten source on a known Amazonian language — Yameo, I guessed (see below). Of course, I immediately went to the obvious sources: Campbell (1997) and Gordon (2005), and turned up absolutely nothing. Odd. Then onto Google, which turned up a couple of confusing results. My first major clue that something very peculiar was going on was an entry for de Nájera in the online Pequeña Encyclopedia Franciscana, the first part of which I reproduce here:

NÁJERA, José de (1621-1684). Capuchino, misionero y escritor ascético. Nació en Nájera (Logroño) el año 1621 y vistió el hábito de san Francisco en 1643. Estuvo de misionero en Arda (Africa) en 1660, pero por poco tiempo. En septiembre de 1661 llegó a la misión de Cumaná (Venezuela), donde permaneció hasta 1670, fundando la población de Nuestra Señora del Pilar. Volvió a España por enfermo y, en 1673, marchó a la misión de Los Llanos, de Caracas; aquí fundó el pueblo de San Antonio de Araure (Venezuela), donde falleció en 1684. Aparte de sus méritos como misionero, se distinguió por la santidad de vida.

Al P. José de Nájera se debe el primer impreso que se conoce de la lengua arda: Doctrina cristiana y explicación de sus misterios en nuestro idioma español y en lengua arda, Madrid 1658.

So first, Arda is identified as being in Africa (Arda was a West African kingdom, as well as the name used briefly by Europeans for the language spoken there), but even more tellingly, the publication of de Nájera’s work precedes his arrival in the Americas by three years. For that matter, even when de Nájera was in the Americas, he was over a thousand kilometers from where the nonexistent ‘Amazonian Arda’ was supposedly spoken. So Arda was obviously not spoken in the Americas, let alone on the banks of the Amazon. This page, from Dalby (1998) confirms this, and supplies the fact that the language in question belongs to the Gbe family of languages.

The story behind the confusion seems to be this: de Nájera’s work was published and quickly fell into obscurity, leaving apparently one known surviving copy. This one surviving copy was subsequently misclassified in the library that held the work as treating an American, and not an African, language. I infer de Nájera’s work must not provide the necessary geographical information to resolve the location of the language it treats, and I suspect that the emergence of the genre of Spanish missionary grammars in the New World led the misguided librarian to assume that de Nájera’s work — also a Spanish missionary grammary — must treat an American language.

The first clear sign of this error is found in a work by Henry Stevens, an American bibliographer who was heavily involved in collecting North and South American materials for the British Museum (among other employers). Google Books makes clear that he mentions de Nájera’s work in his 1851 Catalogue of the library of the Count Mondidier (actually a fictional personage invented by Stevens), but the first actual prose I could find is from his 1862 Bibliotheca Americana, where Stevens mentions the de Nájera volume (which Stevens oddly attributes to one Domingo Garcia Morrás), and remarks:

The Ardas are a barbarous tribe of Indians dwelling between the rivers Napo and Marañon, in the Province of Quijos, Quito. This, as far as we can learn, is not only the sole book published in the Arda Language, but is only copy known of it.

Stevens inter-continental error propagates through several other 19th century works on American Indian languages, such as Ludewig and Turner (1858), and Sabin and Eames (1873). (Actually, Ludewig and Turner (1858) is the earliest reference I have laid my eyes on that explicitly makes the intercontinental Arda error. Stevens (1851) is a truly obscure work, and I have yet to locate a copy I can check; here I am assuming — perhaps in error — that Stevens (1862) reprises Stevens’ (1851) identification of Arda. So the question of whether the original error can be attributed to Stevens or Ludewig is still open.)

In any event, we know that Paul Rivet located a copy de Nájera’s work. Rivet was understandably excited by this find, and we are told by Chamberlain that Rivet intended to publish de Nájera’s work as part of his broader efforts to republish colonial-era works on South American languages. Eventually, Labouret and Rivet (1929) published a work that correctly identified Arda as an African language, a discovery they attribute to Maurice Delafosse.

Now we come to the second major part of the story — how did linguists come to believe that a language named ‘Arda’ was spoken in the Amazon? The answer, as suggested by a brief allusion in Ludewig and Turner (1858), is that Antonio de Alcedo mentions the Arda in his 1786-9 encyclopedia, which was subsequently translated by G.A Thompson and published in 1812 as The geographical and historical dictionary of America and the West Indies (but see the Alcedo’s orginal title below! They really knew how to title a book in those days.). In the translated version, the entry on ‘Ardas’ reads as follows.

ARDAS, a barbarous nation of Indians, who inhabit the s. of the river Napo, and the n. of the Marañon, in the province of Quijos and kingdom of Quito. They occupy the thickest forests, and are bounded by the Maisamaes. (p. 98)

The geographical distribution that Alcedo gives for the Arda corresponds exactly to that of the Peba-Yaguan group now known by linguists as the Yameo, whose language became extinct in approximately 1950. The fact that Alcedo uses an obscure name for this group is hardly surprising. The 17th and 18th century colonial records for this part of the Amazon Basin provide a bewildering variety of names for the indigenous groups of the region. Rivet (1913), for example, lists Nahuapos, Amaonos, Massamaes, Migueanos, Napeanos, Parranos, Yarrapos, and Alabonos as variant names for Yameo, and it seems fairly clear that Arda is simply another one.

So, the basis of the confusion seems clear enough from this vantage point: based on the assumption that Arda was an American language, either Stevens or Ludewig went prowling thought the historical/geographical/ethnographic materials available at the time, looking for a name that matched. Then, finding a match in Alcedo’s historical-geographical dictionary, they concluded that de Nájera’s (African) ‘Arda’ and de Alcedo’s (American) ‘Arda’ were one and the same.

So, is there any edifying moral to be drawn from this affair? If anything, I think it would have to be this: approach colonial-era names used by Europeans for Amazonian indigenous groups with great care. In fact, although the case of Arda is perhaps the most spectacular error of this type that I have run across, I’ve seen a few errors of this type by modern authors as well. But that’s a matter for a later post!

References Cited (Note that I have not been able to provide full references for all works.)

de Alcedo, Antonio. 1786-89. Diccionario geografico-histórico de las Indias Occidentales o America: es a saber: de los Reynos del Peru, Nueva España, Tierra Firme, Chile y Nuevo Reyno de Granada. Con la descripción de sus Provincias, Naciones, Ciudades, Villas, Pueblos, Rios, Montes, Costas, Puertos, Islas, Arzobispados, Obispados, Audiencias, Virreynatos, Gobiernos, Corregimientos, y Fotalezas, frutos y producciones; con expresión de sus Descubrimientos, Conquistadores y Fundadores: Conventos y Religiones; erección de sus Catedrales y Obispos que ha habido en ellas: Y noticia de los sucesos más notables de varios lugares: incendios, terremotos, sitios, é invasiones que han experimentado: y hombres ilustres que han producido. Madrid, Imprenta de Blas Roman, 5 volúmenes.

de Alcedo, Antonio and George Alexander Thompson. 1812. The geographical and historical dictionary of America and the West Indies. London

Campbell, Lyle. 1997. The American Indian languages. OUP.

Chamberlain, Alexander. 1910. Sur quelques familles linguisitiques peu connues ou presques inconnues de l’Amérique du Sud. Journal de la Société des Américanistes. 7 (1-2). pp 179-202.

Dalby, Andrew. 1998. Dictionary of Languages: The definitive reference to more than 400 languages. London: Bloomsbury.

Gordon, Raymond G., Jr. (ed.), 2005. Ethnologue: Languages of the World, Fifteenth edition. Dallas, Tex.: SIL International. Online version: http://www.ethnologue.com/.

Labouret, Henri and Paul Rivet. 1929. Le Royaume D’Arda et son Evangelisation au XVII Siecle. Paris. Intitut d’Ethnologie.

Ludewig, Hermann and W. Turner. 1858. The literature of American Indian languages. London: Trübner and Co.

de Najera, José. 1658. Doctrina cristiana y explicación de sus misterios en nuestro idioma español y en lengua arda. Ms.

Rivet, Paul. 1913. La famille linguistique Peba. Journal de la Société des Américanistes. 10 (1). pp 119-171.

Sabin, Joseph and Wilberforce Eames. 1873. A dictionary of books relating to America, from its discovery to the present time. New York: J. Sabin and Sons.

Stevens, Henry. 1851. Catalogue of the Library of Count Mondidier.

Stevens, Henry. 1862. Historical Nuggets: Bibliotheca Americana. Whittingham and Wilkins.

10 thoughts on “Mistaking an African language for an Amazonian one: The case of Arda

  1. Great detective work Lev! I’ve been doing a lot more historical work lately – it’s a lot of fun but you’re quite right that nothing can be accepted without double-checking.
    A different kind of error crops up with the Jivaro (or Jibaro), as they get conflated with the geographically not-too-distant Jebero. From what I can tell, this error was introduced by William Bollaert, who translated a number of works from Spanish into English in the 19th century. Although the names are quite distinct in the original works (at least the ones I’ve seen), Bollaert apparently just assumed that they were orthographic variants.

  2. Ouch! Conflating ‘Jebero’ and ‘Jibaro’ certainly would result in some major confusion. I’m curious how far this error propagated in the literature until linguists eradicated it.

  3. Luckily the error was not widespread, and must have been quickly picked up. Jijón y Caamaño (1919) says:

    “La lengua jíbara fue confundida durante algún tiempo con la jébaro, por la semejanza de nombre y este error, aceptado generalmente, fue rectificado por González Suárez, en 1904, quien distinguió claramente el área en que se usan ambas lenguas.”

    González Suárez. 1904. Prehistoria Ecuatoriana. Quito
    Jijón y Caamaño. 1919. “Contribución al conocimiento de las lenguas indígenas, que se hablaron en el Ecuador interandino y occidental…” Boletín de la Sociedad Ecuatoriana. 2(6) pp.340-413. Quito.

  4. Thanks, Simon. It’s nice that the confusion did not continue to propagate.

    Although not of quite the same nature, one bit of confusion that *has* propagated in certain quarters — even to the present — is the misidentification of Taushiro as a Zaparoan language. My hunch is that the Taushiro and their neighbors, the Andoa (who are Zaparoan) had very close relations, and that this led the early missionaries to assume that the two groups were linguistically related. This error was repeated in Beuchat and Rivet (1908), and although most classifications after about 1960 do not continue to repeat the error, some do.


    Beuchat, Henri and Paul Rivet. 1908. La famille linguistique Zaparo. JSAP 5: 235-249

  5. Efectivamente, Lev, la identificación errada del taushiro como lengua Záparo está presente aún hoy. He visto que en el Multilingüismo en el Perú se comsidera el Taushiro en el tronco Zaparo. Solis Fonseca por su parte dice:
    “la lengua taushiro conocida también como pinche, cuyos hablantes se reconocen como ite´chi y se ubican en el distrito de Tigre del departamento de Loreto, debe ser considerada una rama divergente dentro de las lenguas que coforman el grupo Záparo “(Solis lenguas en la Amazonía Peruana 2002: 200)subrayado mío)
    Así pues, el error entre los lingüistas peruanos está presente, en cambio los antropólogos del medio tienen la cosa más clara:
    “La región que actualmente ellos [los quichua del pastaza] y los del Napo y Tigre ocupan, fue parte del gran territorio de pueblos del tronco záparo: andoa, gae, semigae, aushiri, arabela y záparo propiamente dicho, más otros cuya filiación a esta familia es dudosa, como los roamayna, omurano, abishiri y los pinche o taushiro” (Chirif y otros. Una historia para el futuro: Territorios y pueblos indígenas en Alto Amazonas, IWGIA, 2002: 82)


  6. ah, debo señalar que en el libro de Pozzi Escot cuando se habla de la clasificación genética no se dice enfáticamente que es del tronco záparo sino: “posiblemente miembro de la familia záparo”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s