Texas Beyond History
TBH Home
Tejas Main

Caddoan Languages and Peoples

Map showing approximate areas where Caddoan languages were spoken in the 17th and 18th centuries
Approximate areas where the various Caddoan languages were spoken in the 17th and 18th centuries. The arrows show the general directions of group migrations during this period.

Within the main Caddo Homeland, the cultural continuity is unbroken from prehistory to early history and the link to today's Caddo Nation of Oklahoma is unquestioned.

John Powell
Ethnologist and famed explorer John Wesley Powell formally defined the Caddoan language family in 1891. Courtesy Smithsonian Institution.
Arikara hoe
Arikara bison-scapula hoe. This type of tool was used in farming by main Plains groups. Smithsonian Institution National Anthropological Archives. Click to enlarge.
Walter Ross
Walter Ross, a Wichita, ca. 1927. Photograph by Edward S. Curtis, The North American Indian, Volume 19.
Wichita grass house
Wichita grass house, ca. 1927. Photograph by Edward S. Curtis, The North American Indian, Volume 19.
Kichai woman and Wichita man
"Nasutoeas, Kichai Woman, Akahedik (Wichita)," ca. 1898. Photograph by F. A. Rinehart, courtesy Omaha Public Library. The Kitsai (Kichai) are the least known of the Caddoan language groups. The Kitsai tribe no longer exists as a separate entity; surviving members joined the Wichita in the mid-1800s.
buffalo
Small protected buffalo herd grazing near Wichita Mountains, southwest Oklahoma, 1908.
Buffalo Bill
"Buffalo Bull: A Grand Pawnee Warrior," by George Catlin, 1832. The Pawnee relied heavily on buffalo and, in early historic times, lived in what is today Nebraska.
Pawnee earth lodges
Pawnee earth lodges and corral, Nebraska Loup Fork Village, late 19th century. Smithsonian Institution National Anthropological Archive.
Wichita Tribal Center
Wichita Tribal center near Anadarko, Oklahoma. Photo by Steve Black.
Caddo country
The rolling prairie in west-central Oklahoma where the Wichita and Caddo tribes in the 1860s were at last given small territories and allowed to settle in peace. Photo by Steve Black.
Pawnee Village
Pawnee village, ca. 1875. In the background are two massive earth lodges. Photograph by William H. Jackson. The northernmost Caddoan groups, the Pawnee and their close relatives the Arikara, lived in earth lodges to survive the brutal winters of the Central and Northern Plains.

The Caddoan languages are Caddo, Wichita, Pawnee, Arikara, and Kitsai, the latter four making up the Northern Caddoan languages. The speakers of the Northern Caddoan languages are also referred to as the Plains Caddoans because all four tribes (and their various bands) lived in the Southern and Central Plains during historic times. Caddo is the only Southern Caddoan language.

"Caddo" vs. "Caddoan"

The words "Caddo" and "Caddoan" have been used to mean different things by different researchers and writers, adding considerable confusion to the complex and often impossible task of understanding the relationships among historic tribes and their ancestors. In the Tejas exhibits we have tried to use these terms consistently.

In historical documents the word Caddo is sometimes used to refer just to the Cadohadacho, but today it is more commonly applied to all of the Caddo-speaking groups. Although there was not a united Caddo Tribe until 1874 when, under pressure from the U.S. government, the remnants of the groups speaking various dialects of the Caddo language formally joined together for survival, the term Caddo is still very useful to refer to (1) the united Caddo Tribe; (2) all of the groups known to have spoken Caddo dialects before 1859; (3) the Caddo language; and (4) the direct ancestors of the Caddo-speaking groups as inferred from archeological evidence. Further the word is used both as a singular (i.e., the Caddo village) and plural noun (i.e., the Caddo were corn farmers), although the plural form, Caddos, is also used.

Fine so far: Caddo means anything to do with the Caddo Tribe and its direct ancestors. It is the term "Caddoan" that causes trouble. In normal English language usage, the word can correctly be used as the adjectival form of Caddo (i.e., the Caddoan village). But it took on another meaning in 1891 when ethnologist and famed explorer John Wesley Powell formally defined the Caddoan language family. Powell, like other scholars before him, recognized that the Caddo language was closely related to the Wichita, Pawnee, Arikara, and Kitsai languages and, rightly, lumped them together in one language family.

Since the 1940s, archeologists have used the term Caddoan Area to refer to the southern and easternmost region containing prehistoric and historic sites linked to the ancestors of the Caddoan language family groups, although this term has fallen out of common usage. The Caddoan Area was mainly the home of the Caddo-speaking groups; however, its northern part may have been occupied in prehistoric times by some of the ancestors of certain other Caddoan groups, notably the Wichita and Kitsai. This is uncertain because the ancestors of all Caddoan language groups appear to have migrated over hundreds of miles during their histories. (Except the ancestors of the Caddo language speakers, who seem to have stayed more or less in one general area throughout their known history.) In 1542 when the De Soto entrada traveled through parts of the area, all of the Caddo groups who were encountered apparently spoke a Caddo dialect. Unfortunately, the Caddoan Area as traditionally defined includes both the main area that we will call the "Caddo Homeland" as well as what is often called the "Northern Caddoan Area."

The main Caddo Homeland lies south of the Arkansas River in the valleys and tributaries of the Ouachita, Red, Sabine, and Neches rivers where the historically documented Caddo speakers lived until the 19th century. This area, sometimes referred to as the "Southern Caddo Area," has abundant and unmistakable archeological evidence that the direct ancestors of Caddo-speaking peoples lived there since at least A.D. 800 and probably for 3000-4000 years or longer, perhaps much longer. In other words, within the main Caddo Homeland, the cultural continuity is unbroken from prehistory to early history and the link to today's Caddo Nation of Oklahoma is unquestioned.

In contrast, the cultural continuity of the Arkansas Basin (the so-called "Northern Caddoan Area"), the valleys of the Arkansas River and its tributaries and adjacent southern Ozark Highlands in northeastern Oklahoma northwestern Arkansas, and even southwest Missouri, was broken prior to the arrival of the earliest European visitors. This circumstance is discussed elsewhere (see "Spiro and the Arkansas Basin"). Briefly, it is not known whether the area was occupied by the ancestors of Caddo-speaking peoples, the ancestors of the Kitsai, or the ancestors of the Wichita.

The Caddoan archeological tradition clearly represents the ancestors of the Caddo, but it may also, in part, include archeological sites occupied by the ancestors of the Wichita and Kitsai. In fact, as explained below, thousands of years ago in Woodland and Archaic times it is likely that the ancestors of all the Caddoan language family lived in or near the Caddoan Area.

To avoid the confusion between the Caddoan language family and the word Caddoan as the adjectival form of Caddo, throughout the Tejas exhibits we use the term "Caddo" to refer only to the Caddo speakers, their language, and their direct ancestors as identified archeologically. We use the term "Caddoan" only in the linguistic sense to refer to the Caddoan language family.

Brief History of the Caddoan Language Family

Before modern transportation and communication systems existed, languages spread and changed in similar, fairly predictable ways. When a people speaking a common language split apart, with one group migrating elsewhere and becoming geographically isolated from the other, the "mother" language as spoken by each group gradually changed over time. For instance, new words may be coined, old words may be dropped, and pronunciation changes. (Witness the differences in the English spoken by Britains, Americans, and Australians.) The mother language becomes two dialects that, over time, become more and more different until eventually the speakers of one dialect cannot understand the other. This is how new languages form.

Linguists are the specialists who study languages and how they relate to one another. They have worked out the basic relationships among most of the world's surviving languages and have classified them into various families and branches. Linguists often use a branching tree as a metaphor for how languages are related. English, German, Dutch, and Yiddish, for instance, are all Germanic languages that represent the West Germanic branch of the Indo-European language family. Similarly, the Romance languages that developed out of Latin, such as Spanish, French, and Italian, form the Italic branch of the Indo-European language family.

By systematically studying how different and similar two related languages are, linguists can estimate how long ago they split apart, a technique called glottochronology. Glottochronology is controversial because, among other reasons, not all languages change at the same rate and because it is often difficult or impossible to compare languages that are poorly known. A great many languages have become extinct in the last few centuries including over half the languages spoken by Native American peoples 500 years ago. Still, studying the relationships among languages is a powerful way of reconstructing the early histories of different peoples across the world.

Unfortunately, the last published glottochronology for all the Caddoan languages dates to the 1960s, before the technique was refined. According to this estimate, the Caddoan languages did not begin splitting apart until 3,000 to 3,500 years ago. Some Caddo archeological experts such as Timothy K. Perttula reject this estimate and suggest that the initial split between Southern and Northern Caddoan languages (see below) may have taken place thousands of years earlier. That said, as one goes back in time, confidently linking language and ethnicity with archeological evidence is increasingly difficult, if not impossible.

Nonetheless, the linguistic estimate is that prior to about 3,500 years ago, the distant ancestors of all of the Caddoan groups were a single people who spoke an ancestral language that linguists call Proto-Caddoan. Of course, Proto-Caddoan has long been extinct or, rather, it evolved into the various Caddoan languages as the Proto-Caddoan ancestors split apart and went their separate ways. It is estimated that about 3,500 years ago (1500 B.C.) the ancestors of the Northern Caddoan groups split from the Caddo and the two languages began to change. Sometime after the time of Christ, the Proto-Northern-Caddoan speakers began splitting off from one another, first the Wichita, then the Kitsai, and finally the Pawnee. Still later, only 400-500 years ago, the Arikara split from the Pawnee.

The Caddoan languages are distantly related to the Iroquoian languages (such as Iroquois and Cherokee) and even more distantly related to the Siouan languages (such as Dakota and Crow). Because of certain similarities, linguists theorize that these three language families have had a common origin (a shared ancestry) at some remote point in time, probably in the central part of the country, perhaps somewhere along the central valley of the Mississippi River. But this was so long ago (perhaps more than 10,000 years?) that any historical reconstruction is little more than a guess. Within the Caddoan language family, however, we can reconstruct at least the general patterns of movement over time.

All of the groups that spoke a Caddoan language lived west of the Mississippi River, along its western tributaries. During historic times the Caddoan groups were spread across an area that spanned about 1200 miles north-south and almost 500 miles east-west. At historic contact, the latest groups that had split off among the Northern Caddoans lived the farthest north. The Arikara lived in what is now South and North Dakota, while the Pawnee lived in present day Nebraska. The Northern Caddoan groups that had split off earlier, the Witchita and Kitsai, lived between the Pawnee and the Caddo, in what is today Kansas and Oklahoma. Based on such geographical clues, linguists surmise that the original homeland of the Proto-Caddoan speakers was in the forested western fringe of the Eastern Woodlands, within or very near what has been recognized as the Caddo Homeland.

To replay the outlines of Caddoan history, we can guess that Proto-Caddoan ancestors lived in the Caddo Homeland, perhaps in or near the valleys of the Red and Arkansas rivers and the intervening Ouachita Mountatins. One group stayed on and became the Caddo and another split off and began moving north and west, probably up the Red and Arkansas river systems. The Proto-Northern Caddoan speakers gradually moved farther out onto the Plains and split apart as they moved west and north. The ancestors of the Pawnee and Arikara moved farther and farther north and west up the Missouri River and its tributaries, eventually losing all memory of the Caddo. The ancestors of the Wichita and Kichai stayed in the Southern Plains. Nonetheless, the Caddo were separated from all of the Northern Caddoan groups long enough ago that they had no tradition of a common ancestry, nor could they speak to one another.

By the end of the Plains Woodland era (about A.D. 900), if not before, the ancestors of most (all?) of the Northern Caddoan peoples were Plains villagers, farmers and buffalo hunters who lived in villages scattered through the wooded valleys across the Plains. Some archeologists think that Caddoan-speaking groups spread westward across Oklahoma, north Texas including the Panhandle, and Kansas, as far as the eastern foothills of the Rocky Mountains in what is today northeastern New Mexico and southeastern Colorado. After A.D. 1350 in the 14th and 15th centuries, the "southwestern" Plains villagers abandoned that area and moved north and east, apparently in response to climatic changes and the encroachment from the west and northwest of Apachean peoples. The ethnic affiliations of the southwestern Plains villagers are not known, but some archeologists believe they may have included the ancestors of the Pawnee/Arikara and the Wichita.

In contrast with the Caddo, who stayed put in their original homeland, all of the Northern Caddoan groups appear to have migrated hundreds of miles during the last two millennia. Their inferred early history makes ecological sense. The relatively dry climate of the southern and central Great Plains is prone to periodic drought and thus is often marginal for dry land farming. (The western Caddoan Homeland is also drought prone, but to a lesser extent.) And without modern machinery and irrigation, the great grasslands of the Plains could not be farmed. Hence the Plains villagers lived along the relatively narrow and well-watered river valleys where farming was possible. The bands of each group had to spread out along the narrow valleys and were susceptible to raids from enemy groups. Raiding and climatic change are two of the main factors that explain why the Plains Caddoans moved from place to place. The complex histories and migrations of the Wichita, Kitsai, Pawnee, and Arikara peoples reflect their precarious existence on the Plains.

Northern Caddoan Peoples

Sadly little can be said about the poorly known Kitsai tribe (also spelled Kichai). The Kitsai language is no longer spoken and only a bit of it was recorded before the last Kitsai speaker passed away in the 1930s. The tribe no longer exists as a separate entity; surviving members joined the Wichita in the mid-1800s. The Kitsai appear to have been farmers and hunters, like all Caddoan peoples, and are mentioned in various French and Spanish documents. Throughout their known history during the 18th and early 19th centuries, the Kitsai were relatively few in number and divided into two groups, a northern band allied with the Wichita, and a southern band allied with the Cadohadacho and other Caddo groups. Their known territory was in south-central Oklahoma and north-central Texas, just west of the Caddo, especially along the Red River. Several archeological sites in north Texas have been linked, speculatively, to the Kitsai, including the mid-18th century Gilbert site, although the evidence is not compelling. Some archeologists believe that Kitsai ancestors were the prehistoric people of Spiro and the Arkansas Basin.

The Wichita are much better known than the Kitsai, because they were a more numerous people, and because they survived as a tribe. Like the Caddo, the Wichita were made up of a number of related, but independent groups including the Tawakoni, Yscani, Hueco, and Wichita proper, that probably each spoke a separate dialect. The Wichita groups (along with the remaining Kitsai) became a single tribe in 1835 when they signed a treaty with the United States. Ancestral Wichita groups were first encountered in 1541 by Coronado's expedition in the vicinity of the Great Bend of the Arkansas River in present day south-central Kansas. The Spanish named the area Gran Quivira and reported visiting a series of large villages, some containing 200 large dome-shaped grass houses similar to those built by the Caddo. During the historic era, the Wichita groups moved southward through Oklahoma and into Texas as far south as Waco, which was named after the Wichita Hueco band whose village once stood where the city was built. Like the Caddo, the Wichita were resettled in Indian Territory after the Civil War and today maintain a tribal center near Anadarko, Oklahoma.

The Pawnee and Arikara had a shared history (i.e., were one people) until splitting apart perhaps 400-500 years ago, just before historic contact. Their ancestors have been identified archeologically as the Upper Republican phase of the Central Plains Village tradition in Kansas and Nebraska. After they split apart, the Arikara moved farther north into what is today South Dakota. Both groups lived in earthen lodges in compact villages that were sometimes fortified. Like other Caddoans, both groups had a mixed economy with farming and buffalo hunting being important. The Pawnee relied heavily on bison, while the Arikara were also fishermen as well as traders. Prior to consolidation during the 19th century, both the Arikara and Pawnee were made up of independent bands speaking their own dialects. Today the Arikara remain in North Dakota, where they settled on a reservation with the Sioux-speaking Mandan and Hidatsa. The Pawnee have a tribal center in north-central Oklahoma, where they were given land in 1876 in exchange for giving up much of Nebraska.

 

"Arikara Village of Earth-covered Lodges"
"Arikara Village of Earth-covered Lodges, 1600 Miles above St. Louis," by George Catlin, 1832. The Arikara are the northernmost of the groups who spoke one of the Caddoan languages. Today they share a reservation with the Sioux-speaking Mandan and Hidatsa in North Dakota.

Click images to enlarge  

Fred Carruth
John Tatum, a Wichita man, and six others, (on the left is Nasutoeas,a Kitsai woman), ca. 1898. Photograph by F. A. Rinehart, courtesy Omaha Public Library..

Throughout the Tejas exhibits we use the term "Caddo" to refer only to the Caddo speakers, their language, and their direct ancestors and … "Caddoan" only in the linguistic sense to refer to the Caddoan language family.

Fred Carruth
Fred Carruth, a young Wichita man, ca. 1898. Photograph by F. A. Rinehart, courtesy Omaha Public Library.
buffalo
All of the Plains Caddoan groups depended on bison herds for food, clothing, and tools. The Pawnee, in particular, were famed bison hunters. Courtesy Texas Parks and Wildlife Department.
Caddoan language diagram
This diagram shows the relationships among the various Caddoan languages and indicates the order in which the various language groups are thought to have split from one another. How long ago the various splits occurred is very poorly known. Linguistic estimates suggest that, prior to about 3,500 years ago, the ancestors of all of the groups were a single people who spoke an ancestral language that linguists call Proto-Caddoan. About 3,500 years ago (1500 B.C.) the ancestors of the Northern Caddoan groups split from the Caddo and the two languages began to diversify. (Some archeologists think the initial split may have occurred thousands of years earlier.) Sometime after the time of Christ, the Proto-Northern-Caddoan speakers began splitting off, first the Wichita, then the Kitsai, and finally the Pawnee. Still later (not long before Europeans arrived) the Arikara split from the Pawnee and Pawnee split into two groups.
Chief Boss Sun
Pawnee Chief Boss Sun wearing bear claw necklace, peace medal, and holding feather fan. Late 19th century. Smithsonian Institution National Anthropological Archives.
Arikira woman
"The Rush Gatherer," an Arikira woman, ca. 1908. Photograph by Edward S. Curtis, The North American Indian, Volume 5. The Arikara were the last of the Caddoan language groups to develop, they split apart from their close relatives, the Pawnee, about 400-500 years ago.
corn
All Caddoan groups were corn farmers to varying degrees. Corn was more important to the Caddo in large part because climatic conditions in the Caddo Homeland was much more favorable to growing corn than it was on the central Plains were the Pawnee and Arikara lived. Photograph by Frank Schambach.
Chief Knee-War-War
Kitsai Chief Knee-War-War in partial native dress with ornaments, 1872. Smithsonian Institution National Anthropological Archive.
Whichita grass house
Wichita grass house on display at Indian City, Anadarko, Oklahoma. Photograph courtesy Dee Ann Story.
Wichita grass-house ceremony
Wichita grass-house ceremony, ca. 1927. Photograph by Edward S. Curtis, The North American Indian, Volume 19.