A21_MISC_(Attachment_3_-_EXHIBIT_C_Van_Winkle_Dissertation).pdf

Copyright

By

Jamie Chad Brandon

The Dissertation Committee for Jamie Chad Brandon certifies that this is the approved version of the following dissertation

Van Winkle’s Mill:

Mountain Modernity, Cultural Memory and Historical Archaeology in the Arkansas Ozarks

Committee:

Maria Franklin, Supervisor

Samuel M. Wilson

John Hartigan, Jr.

Richard R. Flores

George Sabo, III by Jamie Chad Brandon, B.A., M.A.

Dissertation Presented to the Faculty of the Graduate School of the University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

The University of Texas at Austin December, 2004

UMI Number: 3150548

3150548

Copyright 2004 by Brandon, Jamie Chad

UMI Microform Copyright

All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road

P.O. Box 1346 Ann Arbor, MI 48106-1346

All rights reserved.

by ProQuest Information and Learning Company.

For T. J.

Acknowledgements

As with any research such as this there a number of people to thank. I would, of course, like to thank my committee—Drs. Maria Franklin, Samuel Wilson, John

Hartigan, Richard Flores and George Sabo—for their support and tolerance. They have, in varying ways and in varying aspects, helped and encouraged me to complete this work.

Maria Franklin and Sam Wilson, in particular, have supported me and my work throughout my career at the University of Texas. I could not have asked for better mentors.

Thanks are also due to George Sabo for allowing his research station’s resources to be used during the project, encouraging students from his historical archeology class to participate in the excavations, sponsoring the 2001 Van Winkle archaeological field school, and his never-failing support of my work and my professional development.

I would like to the institutions and agencies that funded and supported the fieldwork. A portion of these investigations and this dissertation were made possible by a cost-share agreement between the Arkansas Archeological Survey, the Little Rock

District of the United States Army Corps of Engineers and the Arkansas Department of

Parks and Tourism. A great debt is owed certain individuals who negotiated this agreement and navigated the many pitfalls that occur any time multiple state and federal agencies engage in a joint endeavor. Christopher Davies, Dr. Thomas J. Green, Steve

Chryrchel, Mark Clippinger, Dr. George Sabo, Dr. Robert C. Mainfort, Sean Harper and many others were all instrumental to the process.

v

Thanks to the Arkansas Archeological Survey (AAS) for the years that they have encouraged and supported me in my research. Dr. Robert Mainfort, Randall Guendling, Mary Kwas, Kathy Cande and many others have offered sage advice and support. I would like to acknowledge Jami Lockhart for conducting extensive geophysical explorations at Van Winkle’s Mill on multiple occasions and Jared Pebworth and Mike

Evans for their efforts at mapping and assisting with the archeo-geophysical investigations. Thanks also go to Dr. Patrick Martin of Michigan Technological

University for participating in the Van Winkle’s Mill cost-share project and offering his insightful advice on industrial archaeology.

A large debt is owed to Jerry Hilliard, the station assistant at the UAF Station of the

AAS, who elsewhere I have called “the guiding spirit” behind the Van Winkle’s Mill project. Jerry not only introduced me and James Davidson to the Van Winkle’s Mill archaeological project in 1997, he has always supported and actively aided the archaeological investigations. He has also keep James and I busy on a number of other

“side projects” during our stints at the University of Arkansas and the University of

Texas. He has also, incidentally, always been a true friend.

Thanks to all those students, AAS employees and other volunteers who helped out during the various excavations 1997-2003: Ben Adams, Jenny Bales, Lorna Beard, Brynn Berry, Michelle Berg Vogel, Carrie Berryman, Robin Bowers, David Bowman, Mary Brennan, Queszarrah Bunch, Minnie Burford, William Chesser, Pritam

Chowdhury, Edma Delgado, Robert Depper, Jason DeWitt, Clark Donat, Lela Donat, Carl Drexler, Donna Edgemon, Gregory Edgemon, Randall Guendling, Stacey Gustafson, vi

David Humphrey, Cara Huwieler, Marilyn Johnson, Molly Kerr, Roula Khawam, Marian

Kunetka, Melody Kinnard, Don Lee, Sam Littrell, Christy Longlois, Doyle Loughren, Shannon McElmurry, Robin McMullen, Amanda Melton, Kevin Miller, Pathe Miller, Michael “Buffalo” Morgan, Kim Newman, Michael O'Connell, Lydia Rees, Amy

Reynolds, Jennifer Richardson, Allyssa Riley, Suika Rivett, Barbara Scott, George Stair, Heide Stair, Mark Still, Melissa Stroud, Will Taggert, Katherine Taylor, Maria Tavaszi, Edward Tennant, Chris Torantali, Alicia Underdown, Greg Vogel, Leslie Walker, Susan

Wheatley and Deborah Weddle. My apologies to those who I have forgotten to include here.

Thanks to my colleagues at the University of Texas at Austin: Kerri Barile, Whitney Battle, Peggy Brunache, Paula Sanders, Sean Maroney, John Schafer, Rissa

Trachman, Mary Jo Galindo, Mindy Boninie, Chet Walker, Clay Schultz, Nesta

Anderson, Nusrat Chowdhury, Jemima Pierre, Leighton Peterson, Keisha-Khan Perry, Scott Webel, Carmen “Apen” Ruiz, Dan Sharp, Galeet Dardashti, Ronda Brulotte, Brandt

Peterson and many, many more who helped me, challenged me, encouraged me, critiqued me, and had more than a few beers with me. Thanks also to my colleagues (and some of my students) at the University of Arkansas: Greg Vogel, Michelle Berg Vogel, Edward

Tennant, Mary Brennan, Leslie Walker, Brynn Berry, Robin Bowers, Bryan Renfro, Maria Tavaszi, Alicia Valentino, Carl Drexler, and many others.

A debt is also owed to the authors of the two works which inspired my approach:

Remembering the Alamo (Richard Flores) and Memories of Revolt (Ted Swedenburg).

Their works have been both informative and inspiring. Additionally, I would like to vii thank the Hicks family (especially Marilyn Larner Hicks) and other Van Winkle descendants who have been also incredibly helpful and provided a wealth of information about the history of their families.

I cannot begin to stress the debt I owe my family—Jerry and Jeanette Brandon, Arch and Mae Brandon, Dorris and Billie Deason, A.J. Brandon, and now Mike Hilton and Davis Brandon See—for their support and tolerance throughout my life. I would have never thought I could accomplish a project such as this if you hadn’t been there, and

I apologize for the long absences caused by my work on this project.

I would definitely be remiss if I did not thank James M. Davidson who has literally been working with me on the Van Winkle’s Mill project since “day one.” It is really as much his project as it is mine. Aside from being one of the best research partners an archaeologist could ask for, he has been one of my best and closest friends. From our meeting in 1996, he has spurred me on to do more and better work. We have worked together on Van Winkle’s Mill, Juliette Street in Dallas, Eagle Mountain Lake, the Peel

House, Cross Hollows and countless other small research projects. I can only hope that our partnership will continue in some form well into the future.

Finally, and most importantly, thanks to T. J. Vestal. There are not words for the support, love and understanding that she has given me throughout graduate school. For all of that, and much more, this dissertation is dedicated to her.

viii

Publication NO.______________

Jamie Chad Brandon, Ph.D.

The University of Texas at Austin, 2004

Supervisor: Maria Franklin

The Arkansas Ozarks hold a particular place in our collective cultural memory. A place that is decidedly rural, anti-modern and white. This dissertation explores how these notions came to dominate our historical consciousness and uses the archaeological excavations carried out at Van Winkle’s Mill (3BE413) between 1997 and 2003 as a platform to challenge and complicate our notions of the history of the Ozarks. Van

Winkle’s Mill was a thriving Northwest Arkansas sawmill community from the 1850s through the first decades of the twentieth century. The mill’s founder, Peter Van Winkle, was a regionally important industrialist who identified with both the modern Victorian ideologies and those of the Confederate South. Enslaved labor made up a portion of the mill’s labor force before the war, and several freedmen families continued to work at Van

Winkle’s Mill following emancipation. Topics covered by this dissertation include an analysis of the role of popular culture in the formation of cultural memory, a landscape analysis of the mill community, an examination of symbolic consumption by the mill’s

African-American workers and a critique of efforts to interpret the mill’s history to popular audiences.

ix

Table of Contents

List of Tables xiii

List of Figures xiv

Chapter 1: Introduction Introduction 1 The Van Winkle Educational Program 6 Synopsis 9

Chapter 2 “A Frontier More South than West”:

A Historical Overview of the Arkansas Ozarks and Van Winkle’s Mill Introduction 12 Arkansas, the Ozarks, and Northwest Arkansas:

An Explanation of Geography and Scale 13 “Who Are These Ozark People?”

Ethnicity and Immigration in the Trans-Upland South 17 "Incongruous, Implausible and Unworthy of Examination":

Slavery in Northwest Arkansas 20 The Civil War in the Arkansas Ozarks:

Ambivalent Rebellion and the Collapse of Ozark Society 31 “Ozarks as Destination”: The Postbellum Ozarks 37 “We All ‘Jes Scattered, Somehow”:

African-American Life in Postbellum Northwest Arkansas 42 Peter Van Winkle and Van Winkle’s Mill 52 Life Histories of the Mills and Laborers in Van Hollow 61 Van Winkle’s Mill into the Twentieth Century 72

Chapter 3 “The Most Backward and Deliberately Unprogressive Region in the United States”: Cultural Memory and the Arkansas Ozarks Introduction 78 Cultural Memory 79 Cultural Memory and the Ozarks 84 Narrative Tropes and Ozarks:

The Frontier, The Civil War and The Hillbilly 86 Memory, Modernity and the Hillbilly 92 The Shepherd of the Hills: Ozark Arcadian Master Narrative 96 Construction of the Hillbilly Trope 109 J. Vance Randolph, Ozark Folklore and Cultural Memory 121 x

“There is No Melting Pot in These Mountains”:

The Ozark Hillbilly and Race 129

Chapter 4 Archaeology, Cultural Memory and Van Winkle’s Mill Introduction 138 Archaeology and the Prehistoric Hillbilly 139 Historical Archaeology and the Arkansas Ozarks 141 The Archaeology of Van Winkle’s Mill 146 Excavations at the Home of Peter Van Winkle and Family 153 3BE652, The Mill’s African-American Cemetery 155 Feature 9: Mill Worker’s Quarters 158 Features 27 and 28: Trash Dumping Areas 163 Feature 29: Vernon T. West’s Portable Mill Location 166 Features 31 & 30: The Blacksmith Shop 167 Feature 32: Livestock Enclosure/Barn 177 Feature 33: Possible Slave Quarters 180 Excavations in the Van Winkle Mill Complex 183 Other Features of Note 190

Chapter 5 “Pines in Place of Magnolias”: The Landscape of Van Winkle’s Mill Introduction 195 Archaeologies of Landscape 198 Peter Van Winkle’s Landscape 199 Van Winkle’s Mill and Settlement Patterns in the Regional Landscape 202 The Plantation Landscape of Van Winkle’s Mill 203 Mapping Cultural Change in Van Hollow 208 African-American Perceptions of the Van Winkle Landscape 216 Cultural Memory and the Landscape of Van Winkle’s Mill 218

Chapter 6 “Independent But Not Isolated?”: Consumerism, Racism and African-American Identity Introduction 225 Modernity, Race, Consumption and Desire 227 Archaeologies of Consumption 231 Antebellum Artifact Assemblage: Feature 33 232 Postbellum Artifact Assemblage: Feature 9 224 Isolated and Independent: African-American Subjectivity, Consumerism and Identity at Van Winkle’s Mill 235

Feature 33: Antebellum Enslavement 237 Feature 9: The Postbellum Marketplace 242

Brick-a-Brac 242 xi

Harmonica Fragments 245 Toys 247

Chapter 7 Engaging Cultural Memory and Presenting the Past to the Public at Van Winkle’s Mill Introduction 251 The Rogers Historical Museum and The Van Winkle Educational Program 253 Student and Teacher Response to the Van Winkle Program 256 A Better, Kinder Slavery?

Student Responses to Industrial Slavery in the Ozarks 260 Evaluation of the Van Winkle Program 266 From the Wythe House to the Van Winkle House:

Critical History and Cultural Memory 268 Cultural Memory, Archaeology and Race in the Ozarks:

Recommendations for the Van Winkle Program & State Park Interpreters 271

Chapter 9 “When With Sighs We Seek to Find Thee”: Conclusions and Summary Introduction 284

Appendix A: Van Winkle Program Student Evaluations on Slavery 289

Bibliography 296

Vita 325 xii

List of Tables

Chapter2

2.1 African-American population in the Arkansas Ozarks, 1840-1970 (After Morgan 1973:62). 44

2.2 Enslaved persons listed in the 1860 US Census slave

schedule for Peter Van Winkle. 62

2.3 Mill related persons and families listed in the 1860 US Census. 65

2.4 Mill related persons and families listed in the 1870 US Census. 66

2.5 Mill related persons and families listed in the 1880 US Census. 67

Chapter 4

4.1 Historical archaeology projects conducted in the Northwest

Arkansas Ozarks. 142

4.2 Cultural features identified in Van Hollow, 1999-2003. 149

4.3 Artifacts recovered from Feature 27. 164

4.4 Artifacts recovered from Feature 28. 165

4.5 Artifacts recovered from Feature 31, general surface collection. 168

4.6 Artifacts recovered from testing at Feature 31 (without nails). 169 xiii

List of Figures

Chapter1

1.1 Remaining stairway to the formal raised garden

in Van Winkle’s Mill, 2002. 3

1.2 Foundations for the flywheel at the site of the

Van Winkle Mill complex, 1997. 3

1.3 Students portraying historical figures at Van Winkle’s Mill for the Rogers Historical Museum’s open house, 2003.

Standing left to right: Aaron Anderson Van Winkle, Peter Van Winkle, Temperance Van Winkle, Mary Van Winkle.

Kneeling left to right: a confederate soldier and J. A. C. Blackburn. 7

1.4 Student art project portraying Van Winkle’s Mill. 7

Chapter 2

2.1 The Ozarks as geographic region (after Rafferty 2001:xiv). 14

2.2 Van Winkle project area showing surrounding towns. 16

2.3 Van Winkle’s Mill and Hobbs State Park and Conservation Area. 16

2.4 Peter Van Winkle and family circa 1870. Clockwise from top:

Norman, Temperance, Peter, Jr. (being held), Jefferson Davis, Robert E. Lee, Wallace (being held) and Peter Van Winkle (after Hicks 1990:14). 53

2.5 The Van Winkle home, circa 1870 (after Hicks 1990:28). 57

2.6 Raised gardens across from main house, circa 1970

(after Hicks 1990:28). 57

2.7 The Van Winkle Hotel (also called the Van Winkle House)

circa 1890 (after Hicks 1990:35). 58

2.8 Van Winkle’s Mill ca. 1870 (after Hicks 1990:26). 58

2.9 Van Winkle family gathered on the front porch of the Steele home in Rogers, 1901. Temperance Van Winkle is sitting center (after Hicks 1990:104). 70 xiv

2.10 Detail of 1901 Van Winkle photograph showing Aaron Anderson

Van Winkle standing in the shadows behind Temperance.

This is the only known photograph of Aaron. 70

2.11 “The Old South,” a nostalgic composite photo of five freedmen who had been enslaved in northwest Arkansas.

Sam Van Winkle is second from left. 71

Chapter 3

3.1 A Young Harold Bell Wright (after Shepherd of the Hills

Outdoor Theater 2004). 97

3.2 Title card from the 1912 silent film version of The Shepherd of the Hills (after Chudleigh 2003). 97

3.3 1870 Currier and Ives print based on The Arkansas Traveller. 111

3.4 “Barney Google and Snuffy Smith” Sunday page

May 3, 1936 (after Inge 1994:101). 111

3.5 “Old Ways are Best,” the frontice piece from the first edition of J. Vance Randolph’s The Ozarks: An Survival of Primitive Society (1931). 128

3.6 Front view of the church and high school on Community Hill, the Presbyterian mission established in Kingston, Arkansas (after Burnett 2000:82). 128

Chapter 4

4.1 Map showing the features identified by the original 1997-1999 mapping and testing project at Van Winkle’s Mill (after Brandon et al. 2000:16). 150

4.2 Survey area and features identified by the June 2000 systematic survey of the northern portion of Van Hollow (after Brandon and Davidson 2003:25). 152

4.3 Plan map of main house excavation area (after Brandon et al. 2000:25). 154 xv

4.4 North profile of backhoe trench N487-488/W488 showing buried midden associated with the first Van Winkle house (Feature 24). 154

4.5 Field stones marking probable slave cemetery 3BE652. 157

4.6 Extent of 1999 excavations at Feature 9 uncovering one

chimney box and part of the foundation. 159

4.7 Extent of 2000 excavations at Feature 9 uncovering the back foundation wall. 160

4.8 Plan view of the entire excavated footprint of Feature 9, a postbellum mill worker’s quarters. 161

4.9 Artifacts recovered from Feature 27. a) alkaline-glazed stone ware; b) molded ironstone; c) ironstone base with maker’s mark;

d) hand-[painted whiteware; e) patent medicine bottle with hand-tooled lip; f) ceramic doll leg; g) mule shoe; h) hand-forged hook. 164

4.10 Artifacts recovered from Feature 29. a) Albany-slipped stoneware;

b) alkaline-glazed stoneware; c) automatic machine-made bottle neck; d) bottle base with Owens scar and maker’s mark; e) railroad spike. 165

4.11 Artifacts recovered from Feature 31. a-b) wagon axle “box”;

c) “L” bracket; d) wagon bolster plate; e) wagon strap bolt. 168

4.12 Topographic map and geophysical grid area for Feature 31

(blacksmith shop). 172

4.13 Magnetometer (left) and resistance data (right) from geophysical survey of Feature 31 (blacksmith shop). 173

4.14 Planview of test excavations at forge box, Feature 31 (blacksmith shop). 174

4.15 Test excavations at Feature 31 showing blacksmith’s forge box, view from the east. 174

4.16 Profile of forge box excavations at Feature 31. 175

4.17 Photographic view of excavations along the back wall of the blacksmith shop. 175 xvi

4.18 Plan view map showing the distribution of shovel test at Feature 32

(possible mule paddock). 178

4. 19 Artifacts recovered from Feature 32. a-b) mule shoes; c) L-bracket;

d-h), large penny cut nails from various shovel tests. 178

4.20 Results of geophysical survey at Feature 33 (possible slave quarters). 181

4.21 Extent of 2001 test excavations at Feature 33 (possible slave quarters). 181

4.22 View of the boiler platform (left) and the flywheel support (right) during the 2003 spring break excavations in the mill complex. 184

4.23 Topographic map of the mill complex area delineating the geophysical grids and excavation areas. 185

4.24 Results of geophysical investigations in the mill complex area:

magnetmoetry (left) and electrical resistance (right) from RSG3;

in the inset, electromagnetic conductivity (left) with possible data trends (right) from RSG1. 186

4.25 Artifacts from the mill complex testing. A-d) large rivets; e) large square-headed bolt; f) belt staple; g) file. 189

4.26 Extent of 2001 excavations on the boiler platform in the mill complex. 189

4.27 Feature 2, the Van Winkle Spring house, 1997. 192

4.28 View of the Van winkle house showing the semi-detached kitchen and kitchen addition. 192

Chapter 5

5.1 Map of the late nineteenth century of the cultural landscape of

Van Winkle’s Mill 196

5.2 Map of the late nineteenth century of the cultural landscape of

Van Winkle’s Mill showing industrial and residential zones and activity flow. 197

5.3 A partially reconstructable shallow refined earthenware bowl with an impressed makers mark found under the foundation wall at Feature 9. 213 xvii

5.4 Drawing by Darby Hicks included in the Van Winkle family

history (Hicks 1990:15) demonstrates the importance of the frontier trope to the Hicks family. 213

Chapter 6

6.1 Door hinge and ceramic doorknob recovered from

Feature 9 excavations. 236

6.2 Clothing buttons recovered from Features 9 and 33.

Specimens a-d, f-h, j, k, m-v and x recovered from Feature 9.

Specimens e, i, l, and w recovered from Feature 33 237

6.3 A sample of some of the children’s toys recovered from Feature 9

Excavations: a-c) transfer print alphabet plate Fragments; d-h) porcelain doll fragments; i) cast iron cap pistol Fragment; j) a porcelain marble; k-l) bakelite children’s rings. 238

6.4 Harmonica reed plate fragments recovered from Feature 9. 239

6.5 Small objects recovered from Feature 9 excavations:

a) a copper “shield” nickel; b) a “Granger Twist” tobacco tag excavated from Feature 9; c) a pristine “Granger Twist” tobacco tag for comparative purposes (from the collection of the author). 239

6.6 A variety of decorated ceramics recovered from Feature 9 and

Feature 33 excavations. 241

6.7 A fragment of a rooster ceramic figurine recovered from Feature 9. 241

Chapter 7

7.1 Teacher evaluation forms from the Rogers Historical Museum’s

Van Winkle education program. 263

7.2 Student evaluation forms from the Rogers Historical Museum’s

Van Winkle education program. 264 xviii

Chapter 1

Introduction

Archaeology is often about what is no longer there—at least what is no longer visible on the surface of things. I was reminded of this one day while standing in the narrow Ozark hollow known as Van Hollow. It was a pretty, temperate day in the late spring of 2003. The sunlight was filtering through the green leaves of the abundant sycamore trees near the ruins of a large saw and gristmill once known as Van Winkle’s

Mill. Now the area was being developed, and interpreted as an historical site, as a part of the Hobbs State Park and Conservation Area.

When archaeological investigations of the Van Winkle’s Mill site began in 1997, Van Hollow (an abbreviated form of Van Winkle Hollow) was an overgrown jungle, with few discernible features. The mill had been destroyed around the time of First World

War in order to salvage its metal for scrap. The White River was dammed in 1966 in order to create Beaver Lake, which flooded the northern portion of the hollow. The Van

Winkle house, home of the mill’s owner Peter Van Winkle, was torn down in 1969.

Although the lake rarely floods as high as the location of the house site, time and the destructive forces of nature have slowly caused the site to deteriorate. Due to the site’s accessibility, looting has also occurred; including the scavenging of bricks from the house site and the removal of large limestone blocks from the raised formal garden.

Today, the site offers visitors only glimpses of what it once was. The only discernible surface features include the steps leading up to the elevated garden (Figure 1.1), a limestone spring house, the old Huntsville or Van Winkle Road, which ran in front of the house, and the foundation of the steam engine and boilers (Figure 1.2).

In its day, however, this mill was described as “the most extensive lumber business in the southwest” (Fayetteville Sentinel February 15, 1882). Additionally, the mill served as the anchor for a small community whose lives were connected with the mill in many different ways⎯the owner and his family, lathe operators, bookkeepers, sawyers, teamsters, store operators, blacksmiths, unskilled white wage laborers, enslaved African-

American laborers before the war and free black laborers after emancipation all lived and worked in the narrow Ozark hollow between the 1850s and the early-twentieth century.

By 1997, all that was left of this mill was the large limestone foundations for the platform that, in the nineteenth century, housed the three cast iron boilers and the steam-powered engine that ran the mills. Around this platform were several other features related to the mill. To the west was a deep but narrow trench approximately three meters deep with a wall built of massive pieces of cut limestone extending from the trench’s west face. This trench was where the mill’s legendary flywheel was mounted.

This flywheel, long demolished for scrap metal, gave me my first encounter with how my work as a historical archaeologist at Van Winkle’s Mill was entangled with the larger cultural memory of the Arkansas Ozarks. Historical accounts refuse to agree on the most basic facts about the mill’s equipment. There was no doubt that the flywheel was large and impressive. Beyond that fact, it was reported as being anywhere between

18 and 24 feet in diameter and weighing between 10,000 to 20,000 pounds (Bowers

2003:33; Elliott 1959:6, Rothrock 1972:93, D. Miller 1966:22, Scott 1962:171, Hicks

Figure 1.1. Remaining stairway to the formal raised garden in Van Winkle’s Mill, 2002.

Figure 1.2. Foundations for the flywheel at the site of the Van Winkle Mill complex, 1997.

1990:25, Goodspeed 1889:108). Moreover, none of the measurements given in the historical documentation correspond with the projected size of the wheel as ascertained through industrial archaeology (Bowers 2003:33-34).

My colleague, Robin Bowers, has pointed out that what is behind these contradictions is the place of the mill and its steam-powered machinery in the local imagination.

There is no doubt that the mill was a source of wonder for the inhabitants of Northwest Arkansas. Steam power was very much a new technology when the first steam engine was installed at the mill in the early 1860s. All other mills in the area, including flour mills, were water driven. The spinning of the flywheel, the noise from the engine and saws, and the smoke pouring from the smoke stack must have been impressive for local people visiting the mill⎯people who had likely never seen a locomotive. For rural residents who, even in the 1860s, were considered “backward hillbillies,” there must have been a tremendous pride in this symbol of modernity that graced their woods. (Bowers 2003:20)

Thus the flywheel cannot be discerned through solely historical or archaeological methodologies. To really understand the flywheel, how it becomes “legendary” in local history and how its real size gets obscured, one must take into account the cultural frameworks at work in the construction of our historical narratives. These frameworks have been referred to as our cultural memory.

The above quotation also offers a glimpse at a powerful theme in both local and national cultural memory that I will be examining in much detail later⎯the trope of the hillbilly. Bowers overtly evoked this trope when she pointed out that the nineteenth-century rural Ozark residents may have been considered “backward hillbillies.”

Although the term itself would not become widely used until the twentieth century, Bowers’ point is well taken. There is an interesting disjuncture between these people who may have been conceived of as “backward” or “isolated” and the “symbol of modernity” that is Van Winkle’s Mill. Bowers, however, was also playing into the trope when she stated that many of these inhabitants “had likely never seen a locomotive.”

Each time I have given tours, lectures or interviews about the work being conducted in Van Hollow, I have encountered this disjuncture. I have found the listeners filtering what I was saying through various tropes of cultural memory. After listening to me explain the history and archaeology of Van Winkle’s Mill⎯complete with the New York born, Illinois raised owner and his family, the large industrial saw mill complex that dominated the regional markets, and the enslaved African-American labor which made it possible⎯many people would go out of their way to tell me of their surprise at (or sometimes disbelief of) my narrative. Sometimes they were surprised by the presence of such a powerful industrial capitalist in the Ozarks, sometimes it was surprise that the institution of slavery was prevalent in the upland South or industrial settings, sometimes it was a surprise that any racial or ethnic diversity ever existed in the Ozarks at all. The further I delved into this disjuncture, and the more interactions I had with public interpretations of Ozark history, I began to suspect that the origins of some of the tropes at work in our cultural memories were entangled with the forces at work at Van Winkle’s

Mill in the mid-to-late-nineteenth century. The onset of modernity, industrialization, the creation of tropes such as the hillbilly, changing conceptualizations of race and even the formation of modern whiteness itself all left a mark on either the physical landscape or the cultural memory of Van Winkle’s Mill.

The Van Winkle Educational Program

By that afternoon in 2003, the area where the three-story mill complex once stood was now a clearing with grass and the occasional tree; it was beginning to look like a state park. I was struck by the contrast of the scene before me and the one I remembered from the first time I saw the mill’s ruins in 1997. Then there was a dense tangle of underbrush⎯saw briers, poison ivy, and all variety of small bushes and trees. These plants tugged at every piece of equipment and clothing we had as we tromped around the hollow in the attempt to figure out where the archaeological features associated with this mill were and map them.

I was there on that afternoon to attend an open house in the hollow hosted by Hobbs

State Park and the Rogers Historical Museum. The Rogers Historical Museum, in collaboration with Hobbs State Park, the Arkansas Humanities Council, the Arkansas

Archeological Survey and the Bentonville and Rogers School Districts had launched an ambitious project as a part of their educational programs for schools. Using historical documents, archaeological information, family history and other sources, they proposed to expose some 1,300 fifth-grade children in Benton County schools to critical history through the story of Van Winkle’s Mill.

The curriculum was multidisciplinary and sought to introduce students to aspects of the natural environment that made the lumber mill possible, the social, political and economic history of Northwest Arkansas, the concept of industrial slavery, the impact of the Civil War on the region, and how different types of sources (including historical archaeology) come together to reconstruct the history of a site. The program included a

Figure 1.3. Students portraying historical figures at Van Winkle’s Mill for the Rogers Historical Museum’s open house, 2003. Standing left to right: Aaron Anderson Van Winkle, Peter Van Winkle, Temperance Van Winkle, Mary Van Winkle. Kneeling left to right: a confederate soldier and J. A. C. Blackburn.

Figure 1.4. Student art project portraying Van Winkle’s Mill.

class visit to provide background information for (and primary documentation of) Van

Winkle’s Mill and a field trip to Van Hollow where students rotated through several key interpretive locations.

On the day of the open house the park employees, teachers, students and even

Van Winkle descendants (who traveled from Texas and California to attend) all gathered in the grassy clearing near the mill’s foundations. For this special occasion several extra events were taking place. A group of seventh grade students in historical garb

“reenacted” first person monologs at various locales throughout the hollow for the teachers and descendants (Figure 1.3) and student art projects depicting aspects of life at

Van Winkle’s Mill were on display (Figure 1.4).

It was that afternoon of being on a historical site full of interpretation, with actually very little to materially indicate that anything historic had happened, that lead me to think about archaeology and absence. Everyone in the hollow that day, or any day for that matter, read into the remains of Van Winkle’s Mill its totality. They all looked at the few above-ground features and imagined what they thought the whole of the community must have been like. Dydia DeLyser (2001) has elsewhere examined this phenomenon in western ghost towns where returning visitors actually remember things that never existed and comment on the perceived changes. This was the most active experience of cultural memory I had encountered, and where I have chosen to begin and end this dissertation.

My questions seemed relatively straightforward and centered around the disjuncture between the history and archaeology I worked with and what people expected from the

Ozark past. What work does cultural memory perform on and in the Arkansas Ozarks and Van Winkle’s Mill in particular? How did the tropes of cultural memory come to be and situationally change? What was differentially silenced and remembered in Ozark history?

Synopsis

Both Van Winkle’s Mill and the questions to which it points lend themselves to interrogation using a combination of excavated archaeological materials, historical archival records and an examination of cultural memory. Using these methods and resources in combination will hopefully allow for a more nuanced understanding of each question in context and provide a clearer, interlocking picture of Van Winkle’s Mill and its place in the local and regional cultural memory. This dissertation will make an attempt to do this by moving heuristically between archaeological methodologies and other approaches.

First, in Chapter 2, I will provide the historical context(s) of the Arkansas Ozarks and Van Winkle’s Mill. This chapter will contain a brief historical overview, but it will also contain a number of small vignettes that will go into detail about a few people and events. This technique has been adopted in an attempt to avoid broad generalizing statements about the Ozarks. Concentration on the particulars of these people and events will hopefully demonstrate the diversity of the region’s historical experiences.

Following this overview, I will turn to the history of Van Winkle’s Mill itself. I trace the community and its inhabitants as we can best discern them through historical documents and oral histories from the mill’s founding in the 1850s to the site’s incorporation into the Hobbs State Park and Conservation Area.

In Chapter 3, I will examine cultural memory; what constitutes cultural memory, how it is produced and what role it plays in the creation of Ozark historical narratives.

After a brief explanation of the theoretical orientation I deploy when evoking cultural memory, I turn again to the particulars of the Arkansas Ozarks. Here I follow the construction and use of various “tropes” in cultural memory that help us make sense of

Ozark history. I then concentrate on one of the more powerful tropes⎯the aforementioned hillbilly⎯in order to assess how that trope came to be and what it emphasizes and silences in the Ozark past(s).

Chapter 4 will outline both archaeology’s engagement with the hillbilly trope and the archaeological investigations that have taken place in Van Hollow from 1997 to 2003.

I will not provide all of the technical details of these excavations as I have done so elsewhere (Brandon et al. 2000; Brandon and Davidson 2003; Brandon 2005), but I will endeavor to provide the information that is pertinent to understanding the analyses that will take place in the following chapters.

Following these three overviews and interpretive contexts, I provide chapters that examine two distinct interpretive cases at Van Winkle’s Mill. First, I will examine the cultural landscape of the site (Chapter 5). Here I will use the archaeological recoveries in dialog with the historical materials and cultural memory in an attempt to understand the landscape as seen through nineteenth, twentieth and twenty-first century eyes. What were the messages structured within the landscape of the working nineteenth-century milling community and how have those messages been interpreted by later observers through varying lenses of cultural memory.

I will then turn to examine the artifacts recovered from the site (Chapter 6), especially those recovered from the workers quarters, in order to examine material consumption and what these artifacts might suggest about the daily lives and identities of the occupants of Van Winkle’s Mill.

Following these two cases, I then return to the framework of cultural memory.

Chapter 7 will outline the Rogers Historical Museum’s Van Winkle Program and its intent to expose students in the Benton County school system to critical interpretations of

Van Winkle’s Mill, race, slavery, the Civil War and, less directly, cultural memory. I will also analyze the student and teacher evaluations of the program in order to ascertain the results of such an ambitious project. Finally, Chapter 8 will attempt to draw on the preceding chapters to form some conclusion regarding the dense, entangled thicket of history, cultural memory, modernity and archaeology. A thicket much more entangled than the vegetation that we encountered when we first saw Van Winkle’s Mill in 1997.

Chapter 2

“A Frontier More South than West”:

A Historical Overview of the Arkansas Ozarks and Van Winkle’s Mill

In the early nineteenth century, Arkansas was “a reincarnation of the American frontier” and thought of in cultural memory in the same fashion as places like Kentucky had been in the popular imagination of the late 1700s (Rohrbough 1990:219). It was seen as a “wild borderland in the process of being tamed and brought within the sphere of civilization through pioneer settlement” (Sabo 1990:139). Even at that point in history, however, the “forces of movement” and modernization which had overtaken the myriad of previous frontiers had been put into motion in Arkansas and the Arkansas Ozarks

(Blevins 2002:9; Rohrbough 1990:219). The decades following the 1803 Louisiana

Purchase, which included the land that was to become the Arkansas Ozarks, saw the arrival of what some have called “ambitious pioneers”—merchants, politicians, lawyers, farmers, aspiring planters and their families—until “the Arkansas Ozarks were home to over a quarter of a million American settlers” by the end of the nineteenth century, perhaps more than the land could support (Blevins 2002:9).

The story of Van Winkle’s Mill takes place in this period of exponential growth, a time that regional historian Brooks Blevins calls “an era of starts and stops, false hopes, important transformations and stubborn continuities” (Blevins 2002:69). Of course, this period was a time of rapid change in a region which is usually portrayed as “timeless” and “unchanging.” This is just the oppositional dichotomy that this work seeks to explore through the device of Van Winkle’s Mill.

Below, I will briefly outline some of the historical context(s) of the Arkansas

Ozarks and Northwest Arkansas in particular. I do not intend this to be an exhaustive treatment of the subject. Instead, I intend to give the reader a feeling for the historical and cultural setting of the region. In doing so, I also set the stage for an examination of how we view this region’s history—both locally and on a national level.

Arkansas, the Ozarks, and Northwest Arkansas:

An Explanation of Geography and Scale

For those unfamiliar with the geography of the region, a bewildering number of terms are used in this work. First, it should be said that shifts in terminology in the discussions that follow are largely representative of shifts in analytical scale. That is, I typically will begin the discussion of a topic with the state of Arkansas as my analytical unit and then I will shift my discussion to either the Ozarks or, more restrictively, the

Arkansas Ozarks (Figure 2.1).

The Ozarks, unfortunately, are a region whose boundaries “are vague to most people and subject to interpretation and disagreement by experts” (Rafferty 2001:1). For the purposes of this work, the Ozarks are the interior highlands as delimited by cultural geographers and include parts of four states—Arkansas, Missouri, Oklahoma and Kansas.

The “unifying criteria” of the Ozarks as a region would be the “greater relief and steeper slopes” than the surrounding geographic provinces (Rafferty 2001:1)—bounded to the south by the Arkansas River valley, the Missouri River to the north, the Mississippi

Figure 2.1. The Ozarks as geographic region (after Rafferty 2001:xiv).

lowlands to the east and the Osage Plains to the west. More importantly, I believe, is the

“sense of place” surrounding the Ozarks (McNeil 1995:3). Although experts in geography, history and folklore may not agree on the physical boundaries, the region— both internally and on a national level—is held to be more than “simply another rural region quite like any other in most regards” (Blevins 2002:6). The “region’s past and present are riddled with paradox” and seen through a “thicket of myth, nostalgia and stereotype” formed and remembered through lived experience, popular culture and the tourism/recreation industries with which the region has long been associated (Blevins

2002:6-7; see also Morrow and Myers-Phinney 1999). It is this aspect of Ozark historical and cultural memory that this work attempts to examine.

In the pages that follow, my discussions will focus on Northwest Arkansas—the finest of my analytical scales (Figure 2.2). Like its larger counterpart, the Ozarks, there is little agreement about what counties actually make up Northwest Arkansas. For my purposes, I have attempted to take a cultural approach by roughly equating Northwest

Arkansas with what Blevins (2002) refers to as “the western counties” which includes parts of the Boston Mountains, the Springfield Plateau and small portions of the White

River Hills. The core counties of Northwest Arkansas are Washington, Benton, Carroll and Madison Counties, but I have also included northern portions of Crawford County, in the steep Boston Mountains, in my discussions.

Finally, the site of Van Winkle’s Mill is located in the physiographic subdivision of the White River Hills of the Ozark Mountains in the southeastern portion of Benton

County—along Little Clifty Creek and between War Eagle Creek to the south and Beaver

Figure 2.2: Van Winkle project area showing surrounding towns.

Figure 2.3. Van Winkle’s Mill and Hobbs State Park and Conservation Area.

Lake to the north (Figure 2.3; Rafferty 2001:12). This region of the Ozarks is known for its rugged relief and karst topography (Rafferty 2001:15-16). Historically, it was described as “so broken and uneven that it [was] mostly unfit for cultivation, except in the valleys of the streams” (Goodspeed 1889:2).

“Who Are These Ozark People?”

Ethnicity and Immigration in the Trans-Upland South

Prior to the 1803 Louisiana Purchase, the Spanish and French explored and settled

Arkansas and the Ozarks. Spanish settlement in the Ozarks was largely negligible and

French settlement in the eighteenth century centered upon the lead-rich district near St.

Genevieve in what is now Missouri (McNeil 1995:9-10; Rafferty 2001:42-44; Sabo

1990:138). In some popular histories, the Ozarks owe their name to an Anglicized version of either the French phrase bois aux arcs, which refers to the wood used to make

Native American bows, or a French abbreviation for the Osage tribe (McNeil 1995:2;

Steele and Cottrell 1993:11).

This evocation of the Osage serves to remind us that the Ozarks have been, in fact, populated by Native Americans for some 12,000 years before the coming of the Spanish or French (Sabo 2002:1; Sabo and Early 1990:36-40) and, despite some opinion to the contrary, have not yet disappeared. In fact the Osage, Quapaw, and other Native

American immigrant tribes (such as the Cherokee) are intertwined with the history of

Northwest Arkansas—especially in its role as a frontier abutting the infamous Indian

Territory. Moreover, it is the 1828 establishment of the western territorial boundary between Indian Territory and Arkansas which opened up Northwest Arkansas for full-scale settlement.

Although a vanguard of European Americans were living in the region illegally since the first decades of the nineteenth century, the 1817 Cherokee Treaty reserved much of western Arkansas as Cherokee land—in 1828 the tribe exchanged this land for seven million acres north of the Arkansas River in Indian Territory (McNeil 1995:8).

The immediate impact of the establishment of the 1828 boundary was to officially open

Northwest Arkansas for white settlement (Hilliard 1983; Rafferty 2001:52). Although

Cane Hill was an established community prior to 1828, it was the newly established town of Fayetteville which would become the county seat of Washington County and territorial center of Northwest Arkansas (Hilliard 1983). Benton County, where Van Winkle’s Mill is situated, was largely settled during the 1830s and became the first county established after Arkansas became a state in 1836 (Black 1975:7; Goodspeed 1889:24-25).

Numerous histories have characterized these early Ozark settlers as white immigrants from other places in the upland South (i.e., eastern Tennessee, southern

Kentucky, western North Carolina and southwestern Virginia) who were descended from

British or “Scotch-Irish” stock (e.g., Blevins 2002:19; :Flanders 1979:176-178; Rafferty

2001:2, etc.). Although it is true that other groups represent a minority of the region’s population, they nonetheless “constitute an important element in the population of the

Ozarks” (Gerlach 1976:8). Waves of Germans (1830-1850), Mennonites (1850-1890), Swiss (1860s), Poles and Bohemians (late 1800s), Moravians and Italians (1900s) all came to the Ozarks in search of the same things that had brought those who came from elsewhere in the Southern United States (Gerlach 1976; McNeil 1995:9-15; Rafferty

2001:62-73).

In Northwest Arkansas in the 1830s, a small German settlement founded Dutch

Mills (Hermannsburg) in Washington County, Arkansas, and a group of German-speaking Swiss settled near Altus on the edge of the Boston Mountains in the 1880s

(Rafferty 2001:66). More locally acknowledged is a group of forty Italian families fleeing Sunnyside Plantation—a failed attempt to use European immigrants as plantation labor in lowland Chicot County—who settled in the Ozark Mountains on the border with

Indian Territory where they founded Tontitown, Arkansas in 1898 (Whayne 1993:21). A decade later this group had grown to seventy families living on 1,400 acres (Blevins

2002:44). Tontitown, west of Springdale and north of Fayetteville, still strongly self-identifies itself as Italian and catholic.

This surprising number of minority communities are small when compared to the much larger number of settlers from Tennessee, Mississippi, Alabama, Kentucky, North

Carolina, Georgia and Virginia—not to mention a later and larger wave of settlers from

Missouri, Kansas, Illinois and other Midwestern states (Blevins 2002:38-39)—but they existed nonetheless. Thus the history of the Ozarks included the entanglement and interaction of these southern white immigrants (the so-called “Scotch-Irish”), Midwestern immigrants, Germans, Poles, Italians, indigenous Native groups (such as the Osage), recently immigrated Native Americans (such as the Cherokee), free blacks and the enslaved African Americans brought in bondage by their masters to the mountains of the

Ozarks.

It is to the last group that I wish to turn next. As mentioned in the introductory chapter of this work, part of the labor at Van Winkle’s Mill was provided by slaves. Like the other groups mentioned here, African Americans (and slavery) are not a common part of the traditional cultural memory of the Arkansas Ozarks. For this reason, I will provide a brief background of slavery in the region.

"Incongruous, Implausible and Unworthy of Examination":

Slavery in Northwest Arkansas

In the 1980s, John Solomon Otto, a member of the earliest generation of archaeologists to tackle the subjects of slavery and the African Diaspora, conducted a small historical research project in Yell County, Arkansas. Although his work was brief, he made some important, timely observations:

Historians of American slavery have traditionally studied the institution on a regional basis, searching for generalization that might hold true for the entire South. Despite their attempts at regional generalization, however, they have used sources that come largely from the plantation belt—the lowland and piedmont areas of the Old South…by using these sources from the plantation belt, historians have made sweeping judgments about slavery that may not be applicable to the entire South, since roughly half of the slave territory of the United States lay outside the plantation regime. (Otto 1980:35)

Since the time of his writing there have been exceptions to this rule (e.g., C. Dew

1994; Hahn 1982; Inscoe 1989; Schlotterbeck 1982, etc.), although slavery in the upland

South and slavery in non-agricultural settings, in particular, remain grossly understudied.

As late as 1995, Ted Smith would write in his study of slavery in Washington County, Arkansas that “[t]o historians and non-historians alike, the idea of slavery in the mountains has seemed incongruous, implausible or even unworthy of examination”

(Smith 1995:2; see also Bolton 1999).

The reasons why this has become the case will be examined in later chapters, but for now it is important to note that in 1840, the first United States census for the state of

Arkansas recorded over 20,000 inhabitants in the eight counties that made up the

Arkansas Ozarks⎯including almost 2,000 enslaved African Americans (Blevins

2002:18).

In the decades that followed the opening of Northwest Arkansas to legal white settlement (1828), immigrants filled the Ozark counties. Throughout the 1820s and

1830s, it was the northwestern Arkansas counties, with their “healthful upland valleys” that were the “principle attraction” for farmers—and slave holders—coming to Arkansas

(Taylor 2000:27, 50). These migrants brought the labor which would become the foundation for “a workforce that would out-produce the rest of the state in manufacturing and whose farmers would grow more corn, wheat, oats and Irish potatoes” than any other region in the state—a workforce that, of course, included not only white labor, but their

African-American (mostly enslaved, but a few free) counterparts as well (Smith 1995:5, 16-17). In fact, by 1830, approximately 68% of the state’s total population and 61% of the slave population was located in the northwestern part of the state (Catalfamo-Serio

1979:105; Mintz and Hilliard 1992; Taylor 2000:27). In 1831, the Arkansas Gazette declared that Washington County (which then included what would become Benton

County in…