Plains Earthlodges: Ethnographic and Archaeological Perspectives

Plains Earthlodges: Ethnographic and Archaeological Perspectives

NOOK Book(eBook)

View All Available Formats & Editions

Available on Compatible NOOK Devices and the free NOOK Apps.
WANT A NOOK?  Explore Now
LEND ME® See Details

Product Details

ISBN-13: 9780817384241
Publisher: University of Alabama Press
Publication date: 04/10/2005
Sold by: Barnes & Noble
Format: NOOK Book
Pages: 248
File size: 9 MB

About the Author

Donna C. Roper is Adjunct Associate Professor of Anthropology at Kansas State University and editor of Medicine Creek: Seventy Years of Archaeological Investigations.

Elizabeth P. Pauls is State Archaeologist and Director of the University of Iowa's Office of State Archaeologist.

W. Raymond Wood is Professor Emeritus at the University of Missouri-Columbia and author of numerous publications, including Prologue to Lewis and Clark: The Mackay and Evans Expedition.

Read an Excerpt


Ethnographic and Archaeological Perspectives


Copyright © 2005 The University of Alabama Press
All right reserved.

ISBN: 978-0-8173-5163-2

Chapter One

What, Where, and When Is an Earthlodge?

Donna C. Roper and Elizabeth P. Pauls

The earthlodge, as we discuss it in this volume, is a class of permanent, timber-frame, earth-covered dwelling (Figure 1.1) built by the agricultural village tribes of the Missouri River valley and its tributary valleys in the North American Great Plains. Plains earthlodges (also "earth lodges" or "earth-lodges" in the various sources) are prominent in the oral traditions, mythologies, and cosmologies of the people who lived in them; they are amply described and illustrated in nineteenth-century sources by explorers, traders, and travelers, as well as by later ethnologists; and they are well attested in the archaeological record of the Plains. Untold thousands of these structures were built over an area encompassing nearly eight degrees of latitude and about six or seven degrees of longitude and over a period of about a millennium (depending on what one actually admits to the earthlodge class, as we shall discuss), by around a dozen tribes of Northern Caddoan and Siouan affiliation. They were, truly, a major native North American house type.

In this volume, we explore various aspects of earthlodges asstructures, as living space, as symbolic space, and as important features of the Plains archaeological record. Although space limitations and author self-selection prevent us from covering all aspects of earthlodges, we believe that the papers assembled here sample the diversity of topics and approaches that are necessary for an understanding of these dwellings and of the context in which they were used. To make it clear to our readers exactly what set of phenomena we are considering, we introduce the book by reviewing sources of information about earthlodges, their physical description and construction, and their distribution in space and time. The historic-period Plains earthlodge is a distinctive dwelling type, but it is not totally unique in North America, nor is it a late invention. We therefore also consider how these lodges compare and contrast with structures that also have been called earthlodges in other culture areas of North America, particularly the Southeast, and we compare and contrast the historic lodges of the Plains with their late prehistoric predecessors in the region.


It is likely that European, particularly French, traders were in earthlodge villages as early as sometime in the seventeenth century, and it is a virtual certainty that men such as Bourgmont, who ascended the lower Missouri River in 1714 (Norall 1988), were in earthlodge villages in the early eighteenth century. However, the first European to specifically record observing an earthlodge and to attempt to describe one was Pierre Gaultier de Varennes, the Sieur de La Vérendrye. Entering a Mandan village in what is now central North Dakota in December 1738, La Vérendrye described the village, its fortifications, the people, and their dwellings, saying of the latter that they "are large and spacious, and are divided into apartments by broad planks. Nothing is left lying about, all their belongings being kept in large bags hung from the posts. Their beds are made like tombs, surrounded by hides" (Smith 1980:59)-a brief account and recognizable only if you know what he was describing. Six decades later, at the end of the eighteenth century, in 1798, the British trader David Thompson was in five Mandan and Hidatsa villages, also in present-day central North Dakota. There, he recorded the number of houses and contrasted "houses" with "tents" (Wood 1977; Wood and Thiessen 1985:117-119), obviously recording the presence of earthlodges but not describing them.

In light of the brevity and vagueness of the La Vérendrye description and of Thompson's stopping short of actually describing the "houses" he was counting, our first real recognizable descriptions of earthlodges are those in the journals of the Lewis and Clark expedition. By early October 1804, the Corps of Discovery was in Arikara territory in present-day north-central South Dakota. On the tenth of that month, some members of the party went to one of the villages and several of the men, including John Ordway (Moulton 1983-2001:9:78-79), Joseph Whitehouse (Moulton 1983-2001:11:98), and Patrick Gass (Moulton 1983-2001:10:52-53), described the lodges in their journals. Ordway's and Whitehouse's descriptions are no longer and scarcely more recognizable than La Vérendrye's. That by Gass, a carpenter by trade (Moulton 1983-2001:10:xiv), however, is 191 words long and gives dimensions, mode of construction, and overall plan, painting a verbal picture usable to an unfamiliar reader and justifiably frequently cited. Gass's journal was published in 1807 and was soon accorded the dubious honor of being plagiarized by Charles LeRaye, whose purported account of an 1802 trip to the Arikara, complete with an earthlodge description that resembles Gass's in many ways (Cutler 1971 [1812]:171-172), is now regarded as a forgery that drew on Gass's journal (Wood 1990:86).

Over the next six decades, travelers and sometimes traders would record many more descriptions of earthlodges in the villages of a number of tribes of the Missouri River drainage (Table 1.1), and ethnologists would later add their descriptions (Table 1.2). Some would be shorter than Gass's, others would be longer, with Clayton's 1846 description of Pawnee lodges (Clayton 1973:97-100) and Morgan's 1862 description of Arikara lodges, complete with sketches (Morgan 1965:133-136), being perhaps the longest and most meticulous of them all. Additional information on various aspects of earthlodge construction, life, and meaning is found in various compilations of texts (also listed in Table 1.2).

The earliest drawing, sketch, or painting of an earthlodge we know of is Samuel Seymour's 1819 drawing of a dance inside a Kansa lodge in the Blue Earth village near present-day Manhattan, Kansas (reproduced in Bailey and Young 2001:469; Bushnell 1922:plate 30b; and other sources). The first exterior views apparently were those done in the 1830s by George Catlin (1973 [1844]) and Karl Bodmer (Hunt et al. 1984; Maximilian and Bodmer 1976; Wood et al. 2002), each of whom also painted interior views. One of Bodmer's interior views is particularly well known (Figure 1.2). Many more sketches, drawings, paintings, and, eventually, photographs followed (Matthews 1902 may be the largest single compilation of earthlodge illustrations extant). Therefore, we do not lack for information on how nineteenth-century earthlodges were built or what they looked like.

In general, earthlodges were circular, with an entryway extending from some part of the otherwise continuous wall. Except for a smoke hole at the apex of the roof and sometimes excepting the entryway as well, the conical roof was covered completely with sod or with earth that eventually sprouted vegetation (Figure 1.1). When viewed from the exterior they do indeed appear to have been made of earth, as Catlin (1973 [1844]:32), writing at a Mandan village in 1832, observed: "their lodges ... appear from without, to be built entirely of dirt." And it was Morgan (1965:134) who, noting the earth covering of Arikara lodges, commented that it was "for this reason they have usually been called 'dirt lodges.'" Indeed, tracing usage, we find the term dirt lodge appearing in the accounts possibly as early as 1805 in Larocque's journal of a trip to Yellowstone via the Mandan and Hidatsa villages (Wood and Thiessen 1985:164). The common, if not nearly exclusive, use of this lexeme in many nineteenth-century documents suggests it was common lingo in the region at that time and undoubtedly earlier. Accounts written toward the later part of the century, however, seem to alternate the terms dirt lodge and earth lodge, with the latter becoming the more common term by the early twentieth century. It is, in fact, the term Matthews (1902:2) used for Missouri River valley dwellings in 1902 and is the heading under which Fletcher (1907) described the Plains earthlodge in the Handbook of American Indians North of Mexico a few years later. We should note, though, that the Pawnee have traditionally referred to these structures as mudlodges and that Whitman (1937:1-3 and passim) used that same term when discussing Otoe lodges.

Of course, an earthlodge was not really an earthen structure and is more correctly classed as a timber-frame or post-and-beam structure (Nabokov and Easton 1989:16), as Morgan (1965:133) was aware. It was framed by two concentric sets of upright posts and connecting beams (Figure 1.3). The interior was the taller of the two sets of posts and, according to some traditional accounts, often the first set erected. Four is considered the ideal number of interior posts, and this number does characterize Arikara (Figure 1.4a), Mandan, and Hidatsa lodges (Matthews 1877:5; Will 1930:40; Wilson 1934:356). A Pawnee account, though, states that "in a large mud-lodge there would be ten center poles, but in smaller lodges there would be eight, and in quite small ones, six" (Weltfish 1937:147). The Omaha and probably also the Ponca often used more than four interior posts, depending on the size of the lodge (Dorsey 1896:269; Fletcher and La Flesche 1972:97). Archaeological evidence from excavated historic Pawnee (Figure 1.4b) and Omaha lodges bears this out and also shows that lodge diameter and the number of interior posts are strongly correlated (O'Shea and Ludwickson 1992:75-78). This must have been a late eighteenth- or early nineteenth-century development, though, for protohistoric-period (Lower Loup phase, ca. seventeenth-eighteenth century) Pawnee lodges were still of four-interior-post construction (Dunlevy 1936:160-169). Interior posts were one to two feet in diameter and, according to Wilson (1934:357), placed in pits about 27 inches deep and the diameter of the post. Wood species might have affected post size, with larger posts needed if cottonwood as opposed to a stronger wood was used (Will 1930:42; see Roper, this volume, for more on this point). After the interior posts were erected, beams were laid across the top to connect them and form a ring (actually a polygon, with the number of sides depending on the number of interior posts). The height of these posts above the floor varied. Wilson (1934:357) suggested about 11 and a half feet; Fletcher and La Flesche (1972:97) reported 10 feet, as did Matthews (1877:5); and Gass (Moulton 1983-2001:10:53) said they were 15 feet tall. Some of this variation undoubtedly is the difference between estimates and actual measurements, some is the difference between an ideal and its actualization in a given case, and some is probably due to actual variability among lodges.

The wall posts were placed next, in a ring varying in diameter from 20 feet or so to as much as 60 feet, with most in the roughly 30- to 40-foot range. These posts were shorter and somewhat smaller in diameter than the interior support posts. Wilson (1934:363) indicated that, in a Hidatsa lodge, the wall posts should be about the height to which a woman could reach with her hand and arm outstretched above her head, which would make them about six and a half to seven feet. Fletcher and La Flesche (1972:97) said the walls were about eight feet high, while both Matthews (1877:5) and Gass (Moulton 1983-2001:10:53) suggested wall posts were about five or six feet high. The number of posts varied. Most accounts, when they comment on the matter, suggest about a dozen to a dozen and a half wall posts, and data from excavated lodges tend to bear this out. Floor plans from excavated lodges, however, show that some larger lodges had 20 or even up to around 30 wall posts. As with the interior support posts, beams were laid across the top of or in the forks of the wall posts to form a ring.

Roof rafters and wall leaners were laid above the frame (Figure 1.5). Rafters were laid in a radial pattern with the upper ends on the interior post beams and the lower ends on the wall post beams. The roof typically was pitched, thus producing the conical or dome-shaped appearance of the structure. The rafters, however, did not meet at a point. Rather, an open square was left at the roof apex to serve as a smoke hole. This smoke hole lay directly above a hearth in the center of the lodge floor. Logs were later added around the edges of the smoke hole and, in the day-to-day use of the lodge, it may have been covered by a frame to prevent children and puppies from falling through the hole or by a bullboat to prevent heavy precipitation from entering (Wilson 1934:368-369). The upper ends of the wall leaners were laid on the wall post beams, where they were interspersed among the lower ends of the roof rafters. The leaners were laid at an angle, with their lower ends placed in the ground several feet outward from the wall posts. The space between the wall posts and the lower ends of the leaners is sometimes called by its Hidatsa name, atúti (Wilson 1934:363; see also Matthews 1877:6).

At least two variations on this basic form are recognized. One of them, described for the Hidatsa (Wilson 1934:364-365), is a flat-roofed structure. In this variation, roof rafters reached only to the interior support beams. Cross-stringers then were laid flat across the beams and slabs or planks were laid above these to form a flat roof (see Bowers 1965:plate 2 for a photograph of one such lodge and Wilson 1934:360 for a drawing of construction details). In the other variation, two concentric rings of wall posts of equal height were erected and each was connected by beams. Short rafters were laid flat across the two rings of beams. The finished lodge thus had a stepped appearance. The 1872 W. H. Jackson photographs taken in the Pawnee village (25NC6) near Genoa, Nebraska, show lodges with this form of construction (see Bushnell 1922:plates 49-50; Hyde 1974:35; and other sources for reproductions of these photographs). In addition, the floor of the early nineteenth-century lodge over which the museum at the Pawnee Indian Village (or Kansas Monument site, 14RP1) in Republic County, Kansas, is built indicates that this lodge took that form, which appears to be a late variation of limited geographic distribution.

The wood shell formed by the rafters and leaners was next covered with a layer of willow withes placed parallel to one another and perpendicular to the wood posts. A few accounts suggest these were bundled to form mats (Abel 1939:147; Benson 1988:88; Catlin 1973 [1844]:82), but most accounts do not. In any event, the layer of willows completely covered the lodge, except for the smoke hole and entryway openings. Above it was a matting of dry grass, characteristically either bluestem or slough grass (Carlton 1983:66; Gilmore 1931:69, 1977:14, 16; Howard 1995:56; Will 1930:44; Wilson 1934:36). Wilson (1934:365) suggested that this layer was about six inches thick and Dorsey (1896:270) suggested it was about a foot thick, but, of course, it would be compressed by the earth covering above it.

The earth covering was the outermost layer of the earthlodge and was what gave these structures their misleading name. Earth may have been laid on either as blocks of sod (Dunbar 1880:274; Fletcher and La Flesche 1972:9) or as loose earth (Brackenridge 1904:115; Gilmore 1931:57; Will 1930:44). Wilson (1934:366) said the Hidatsa used both, with the sod or turf method preferred. The thickness of the earth covering is variously reported as "four to six inches" (Will 1930:44), "five to six inches" (Abel 1939:147), "nine inches" (Dunbar 1880:274), "about 12 inches" (Dunbar 1918:600), "one or two feet" (Coues 1987:534; Dorsey 1896:270), "two or three feet" (Catlin 1973 [1844]:82), "nearly three feet thick at the bottom and eighteen inches, or so, at the top" (Carlton 1983:66), "thick" (Fletcher and La Flesche 1972:98), "deep" (Morgan 1965:135), or simply "to a sufficient thickness to prevent the rain and snow water from washing it away" (Gilmore 1931:57). As Scullin and Roper discuss in later chapters, it is the lower numbers that are the more credible because of the great weight of soil and the stresses even a few inches of earth place on the lodge's wood frame.


Excerpted from PLAINS EARTHLODGES Copyright © 2005 by The University of Alabama Press . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Table of Contents

A Note on Plains Village Taxonomy and Chronology
1. What, Where, and When Is an Earthlodge? Roper Donna C. Pauls Elizabeth P.
2. Confounding Stereotypes: Building an Earthlodge for Fun and Edification Scullin Michael
3. Architecture as a Source of Cultural Conservation: Gendered Social, Economic, and Ritual Practices Associated with Hidatsa Earthlodges Pauls Elizabeth P.
4. From Earthlodge to Medicine Lodge? Probable Cheyenne Origins of the Sun Dance Liberty Margot P
5. Middle Ceramic Period Earthlodges as the Products of Craft Traditions Blakeslee Donald J.
6. Earthlodge Dynamics 101: Construction and Deterioration Issues and Their Lessons for Archaeologists Roper Donna C.
7. This Old Earthlodge Village: How Long Were Sites of the Middle Missouri Tradition Occupied? Lensink Stephen C.
8. Geophysical Signatures of Earthlodges in the Dakotas Bales Jennifer R. Kvamme Kenneth L.
9. Future Directions for Earthlodge Research Pauls Elizabeth P. Roper Donna C.
References Cited

Customer Reviews

Most Helpful Customer Reviews

See All Customer Reviews