The Truth About Taensa

In the 19th and early 20th centuries, scholarly attitudes toward colonized cultures became increasingly governed by the idea of “salvage ethnography.” The logic went that these cultures were “dying out,” with traditions and languages being supplanted by those of the dominant groups, and that they needed to be recorded for study.

Stop me if you’ve heard this one—Europeans and white Americans motivated by complex mixtures of greed, pity, scientific interest, lust for fame, adventure-seeking, exoticism, and a desire to help swooped down on Native groups to see what they had, what they thought, and how they spoke. Many Native groups had cultural treasures, even human remains, taken from them in the name of “safekeeping,” which often meant they were sprayed with arsenic (as a pesticide) and locked in a vault. Photographs and recordings prioritized “authenticity,” which in some contexts meant trying to get actual Native Americans to look and act more like preconceived notions of stereotypical “Indians” to create more compelling narratives.

But one silver lining was that some linguists and anthropologists did record examples of indigenous languages that were falling out of use. The stories, glossaries, and grammars in Native American languages preserved through this work have emerged as valuable assets for modern tribes and researchers (Natives, linguists, and Native linguists) hungry for information about and examples of these languages. In some cases, this was where the idea of “salvage” lived up to its most generous interpretation, preserving these tongues until their speakers’ heirs had a plan to bring that heritage home.

When it comes to the case of the Taensa language, though, things get much more complicated.

The language of the Taensa people, for whom is named Tensas Parish in the Louisiana Delta, seems to have been the subject of a bizarre fraud by a young French seminary student named Jean Parisot. Parisot claimed to have found, in his late grandfather’s papers, an undated manuscript with no indication of who authored it, written in Spanish comprising enough material in the Taensa language to release a journal article in 1880, twelve songs in 1881, and a grammar and word list in 1882, for a total corpus of about 600 Taensa words. These works intrigued the French linguistic establishment and seem to have been broadly accepted, but in 1884 they attracted the attention of Daniel Brinton, an American surgeon, ethnologist, and editor.

"Why was this Spanish monk who allegedly composed the original manuscript effectively “going rogue” and evangelizing in what we know to have been a French mission field? Why do no French accounts mention this Spaniard wandering around? And why is his Spanish bad? When Parisot was later pressed to provide the original manuscript, he would claim to have lost it—and that no, it was actually in French, and probably left at his grandfather’s boardinghouse by a traveler." 

Brinton himself seems to have been eccentric—late in life, he became a devoted anarchist—but he was a well-regarded scholar, and in 1885 he wrote a gleeful takedown of Parisot’s Taensa titled “The Taensa Grammar and Dictionary: A Deception Exposed”. (That same year, he published a new translation of the Walam Olum, purportedly a sacred text of the Lenape tribe collected in the 1830s, which would itself be widely considered a fraud after scholarly reassessment in the late 20th century. One is tempted to say “you can’t make this stuff up,” but….)

The provenance is terrible, Brinton points out. Why was this Spanish monk who allegedly composed the original manuscript effectively “going rogue” and evangelizing in what we know to have been a French mission field? Why do no French accounts mention this Spaniard wandering around? And why is his Spanish bad? When Parisot was later pressed to provide the original manuscript, he would claim to have lost it—and that no, it was actually in French, and probably left at his grandfather’s boardinghouse by a traveler.      

Brinton also argues that the grammar of Taensa itself is not much more credible—it is certainly possible that the language could have certain grammatical features seldom, if ever, seen in the indigenous languages of North America—even several of them. But it is unlikely and not convincingly explained.      

Finally, the texts describe a Louisiana full of sugar maples, oxen, and bananas, among other improbable flora, fauna, and customs—complete with snowy winters. “It may be urged that these are all mistranslations of misunderstood native words,” Brinton poses. “To this I reply, ‘what sort of editing is that which not only could commit such unpardonable blunders, but send them forth to the scientific world without a hint that they do not pretend to be anything more than guesses?’”

[Read this poem, written in the Louisiana Creole language, Kouri Vini.] 

Brinton’s thesis is that Parisot picked the name of a relatively obscure Native group from a map, then built a fictional language as a prank; when it caught on, he realized they could make a little money by leaning into the fraud. While some of the people taken in by the language continued to defend it, fueling an existing tension between French and American scholars of American languages, the general consensus has been close to Brinton’s thesis: Taensa was the Elvish of exoticism, the Klingon of colonialism.

The Taensa people did exist, though. They lived, at least for a time, in the general area of Tensas Parish, where in the early 18th century French missionaries met them. In 1908, the linguist John Swanton would use accounts written by these missionaries to publish his own refutation of Parisot’s Taensa. Swanton points out that the missionaries planned to consolidate their missions to the Taensa and the Natchez into one—an unnecessary complication if the two tribes spoke unrelated tongues. In fact, by all indications, the Taensa, Natchez, and French missionaries were able to communicate relatively easily, and the missionaries record the cultural affinity between the Taensa and the Natchez (even though the groups were at war at the time). For Swanton, the fact that the Natchez and the Taensa were so similar and do not seem to have needed to resort to Mobilian, the lingua franca among most central Gulf tribes, indicated the languages were similar or the same.      

As more Europeans arrived, Native groups rearranged their relationships to resist or accommodate them, and the Taensa people pinballed around the southeast, accepting Bayougoula hospitality before turning on their hosts, taking refuge near Mobile and then leaving when it came under British control, and ultimately moving back into modern Louisiana. After many ructions, the Taensa broadly assimilated into Chitimacha settlements, no longer with a distinct territory but maintaining an identity and language until sometime around 1900. (They are not the same as the Avoyel-Taensa, who are related to tribes of the Ohio Valley.) Swanton recounts his own attempt, during a 1907 visit to Charenton, to persuade an older Chitimacha woman whose father had identified as Taensa to share some of the Taensa language, but suffering memory loss and not having used the language in many years, she was unable. (There is, of course, always the possibility that for her own reasons she simply didn’t wish to.)

Perhaps ironically, if the Taensa were Natchez speakers, they did indeed speak a language that has intrigued scholars. Natchez is a language isolate, meaning it has no clear relationship to any other language: it’s part of a cluster of unique languages in the central Gulf that have bedeviled linguists’ efforts to categorize them. Natchez has the complex verbs common in American languages—if the verbal endings in French intimidate you, know that Natchez verbs can take several suffixes and several prefixes for an action-packed action-word juggernaut—and a special “cannibal speech,” used when telling stories involving the cannibal characters common in Natchez lore. A Taensa language that shared traits like these would have delighted researchers and armchair linguists on its own merits.

However, Parisot’s Taensa has itself continued to attract attention. Usually, someone reexamines it and dismisses it again as a hoax, but the Australian linguist Claire Bowern has looked with a provisionally open mind. Harvard-educated and now at Yale, Bowern has primarily worked on language change and documentation in the Aboriginal languages of Australia, but when she was invited to give the prestigious Whatmough lecture for the Harvard linguistics department, she turned to Parisot’s Taensa.

"For Bowern, it’s easier to explain the existing Taensa corpus as the work of a very young man who didn’t know what he had and didn’t know what to do with it." 

Bowern takes issue with Brinton’s glib dismissal of the possibility of a language with grammatical features as described in Parisot’s Taensa. To take one relatively straightforward example, Brinton raised issue with the “three plurals” used in Parisot’s Taensa, which for Bowern, could simply be variations in a pattern that changes depending on the word made plural—“utterly standard for languages the world over.” (Think cats, babies, fishes, oxen.) She goes on to point out that maples are found in parts of northern Louisiana and that icy winters do on occasion happen there. The relevance of the “exotic” fauna depends on when the texts were collected, she said: Native groups had their own trade networks and participated in those the Europeans established—the Taensa might indeed have had bananas and potatoes, or at least words for them, later in their existence. Could the Taensa word for “polar bear” be a misreading for some other term—maybe a pale or shiny-coated black bear? She also points out that multilingualism was extremely common among Native groups in what became the southeastern and south-central United States: “the Taensa spoke Natchez” is not the same proposition as “the Taensa spoke only Natchez.”

[Read about the Natchez War here.]

Bowern also fills in more of Parisot’s personal history. The seminary student went on to work as a linguist in the Ottoman Empire—he was there studying Syriac when Brinton fired off his cannonade. But at least at the time he published on Taensa, he was not very good. Bowern described the work on Taensa, if genuine, as deserving a B minus or a C plus, with inconsistent and incomplete points and labeling any verb with a complex pattern as “irregular”. Parisot also fails, in both the original work and any subsequent defense, to note the existence of possible cognate words between Taensa and the Timucuan language of Florida and Georgia. As Bowern notes—if he had intentionally used those cognates to make the hoax credible, he would have been able to point to them as evidence of the language’s legitimacy. As for the other “exotic features” of Taensa—would a middling linguist, familiar only with European languages, be able to come up with these features that actually occur in some world languages?

For Bowern, it’s easier to explain the existing Taensa corpus as the work of a very young man who didn’t know what he had and didn’t know what to do with it. (She uses phrases like “doing a poor job on challenging material,” a description I have repeatedly feared would apply to the present article). She hasn’t proven Parisot’s honesty or the existence of a separate Taensa language, but she certainly creates reasonable doubt. It would be good to believe that the nearly forgotten language of an assimilated tribe found an imperfect champion in a French seminary student. So many of the world’s languages exist in fragmentary forms that don’t offer linguists or descendant communities much to work with: wax cylinders, weathered carvings, personal names. We can see where a language was, but just a shadow. If the much-mocked Taensa record is real, it is sacred, a last whisper of a way of life, a way of thought, and Parisot deserves praise for amplifying this last utterance—before he lost the damn manuscript.