A New Deal for Southeastern Archaeology

A New Deal for Southeastern Archaeology

Edwin A. Lyon

The University of Alabama Press

Tuscaloosa

Copyright © 1996

The University of Alabama Press

Tuscaloosa, Alabama 35487-0380

All rights reserved

Manufactured in the United States of America

The paper on which this book is printed meets the minimum requirements of American National Standard for Information Science–Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984.

Library of Congress Cataloging-in-Publication Data

Lyon, Edwin A.

A new deal for southeastern archaeology / Edwin A. Lyon.

    p. cm.

Includes bibliographical references and index.

ISBN 978-0-8173-0791-2 (alk. paper)

ISBN 978-0-8173-8381-7 (electronic)

1. Archaeology—Southern States—History—20th century. 2. Archaeology and state—Southern States—History—20th century. 3. Indians of North America—Southern States—Antiquities. 4. Southern States—Antiquities. I. Title.

CC101.U6L96    1996

975'.01—dc20

95-11101

CIP

To my mother and the memory of my father

Contents

Illustrations

Preface

Prologue

1. Southeastern Archaeology before the Depression

2. The Origin of New Deal Archaeology

3. Archaeology in the 1930s

4. WPA Archaeology

5. TVA Archaeology

6. National Park Service Archaeology

7. The Legacy of New Deal Archaeology

Epilogue: New Deal Archaeology Today

Notes

Bibliography

Index

Illustrations

Maps

1. Southeastern archaeology sites before the New Deal

2. Federal archaeology sites, 1933–1934

3. WPA archaeology sites

4. TVA archaeology sites

5. National Park Service archaeology sites

Figures

1. Mound 4 at Marksville, Louisiana, 1993

2. Mound 4 at Marksville, 1933

3. Excavation of semi-subterranean house at Marksville, 1933

4. Mound C, Ocmulgee

5. Norris Basin house pattern

6. Webb and Lewis in TVA laboratory, 1934

7. Chartres street laboratory in New Orleans

8. Greenhouse site

9. Excavation at Greenhouse

10. Excavation at Greenhouse

11. Dog burial at Indian Knoll

12. Robbins mound

13. Adena town house at Crigler Mound

14. Excavation of a Pickwick Basin shell midden

15. Later excavation of a Pickwick Basin shell midden

16. Pickwick Complicated Stamped Pottery

17. Cataloging at Ocmulgee

18. Macon Trading Post

19. 1939 Southeastern Archaeological Conference meeting at Ocmulgee

20. Moundville excavation in 1992

Preface

THIS BOOK IS a history of New Deal archaeology in the American Southeast. The depression of the 1930s created a unique opportunity for archaeologists, with consequences still felt today. Archaeologists often use the term WPA archaeology to describe this period in the history of American archaeology. Unfortunately this usage obscures the true nature of federal archaeology in the depression. New Deal archaeology was a complex system of interrelated projects of the Federal Emergency Relief Administration (FERA), Civil Works Administration (CWA), Works Progress Administration (WPA), Tennessee Valley Authority (TVA), National Park Service (NPS), and Smithsonian Institution working with museums and universities in a number of states. Despite efforts at coordination of these projects, New Deal archaeology never was effectively coordinated and managed as a national program.

The focus of this book is on the Southeast. Major New Deal archaeological projects were active in other sections of the country including the Southwest, the Great Plains, Rocky Mountain states, Midwest, and Northeast. But New Deal archaeology as a whole is difficult to understand because North American archaeology varies regionally. As the historian of anthropology Curtis Hinsley has noted about archaeology in the United States, North American work, for complex historical reasons, has deep local and regional roots, and different parts of the country have come to archaeological attention or prominence at successive stages of national political-economic growth and of professional growth of archaeology. Hinsley is aware of the problems of a regional approach in the history of archaeology but nevertheless recognizes that geographical locus has always been a critical factor in archaeology, and it is equally so in writing the history of archaeology.¹ This is certainly true in the case of the Southeast. As Louisiana archaeologist Jon Gibson has pointed out, Southern archaeology has always been slightly out of kilter with American archaeological development in general.²

The Southeast began to be treated as an archaeological unit before the 1930s, but during that decade it became a real focus of archaeological interest, culminating in the creation of the Southeastern Archaeological Conference (SEAC) in 1938. This concentration on the Southeast continues and was very evident to me during the last SEAC meeting I attended in the fall of 1992. But we should realize that there has never been a consensus about the boundaries of the Southeast, and the Southeast discussed in this book is certainly open to question. My Southeast includes New Deal archaeological projects in Florida, Georgia, North Carolina, Alabama, Tennessee, Kentucky, Louisiana, Mississippi, Arkansas, and east Texas. My inclusion of border areas in this study is designed to make the book as useful to contemporary archaeologists as possible.

This book has been formed out of my unique experience over more years than I care to think about. I never intended to become an archaeologist, and the archaeologists in my office would tell you that I have been successful. Experience digging in Louisiana heat as an undergraduate and working in the LSU museum washing pottery, typing site records, and repackaging part of the WPA collection from its original shoe boxes into more modern containers convinced me that archaeology was not for me. My experience was similar to that of the ethnologist John R. Swan-ton, who in the 1890s did archaeology long enough to enjoy the sound of noon whistles and appreciate the taste of cold spring water and then went into ethnology.³ I also went into ethnology for my M.A., and then history, but I have never been able to escape archaeology.

I became interested in the subject of this study some years ago listening to Bill Haag's stories about his experience in CWA, TVA, and WPA archaeology in Tennessee, Alabama, and Kentucky. I began to work formally on the subject when I wrote a paper on Louisiana WPA archaeology for Burl Noggle's seminar at LSU on American history of the 1930s. My dissertation, completed in 1982, was an effort of a historian with a background in anthropology and archaeology but suffering from many deficiencies. I am grateful to the historians on my dissertation committee for supporting my work in an area foreign to them. Burl Noggle served as my major professor, and John Loos, Robert Becker, and David Lindenfeld were on my committee. Only Haag understood the archaeology, but he did not get to read it until much later. It was an ideal committee.

Late in 1985 my job as historian for the U.S. Army Corps of Engineers in the New Orleans District was abolished, and I took a position as an archaeologist working in cultural resources management (CRM) and historic preservation. This experience has been crucial in creating the book in this form. While always working toward my goal never to get mud on my boots, I now know firsthand the difficulties of survey, the expense of data recovery, and the problems of report preparation. I have struggled with a bureaucracy not very different from the federal agencies of the depression. This experience has enabled me to understand more clearly the problems of archaeologists in the 1930s. I now understand what William Webb, an important WPA and TVA archaeologist, meant when he wrote about New Deal archaeology: Regulations, constraints, limitations, difficulties innumerable all conspired to make this work what it was. It was never possible to do what was best to do at the most propitious time or in a way most satisfactory to science. It was always the case of working in a hurry, under adverse conditions, in the face of many limitations and restrictions.⁴

My daily contacts with federal, contract, and academic archaeologists and involvement in a number of archaeological projects has made clear to me that New Deal archaeology in the Southeast was a major formative experience in the development of professional archaeology in the post-World War II period. At the same time I also understand how different the archaeology of the depression was from what we do now. Archaeologists now devote much more attention to minute examinations of more limited areas of smaller sites than did New Deal archaeologists. It is still difficult for me to grasp the size and scale of some of the relief, salvage, and preservation projects of the depression. The large numbers of laborers available at some of the major New Deal sites allowed much more extensive excavation than would be possible today in our CRM data recovery projects. As a result major New Deal excavations were vastly larger than many of our contemporary projects. At Hiwassee Island, for example, a salvage project in the Chickamauga Basin in eastern Tennessee, excavation of a village and substructure mounds uncovered an area of more than 33,000 square feet. In addition, small midden areas and conoidal burial mounds were excavated. Entire mounds were completely excavated at many sites. At the Wright Mounds in Kentucky forty men removed more than 13,000 cubic yards of earth in nineteen months. Trenching is another impressive component of New Deal projects. Huge trenches were run for incredible distances. At the Greenhouse site in Louisiana archaeologists excavated a 5-foot-wide trench in 3-inch levels for 680 feet. It proved so successful that they dug four other trenches through the site.

I hope that both archaeologists and historians will read this book. The archaeologist will approach the book in a very different way than the historian. Archaeologists use the data produced by New Deal archaeologists, they have heard stories told by the archaeologists, and they have formed definite opinions about New Deal archaeology. They will learn about previously buried archaeological projects and the overall structure and context of New Deal archaeology. For historians the book may be useful in another way. A number of studies of the WPA arts program have made historians aware of the art, music, theatre, writers, and historical records surveys. But these projects are not completely representative of the great variety of WPA projects. Archaeology was not only organized very differently from other WPA projects but was unique because it involved the National Park Service, Tennessee Valley Authority, and the Smithsonian Institution.⁵

This book would not have been possible without the help of many archaeologists, historians, and archivists. During my years as a federal archaeologist I have learned a great deal from a number of my colleagues. Michael Stout, in particular, helped me learn enough to survive in CRM. Over the many years I have worked on this book I have been assisted by archivists, librarians, and archaeologists serving as custodians of the documentation of New Deal archaeology in the Southeast. I am especially grateful to James Glenn of the National Anthropological Archives and his colleagues. A number of archaeologists and historians have read chapters of this book, among them James B. Griffin, Gordon Willey, Lynne Sullivan, Mary Lucas Powell, and Mark Barnes. Edwin Bearss and Barry Mackintosh, chief historian and bureau historian of the National Park Service, respectively, read the discussion of the NPS. Bill Haag and an anonymous reviewer for the University of Alabama Press read the entire manuscript and offered many useful suggestions. Jefferson Chapman, Sylvia Flowers, William Haag, John Hall, Vernon J. Knight, Robert Neuman, Joan Exnicios, and Mary Lucas Powell helped with the illustrations. Judith Knight pushed me for many years to complete the manuscript and provided more help than an author has any right to expect from an editor. Anders Thompson copyedited the manuscript. All readers should be as grateful to him as I am. I thank all for their help.

Prologue

IN JANUARY OF 1993 I visited the Marksville site in Louisiana. Today the site is a state park, the Marksville State Commemorative Area, with a museum that opened in 1950. As I walked through the park trying to understand the mounds, the embankment around the site, and its relationship to Old River at the rear of the park, I thought about the site in 1933. Excavation at this site was the beginning of New Deal archaeology sixty years ago.

In 1933 Marksville was very different. The Smithsonian Institution had been interested in archaeology in Louisiana for some time. Edward F. Neild, an architect in Shreveport, corresponded with Smithsonian archaeologists about Hopewell sites in Louisiana. He had found Hopewell type sherds at Moncla Ferry on the Red River near Marksville. Alexander Wetmore of the Smithsonian planned for Frank Setzler to visit sites in Louisiana after a Texas trip when Neild was to show him Hopewell sites in north Louisiana.¹ Both Setzler and I, Neil Judd wrote, are tremendously interested in this Hopewell influence in the South. It may be that in your vicinity we shall yet find the information to solve the problem of this unknown, but brilliant, people whose remains in Ohio have prompted so many unanswerable questions.² Setzler later visited Marksville and other sites in the spring.

As local amateurs began to be interested in restoration of the mounds, Setzler and Judd began to worry that the site would be destroyed by its restorers. Judd recommended to Neild that restoration should follow careful examination of what now remains and in no case should it be left to the imagination of one unfamiliar with Indian mounds and especially those at Marksville.³

The town of Marksville purchased the site and planned to convert it into a park and recreation center using Federal Emergency Relief Administration funds. Work had started on a swimming pool before local people interested in archaeology persuaded the authorities in charge of the project to allow excavation and restoration under the direction of the Smithsonian.⁴ The city council and the local FER A then requested that the Smithsonian send a representative to supervise excavation and restoration of the site.⁵ Frank M. Setzler, assistant curator of archaeology at the United States National Museum of the Smithsonian Institution, arrived in late August of 1933 and remained until November. Setzler had studied at Ohio State University from 1924 to 1927 while he worked as an assistant field director at the Ohio State Museum. He later was a graduate student in anthropology at the University of Chicago and worked as an Indiana state archaeologist.⁶ His assistant, James A. Ford, aided in the excavation while Setzler was at the site and took charge for the month of November after Setzler left. Ford had graduated from high school in Clinton, Mississippi, in 1927 and went to work for the Mississippi Department of Archives and History, where he worked with Moreau B. Chambers digging mounds. According to Gordon Willey, Ford and Chambers spent three summers at this task of officially sponsored 'pothunting,' traveling from site to site by team and wagon.⁷ In 1930 Henry B. Collins, Jr., of the Smithsonian Institution offered Ford the job of field assistant on an Alaskan field trip. Ford returned to Alaska for eighteen months beginning in the summer of 1931. In 1933 he received a grant from the National Research Council for archaeological investigations in Mississippi and Louisiana.⁸

It was a new experience for Setzler and Ford to supervise a crew of more than one hundred laborers in the excavation of three mounds and village areas.⁹ The site is surrounded by an earthwork from three to seven feet high that was probably built for ceremonial purposes. Near the museum building is Mound 4, a conical burial mound 20 feet high dug into by Gerard Fowke of the Smithsonian in 1926. Fowke had only disturbed part of the mound, and Setzler returned to it in 1933.¹⁰ The current mound was reconstructed in 1933 and now is surrounded by a fence to keep off visitors. Setzler and Ford dug into Mound 5, which was about 3 feet high and 40 feet in diameter, but few records survive. Mound 6, a truncated mound about 13 feet high, was the site of extensive digging but the work is documented today only by a few photographs. Setzler and Ford also placed at least five trenches through the village area. In addition to a number of burials in Mound 4 they recovered artifacts including Marksville pottery, pipes, projectile points, and stone knives. A final report on the project was never published—a common occurrence in many of the later New Deal archaeological projects. Few records of the project survive.

Setzler finished the excavation with a new awareness that the Hopewell culture extended into the Southeast. At first, he resisted the heretical idea that a variant of Hopewell existed in the Southeast. According to Henry Collins, it took Setzler's experience in Louisiana to convince him of the importance of the Hopewell-south-eastern relationship.¹¹ Ford already had seen a relation between some Mississippian sherds and Hopewell, and Collins had tried to convince Setzler of the presence of Hopewell in the South.¹² Finally, Setzler admitted that the data obtained give definite proof that the prehistoric Indians who lived and built the mounds on this site were closely allied in their culture-phase to those known as the Hopewell in the northern Mississippi Valley.¹³

This project would demonstrate to skeptical archaeologists that archaeology was possible using large crews of relief laborers. The large Civil Works Administration relief archaeology projects during the winter of 1933-1934 emerged directly from this experience.

1

Southeastern Archaeology before the Depression

PREHISTORIC ARCHAEOLOGY in the United States developed during the twentieth century as one of four components of the discipline of anthropology: ethnology, physical anthropology, linguistics, and archaeology. Within anthropology, ethnology was the dominant force before the 1930s. Archaeology as it was practiced in the early twentieth century was of little value to many anthropologists. A few archaeologists had done good work, but the typical archaeologist, in the words of J. Alden Mason, was a congenital antiquarian, attracted to the ancient, the rare, the spectacular.¹ Franz Boas, the most important figure in American anthropology in the early twentieth century, was aware that archaeology could contribute to anthropology, but was unimpressed with archaeologists. According to Mason, A cynical remark attributed to him, even if apocryphal, probably expresses his attitude: ‘If a man finds a pot, he is an archeologist; if two, a great archeologist; three, a renowned archeologist!’²

Archaeology had little to offer at that time to anthropologists interested in understanding the history of Native Americans. As Alfred Kroeber pointed out, Incredible as it may now seem, by 1915-25 so little time perspective had been achieved in archaeology that Wissler and I, in trying to reconstruct the native American past, could then actually infer more from the distributions and typology of ethnographic data than from the archaeologists' determinations. Our inferences were not too exact, but they were broader than those from excavations.³ Not only lack of archaeological knowledge but the refusal of physical anthropologists to recognize the presence of humans in the New World before the very recent past limited the importance of archaeology in anthropology. Failure to recognize time depth in eastern North America led to a short prehistoric chronology with changes occurring rapidly as the result of movement of population or spread of cultural traits by diffusion.

This domination of archaeology by ethnology benefited archaeology by expanding archaeologists' interests in broader anthropological questions but also limited the development of the field. Boas and his followers opposed any role for cultural evolutionism in anthropology leading to emphasis on cultural relativism and historical particularism. This opposition to cultural evolutionism effectively prevented concern with broader issues of change in Native American cultures.⁴ As Gordon Willey and Jeremy Sabloff point out, The distrust of evolutionary thinking and the marked historical particularism of American anthropology forced the American archaeologist into a niche with a very limited horizon.⁵

Lack of time perspective led to reliance on space rather than time as the primary interpretative approach of ethnologists and archaeologists. As late as 1923 Kroeber noted that one of the major characteristics of the native cultures of the New World was that they have come to us virtually in momentary cross section, flat and without perspective. In general there are few historic data extant about them. Kroeber argued that the history of the Native Americans had to be studied through the medium of space. As soon, Kroeber said, as knowledge of American cultures became orderly, its organization was inevitably effected in terms of geography. This geographical approach established culture areas, a non-philosophical, inductive, mainly unimpeachable organization of phenomena analogous to the ‘natural classification of animals and plants on which systematic biology rests.⁶ Kroeber listed ten culture areas for North America including the Southeast or Southern Woodland.

While in the southwestern United States stratigraphy was developing as a means of understanding culture history, this approach was not transferred to the Southeast. As Willey observed for all of American archaeology, Frequent gross observations were made on superposition in refuse strata, in structures, or in graves; and, on occasion, differences in pottery or other artifacts were correlated with these observations, but, for some reason, this did not seem to lead on to the establishment of local, regional, or areal culture sequences. Willey concluded that the stratigraphic method did not become truly viable in American archaeology until after 1920.⁷

The approach Gerard Fowke used in mounds in Colbert County, Alabama, was typical of many archaeologists working in the Southeast before the depression. The numerous worked objects scattered throughout that portion of the mound which was excavated, and presumably in all other parts of it as well, being merely derelicts, so to speak, not distinctive in material, form, or in any other respect, cast no light upon the identity of the tribe who may have made them or the time at which the users may have left them here. Consequently no necessity exists for entering into particulars regarding the depth or the part of the mound where they were discovered. Only unusual features will be herein recorded; burials, of course, will be somewhat fully described.⁸

NORTHERN SUPPORT FOR SOUTHEASTERN ARCHAEOLOGY

Southeastern archaeology developed slowly before the New Deal. Southern universities were poor and unable to provide support for research and publication in archaeology. The development of northern anthropological museums during the last half of the nineteenth century increased support for southeastern archaeology during what has been called the museum era of American anthropology. Much of southeastern archaeology before the 1930s was supported by non-southeastern museums such as the Smithsonian Institution, the Peabody Museum, the American Museum of Natural History, and the Heye Foundation.⁹ The National Research Council was the one non-museum institution that assisted southeastern archaeologists.

The Peabody Museum of Harvard University, founded in 1866, was an important supporter of early southeastern archaeology. The Peabody first concentrated its collecting in the eastern United States, accumulating large collections from New England, Florida, the Middle Atlantic area, Tennessee, the Ohio Valley, Michigan, and Missouri. Later the museum collected in the Southwest, the West, and Middle America. Jeffries Wyman, an anatomist, was the first curator at the Peabody Museum from 1866 until he died in 1874. He dug in shell middens in New England and then turned to the St. John's area of Florida in 1867, with additional work in 1869, 1871, and 1874. He published preliminary studies followed by his major report in 1875. Wyman showed that shell heaps were the result of human activities rather than natural processes. He analyzed pottery and faunal material, concluding that mounds without pottery were older than those containing pottery and that plain and incised pottery was older than stamped-decorated pottery.¹⁰

In 1875 Frederic W. Putnam became curator of the Peabody Museum. In the 1880s the museum began work in mounds in the eastern United States with Putnam continuing Wyman's collecting procedure of collaboration with field workers. Putnam concentrated on mounds in Ohio, at Madisonville and the Turner group. He was instrumental in the purchase and transfer of the Serpent Mound to the Ohio State Archaeological and Historical Society. In the Southeast he concentrated his attention on Kentucky and Tennessee. Early archaeological study of the Upper Tennessee River was conducted by the Reverend E. O. Dunning for the Peabody Museum of Harvard and the Peabody Museum of Yale, the only involvement of Yale's museum in the Southeast in this period. Dunning's work in Tennessee was supported by Wyman with $300 a year from 1868 to 1871. The Peabody Museum sent Edward Palmer to Texas and northern Mexico in 1879-1880, and his investigations included visits to sites in east Texas. Edwin Curtis worked during the winter of 1879 in the St. Francis River Valley in Arkansas, collecting from a number of sites, including the Rose Mound, Fortune Mound, Stanley Mounds (Parkin), and the Holcomb Mound.¹¹

The Peabody Museum sent Charles Peabody and W. C. Farabee to Coahoma County, Mississippi, where they excavated two mounds, the Dorr Mound and the Edwards Mound, over three months in 1901 and 1902. Peabody was aware of stratification in the Edwards Mound, identifying layers of soil, ashes, and charcoal. He recognized that pottery, bundle burials, and animal bones were more plentiful in an upper level than in a lower level, leading him to conclude that the mound was built in two periods, the more recent post-Columbian.¹²

The Stalling's Island site near Augusta, Georgia, was excavated early in 1929 by C. B. Cosgrove of the Peabody Museum, using about fifteen workers. William H. Claflin, Jr., had dug into the site beginning in 1908 and again during 1920 and prepared a report on the 1929 project. Using more than 3,500 sherds recovered from the site, he identified a long occupation by the Stalling's Island people followed by later visitors to the site. Pottery decoration was similar at all levels. Claflin concluded that the Stalling's Island people lived in the Savannah River Valley before later makers of paddle-stamped pottery.¹³

The American Museum of Natural History in New York also played an important role in early twentieth-century southeastern archaeology. The museum, founded in 1869, opened its new building in 1873. It supported ethnological and archaeological research in a variety of areas in the United States, primarily in the West. One important early project of the museum was in the Southeast, Harlan Smith's investigations at Fox Farm and a survey of the area in the summer of 1895. Smith knew that the site was prehistoric but could not estimate its age. He recognized that the pottery was part of the Ohio Valley group rather than the Mississippi Valley group. He concluded that the Fort Ancient culture was present in Kentucky.¹⁴

In 1905 when Clark Wissler became head of the department of anthropology at the museum he planned a program of archaeological investigations. The emphasis of the museum would be on the Southwest, Mexico, and Central and South America, but Nels C. Nelson would work in the Southeast for the museum. Nelson, born in Denmark in 1875, received a B.L. from the University of California in 1907 and a M.L. in 1908. He was an assistant curator at the Museum of Anthropology at the University of California from 1909 to 1912, assistant curator at the American Museum of Natural History from 1912 to 1920, associate curator from 1921 to 1927, and became curator of prehistoric archaeology in 1928. In addition to his training in the United States, Nelson studied European archaeological methods in France.¹⁵

Nelson's most important work in the Southeast was in Kentucky and Florida. In May and November of 1916, Nelson investigated Mammoth Cave in Kentucky. He argued that cave archaeology should be as valuable in the United States as it had been in Europe. Caves were more useful in working out chronology than mounds or village sites because nowhere else can we be quite so sure of the validity of stratigraphic results as in cave floor deposits. He saw shell heaps as almost as useful for coastal areas and pointed out that shell heaps have not yet been adequately investigated.¹⁶

E. H. Sellards, Florida's state geologist, informed the museum in April 1917 that a large shell mound at Oak Hill, Florida, was being destroyed for road construction material. When Nelson arrived in Florida for a few days work, he found one-seventh of the mound remaining after two steam shovels had dug out nearly two thousand carloads of shell over four months. No pottery was found in the bottom layer of the midden, plain sherds in the middle, and checker-stamp decorated ware in the upper level of the site. Nelson noted that check-stamped ware probably originated near the Indian River, where it was found deposited above undecorated sherds.¹⁷

Another northern sponsor of archaeology in the Southeast was the Museum of the American Indian. George G. Heye began collecting artifacts in the 1890s. In 1903 Heye acquired his first important collection from New Mexico. Collecting expeditions followed in Arizona, Central America, and the Caribbean. In 1916 Heye founded the Museum of the American Indian, Heye Foundation. In addition to collecting, Heye also dug sites including several mounds in Haywood County, North Carolina, in the spring of 1915. The Nacoochee Mound in Georgia, a domiciliary mound 17 feet high, was partially excavated by Heye in the summer of 1915 in cooperation with the Bureau of American Ethnology. The excavation revealed the stratigraphy of the mound, including layers of red-burned earth, dark clay, and 6 feet of clay with specks of charcoal. Although the excavators did not consider the stratification well defined, it indicated to them that the mound had been built in several periods. They thought it was a typical Cherokee earthwork occupied in both prehistoric and historic times.¹⁸

The Museum of the American Indian became interested in Arkansas after seeing the results of Clarence B. Moore's work. Moore recommended work in the Red River Valley, and Mark R. Harrington began work for the museum in the area in February 1915. Harrington had received his B.S. from Columbia University in 1907 and his A.M. in 1908. He was an assistant anthropologist at the American Museum of Natural History from 1899 to 1902. He worked as a field ethnologist for the Museum of the American Indian from 1908 to 1910 and as an assistant curator in the American section of the University of Pennsylvania Museum from 1911 to 1914. He returned to work for the Museum of the American Indian as archaeologist and ethnologist from 1915 to 1917 and again from 1919 to 1928. Flooding in the Red River Valley drove Harrington out of the valley and into other areas of southwestern Arkansas where he studied approximately twenty sites over two years. In 1922 and 1923 Harrington worked in caves in the northwestern part of the state. The caves were filled with dry dust, forcing the archaeologists to wear a respirator, a kind of gas mask.¹⁹ From August to December 1919, Harrington searched for sites along the Tennessee River between Nashville and Chattanooga but concentrated his attention on the area between the Little Tennessee and Hiwassee rivers. Harrington found the remains of the Round Grave culture at the bottom of each site. Above this level he found a Cherokee-like culture that built most of the burial mounds. Next came the remains of the Cherokee. James B. Stoltman, in his review of the history of southeastern archaeology, viewed Harrington's work as important in the history of southeastern archaeology because he was a leader in the rediscovery of the scientific value of village refuse excavation.²⁰

The Phillips Academy in Andover, Massachusetts, also supported southeastern archaeology even before the establishment of its Department of Anthropology in 1906. Beginning in 1896 Warren K. Moorehead purchased collections for Robert S. Peabody. In 1897 he collected in New Mexico and later dug in the Southwest. Later he employed collectors who dug in Ohio and collected in Tennessee, Arkansas, and other areas. In January 1903 Moorehead and a crew of less than ten men opened more than one hundred graves at Hopkinsville, Kentucky. The laborers dug between twenty and thirty holes daily and examined quite thoroughly about two acres. Moorehead and his crew used a barge to visit sites in Tennessee.²¹

In 1915 Moorehead decided to survey the Arkansas River Valley. He traveled upstream, finding mounds of the valley to be similar to the pottery belt of the middle Mississippi Valley. During winters from 1925 to 1927 Moorehead worked at the Etowah site in Georgia concentrating on Mound C and the village. In the winter of 1924 Moorehead worked at Natchez, Mississippi, because he wanted to compare the Hopewell with the description of the Natchez that John R. Swanton had provided in his 1911 study, Indian Tribes of the Lower Mississippi Valley. Moorehead did preliminary work at a number of sites including the Emerald Mound, the Anna group, and the Ferguson mounds. Despite many large pits and auger tests in a number of mounds, he was disappointed to find no evidence of a high culture in the area.²²

The Smithsonian Institution provided the only federal funds for archaeology in the Southeast, but its resources were limited. Early support for archaeology by the Smithsonian concentrated on the Southwest. But the Smithsonian's role in North American archaeology was more than excavation; it defined important archaeological questions and definitively answered a number of them. The Smithsonian played an important role in two major controversies important in the development of southeastern archaeology: the debate over early humans in the New World and the mound builders.

Before 1890 some writers on the prehistory of North America believed that humans had lived in America in glacial times. In the 1870s Charles Abbott, with support from Putnam of the Peabody Museum, dug at his farm in New Jersey, finding artifacts that looked similar to European paleolithic tools. Other early sites were found in a number of locations in the 1880s based on the criteria of similarity of appearance with artifacts in Europe. This early date for peopling the New World was subject to significant attack in the 1890s.²³ William Henry Holmes (1846-1933) was the leader of the first wave of the attack. He had worked as an artist for the U.S. Geological Survey and became curator of aboriginal ceramics in the U.S. National Museum of the Smithsonian Institution in 1882. In 1889 he moved to the Bureau of American Ethnology and from 1902 to 1909 was chief of the bureau. Holmes studied artifacts thought to be remains of early human beings in the New World and demonstrated that these paleoliths dated to a later period.²⁴

ka's method of discrediting early human occupation in North America was what T. D. Stewart called morphological dating: under this concept the claims to antiquity of any human skeletal remains could be dismissed simply by showing that they resembled living types.²⁵ The significance for archaeology of a late arrival of people in the New World was a Native American past with little time depth. Historians of archaeology have called this a flat past. Native American history had to be squeezed into a short period of development in the New World. Little time was available for development of the great variability of Native Americans. As George Stocking pointed out, this short time period and underdevelopment of archaeology led to a situation where ethnology perhaps seemed a more likely approach to the history of man in the Americas than might otherwise have been the case.²⁶ This is another indication of the low value placed on archaeology and stratigraphy early in the twentieth century.

A second contribution of the Smithsonian to southeastern archaeology was the study of the mound builders. Antiquarians who studied eastern United States prehistory were fascinated with the many large mounds scattered across the country. With little data available, antiquarians found opportunities for much fanciful speculation about the builders of the mounds. Incapable of recognizing that Native Americans had built the mounds, a lost race of mound builders was given credit. For some, the mounds were built by Danes, Vikings, Welsh, Toltecs, or the Lost Tribes of Israel; some even thought Native Americans constructed the mounds. Important early descriptive work resulted from this preoccupation, including Ancient Monuments of the Mississippi Valley by Ephraim G. Squier and Edwin H. Davis, which in 1848 was the Smithsonian Institution's first publication. Squier and Davis's surveys of many mounds and digging in a few resulted in an important compilation of data that convinced them that the mounds had been built by a race of Moundbuilders.²⁷ To answer finally the question of the origin of the eastern mounds, the Bureau of Ethnology of the Smithsonian Institution organized a mound survey. This project had its origins in a congressional appropriation to the Bureau of Ethnology for studies of the mound builder problem. In 1882 Cyrus Thomas (1825-1910) was appointed chief of the Division of Mound Exploration. Under his direction the survey worked in a number of southeastern states including Arkansas, Louisiana, Mississippi, Tennessee, Kentucky, Alabama, Georgia, South Carolina, Florida, and North Carolina, in addition to other eastern states. The survey investigated more than two thousand sites, including not only mounds but refuse heaps, cemeteries, camps, house sites, quarries, caves, and petroglyphs. Thomas concluded in 1894 in the influential report of the survey, Report on the Mound Explorations of the Bureau of Ethnology, that the mounds were built by Native Americans.²⁸

Holmes, in addition to his role in establishing a flat past for Native Americans, was influential in the development of southeastern archaeology in the late nineteenth and early twentieth centuries. His studies of ceramics and publication of general interpretations of North American prehistory were important to southeastern archaeologists. In the 1880s he investigated the pottery of the Mississippi Valley. In 1889, when he moved to the Bureau of American Ethnology, he studied ceramics recovered during the mound survey. He also analyzed pottery found by C. B. Moore in Florida. This experience enabled him to publish his study Aboriginal Pottery of the Eastern United States in 1903. According to Stoltman, Holmes divided the Southeast into six ceramic provinces: Middle Mississippi, Lower Mississippi, Gulf Coast, Florida Peninsula, South Appalachian, and Middle Atlantic.²⁹ His approach was geographic, and he was able to say little about chronology.

Holmes was also concerned with broader questions in archaeology than many of his contemporaries. In 1919 he published his Handbook of Aboriginal American Antiquities in which he argued that many problems of archaeology remained to be solved, but he emphasized substantial progress: The deep mystery which a short time ago enshrouded some of the greater problems is now dispelled and visions of mysterious races and lost civilizations haunt the minds of those only who have failed to keep in touch with the progress of archeological research throughout America.³⁰

Holmes synthesized his work in the study Areas of American Culture Characterization Tentatively Outlined as an Aid in the Study of Antiquities, published in 1914. He recommended that the archaeologist study contemporary tribes and their historical development. He recognized that Native American cultures were very diverse but that similarities existed within geographical areas and that the environment influenced human history. Holmes defined eleven areas north of Mexico based on archaeological rather than ethnological criteria: North Atlantic, Georgia-Florida, Middle and Lower Mississippi Valley, Upper Mississippi and Great Lakes Region, Plains and Rocky Mountains, Arid Region, California, Columbia-Fraser, Northwest Coast, Arctic Coastal, and Great Northern-Central. He based his archaeological classification partially on the large natural geographical divisions used in museums, but he found classification by "ethnic areas, or areas of culture