Time, Typology, and Point Traditions in North Carolina Archaeology: Formative Cultures Reconsidered

By I. Randolph Daniel

A reconsideration of the seminal projectile point typology
 
In the 1964 landmark publication The Formative Cultures of the Carolina Piedmont, Joffre Coe established a projectile point typology and chronology that, for the first time, allowed archaeologists to identify the relative age of a site or site deposit based on the point types recovered there. Consistent with the cultural-historical paradigm of the day, the “Coe axiom” stipulated that only one point type was produced at one moment in time in a particular location. Moreover, Coe identified periods of “cultural continuity” and “discontinuity” in the chronology based on perceived similarities and differences in point styles through time.
 
In Time, Typology, and Point Traditions in North Carolina Archaeology: Formative Cultures Reconsidered, I. Randolph Daniel Jr. reevaluates the Coe typology and sequence, analyzing their strengths and weaknesses. Daniel reviews the history of the projectile point type concept in the Southeast and revisits both Coe’s axiom and his notions regarding cultural continuity and change based on point types. In addition, Daniel updates Coe’s typology by clarifying or revising existing types and including types unrecognized in Coe’s monograph. Daniel also adopts a practice-centered approach to interpreting types and organizes them into several technological traditions that trace ancestral-descendent communities of practice that relate to our current understanding of North Carolina prehistory.
 
Appealing to professional and avocational archaeologists, Daniel provides ample illustrations of points in the book as well as color versions on a dedicated website. Daniel dedicates a final chapter to a discussion of the ethical issues related to professional archaeologists using private artifact collections. He calls for greater collaboration between professional and avocational communities, noting the scientific value of some private collections.
 

• Language: English
• Publisher: University Alabama Press
• Release date: Mar 2, 2021
• ISBN: 9780817393434

Book preview

TIME, TYPOLOGY, AND POINT TRADITIONS IN NORTH CAROLINA ARCHAEOLOGY

ARCHAEOLOGY OF THE AMERICAN SOUTH: NEW DIRECTIONS AND PERSPECTIVES

SERIES EDITOR

Christopher B. Rodning

EDITORIAL ADVISORY BOARD

Robin A. Beck

John H. Blitz

I. Randolph Daniel Jr.

Kandace R. Hollenbach

Patrick C. Livingood

Tanya M. Peres

Thomas J. Pluckhahn

Mark A. Rees

Amanda L. Regnier

Sissel Schroeder

Lynne P. Sullivan

Ian Thompson

Richard A. Weinstein

Gregory D. Wilson

TIME, TYPOLOGY, AND POINT TRADITIONS IN NORTH CAROLINA ARCHAEOLOGY

Formative Cultures Reconsidered

I. Randolph Daniel Jr.

THE UNIVERSITY OF ALABAMA PRESS

TUSCALOOSA

The University of Alabama Press

Tuscaloosa, Alabama 35487-0380

uapress.ua.edu

Copyright © 2021 by the University of Alabama Press

All rights reserved.

A Dan Josselyn Memorial Publication

Inquiries about reproducing material from this work should be addressed to the University of Alabama Press.

Typeface: Adobe Caslon Pro

Cover images: North Carolina point types and projectile point traditions of the North Carolina Piedmont; courtesy of North Carolina Archaeological Collection, Research Laboratories of Archaeology, University of North Carolina, Chapel Hill; I. Randolph Daniel Jr.; American Philosophical Society

Cover design: Lori Lynch

High-resolution color photographs of the figures in chapter four are available for viewing online at https://archaeology.sites.unc.edu/home/rla/archives/north-carolina-projectile-point-types.

All figures are courtesy of the author unless otherwise noted.

Cataloging-in-Publication data is available from the Library of Congress.

ISBN: 978-0-8173-2086-7

E-ISBN: 978-0-8173-9343-4

Contents

List of Illustrations

Preface

Acknowledgments

Introduction

1. Revisiting the Formative Cultures Typology

2. Communities of Practice and Point Traditions

3. North Carolina Stone Types

4. North Carolina Chipped-Stone Point Types

5. Rethinking the Formative Cultures Point Sequence

6. Three Decades of Working with Collectors in North Carolina

Notes

References Cited

Index

Illustrations

FIGURES

High-resolution color photographs of the figures in chapter four are available for viewing online at https://archaeology.sites.unc.edu/home/rla/archives/north-carolina-projectile-point-types.

I.1. Formative Cultures sequence from Coe

3.1. Prehistoric stone quarry locations mentioned in text

4.1. Site locations mentioned in text

4.2. Hardaway Blades

4.3. Clovis points

4.4. Clovis points

4.5. Clovis points

4.6. Clovis points

4.7. Redstone points

4.8. Redstone points

4.9. Cumberland points

4.10. Hardaway-Dalton points

4.11. Hardaway-Dalton points

4.12. Hardaway-Dalton points

4.13. Hardaway Side-Notched points

4.14. Other Side-Notched points

4.15. Palmer Corner-Notched points

4.16. Kirk Corner-Notched points

4.17. Kirk Corner-Notched points

4.18. Southern Hardin points

4.19. Bifurcate points

4.20. Kirk Stemmed points

4.21. Kirk Stemmed points

4.22. Stanly Stemmed points

4.23. Morrow Mountain Stemmed points

4.24. Guilford Lanceolate I points

4.25. Guilford Lanceolate II points

4.26. Halifax Side-Notched points

4.27. Large Savannah River Stemmed points

4.28. Large Savannah River Stemmed points

4.29. Large Savannah River Stemmed points

4.30. Large Savannah River Stemmed points

4.31. Small Savannah River Stemmed points

4.32. Small Stemmed points: Otarre Stemmed points

4.33. Small Stemmed points: Otarre Stemmed points

4.34. Small Stemmed points: Swannanoa Stemmed, Plott Short Stemmed, and Pigeon Side-Notched points

4.35. Small Stemmed points: Thelma Stemmed and Randolph Stemmed points

4.36. Mack point

4.37. Mack point

4.38. Mack point

4.39. Large Triangular points: Yadkin Triangular, Eared Yadkin, Roanoke Triangular

4.40. Large and Small Triangular points: Garden Creek Triangular, Connestee Triangular, Caraway Triangular, Pisgah Triangular

4.41. Small Triangular points: South Appalachian Pentagonal, Pee Dee Triangular, Pee Dee Pentagonal, Clarksville Small Triangular

5.1. North Carolina projectile point types and traditions for the Paleoindian and Early Archaic periods

5.2. North Carolina projectile point types and traditions for the Middle Archaic and Late Archaic periods

5.3. North Carolina projectile point types and traditions for the Early Woodland, Middle Woodland, and Late Woodland periods

6.1. Members of the 1937 Town Creek Excavation Committee

6.2. Artifact forgeries from the Valentine Collection

6.3. Clovis replica sold as a genuine artifact

TABLES

1.1. Point type comparisons in Formative Cultures

4.1. North Carolina projectile point types and traditions

Preface

ONE OF THE MANY OPPORTUNITIES I had as a graduate student at the University of North Carolina at Chapel Hill was to examine the numerous artifacts curated by the Research Laboratories of Archaeology (RLA). In fact, I spent several semesters classifying nearly 48,000 untyped projectile points collected from over 2,800 sites around the state (Daniel and Davis 1996; Davis and Daniel 1990). The points were untyped because they had been collected in the first decades of the twentieth century before the publication of Joffre L. Coe’s The Formative Cultures of the Carolina Piedmont (Coe 1964) (hereafter Formative Cultures). That monograph is the typological touchstone for projectile point classification in North Carolina, if not much of the southeastern United States. In the absence of that typology, students involved in creating the RLA artifact catalog simply listed all points as CSPP (chipped stone projectile points). Even after the establishment of the point typology, no efforts were made to identify specific point types in the collections as Joffre Coe, then director of the RLA, depended on students to make only basic typological decisions in cataloging artifacts. For example, with a little training, students could identify points, scrapers, and flakes among stone artifacts, but they were not expected to make specific point type classifications. Decades later, it remained for an eager graduate student like me to examine each point and classify it as to type. Consequently, important information regarding the cultural-historical association of all those sites was added to the state’s site files and eventually synthesized by Theresa McReynolds (2005).

A personal benefit of that work was that I learned the point types from North Carolina. Indeed, I had the luxury of using the type collections curated by the RLA as my comparative baseline. It is one thing to see artifact photographs and read type descriptions in Formative Cultures; it is another thing to examine those type specimens firsthand. I also had the advantage of asking RLA archaeologists their opinions regarding artifacts that left me scratching my head. While most of the points I examined could be identified based upon the conventional typology, there were instances of points that varied from a formal type description just enough to defy being pigeonholed into an existing type. How were those points to be classified? I suspect that every professional or avocational archaeologist who works with stone tools eventually encounters this dilemma. How archaeologists solve this problem is often referred to as the lumper versus splitter debate in archaeology.¹ One is a typological lumper if the artifact is included in an existing type based upon a few key traits while ignoring some differences. A typological splitter, on the other hand, focuses on the variation rather the similarities to a known type and creates a separate (i.e., new) type. Either approach has its drawbacks. Doing the former appears to belie the type description itself, while doing the latter opens the door to creating a proliferation of types that undermines their usefulness as a classificatory tool. For the record, as a student attempting to classify several tens of thousands of points, I thought it more prudent to lump than to split and that was my approach as I examined drawer after drawer of artifacts.² Nevertheless, I did so with the uneasy feeling that I was missing something important typologically.

I also encountered a second somewhat related typological issue as I worked. How was I to account for previously unrecognized types like bifurcate and fluted points that were present in the collections but were not part of the Formative Cultures sequence? While I was comfortable in departing from the Coe typology in those instances, I still wondered what the cultural-historical implications were for those types with respect to the Formative Cultures sequence. For example, following the practice of scholars in Tennessee where bifurcates were better known (e.g., Chapman 1975), archaeologists in North Carolina simply tacked bifurcates temporally to the end of Early Archaic, placing them between Kirk Corner-Notched and Kirk Stemmed points. But this placement was problematic to me as it belied the direct link between corner-notched and stemmed points that Coe (1964) set forth. Moreover, if bifurcates were part of the Piedmont sequence why were they not identified by Coe originally? As another example, why did Coe tend to ignore the presence of fluted points in North Carolina (Daniel 2006)? Coe (1964:120) does acknowledge their presence at Hardaway as surface finds but emphasizes the Hardaway Blade as the indigenous Paleoindian type in the Piedmont. Given the presence of Clovis points in North Carolina, what was the relationship between the Hardaway Blade and Clovis point types?

These questions, then, illustrate the problem of unaccounted-for artifact variation I struggled with in classifying those 48,000 points several decades ago. Suffice it to state that examining those points was a formative experience with respect to how I view typology today. I came away from that undertaking thinking that I had been schooled in point typology. Yet, I also came away with the (perhaps heretical) notion that archaeologists did not know as much about the typology of North Carolina points as we thought. In retrospect, I view this experience as the genesis of this book. Decades later, I have read numerous site reports with point type descriptions and examined countless other points from collections across the state, all of which has reinforced the idea that a reassessment of the Formative Cultures typology and its cultural-historic sequence is needed. This book is the result.

Acknowledgements

IF THE SEED FOR THIS book was planted during my graduate school days, then its completion several decades later marks a long gestation. Even if one measures its conception only from the date the first words were written, that period of development represents intermittent intervals of work spread out over the last several years. Regardless, enough time has elapsed in the process that I am grateful to several individuals for aiding its completion.

As I recount in chapter 6, I have had the benefit of examining numerous private collections over the years. Uncharitably, I cannot acknowledge them all by name, but a few individuals deserve mention because they allowed me to photograph some personal artifacts used in this book, specifically identified in chapter 4. These individuals include Jimmy Boswell, Sammy Evans, Mark Pace, Warner Williams, and Jim Wilson. Several colleagues, on behalf of several institutions, also facilitated photographs of certain artifacts. These include Emilie Cobb and Gail Benson (Rankin Museum), Linda Carnes-McNaughton (Fort Bragg Cultural Resources Program), Paul Thacker (Archaeology Laboratory, Wake Forest University), John Privett (Raven Rock State Park, North Carolina), and John Mintz (North Carolina Office of State Archaeology). The artifacts from the North Carolina Office of State Archaeology include fluted points donated by Bob Oshnock, whose then extensive and well-documented personal collection I viewed several years ago. The donation of his collection to the North Carolina Office of State Archaeology epitomizes the responsible and responsive collector discussed in chapter 6.

Other colleagues also provided key comments on this manuscript. Thomas Emerson, Dale McElrath, and David Thulman were especially helpful, answering my questions regarding point typologies in general along with comments on an early draft of the communities of practice concept I use in chapter 2. They also provided key references regarding the latter. All of this is to say that while I drew inspiration from their work, they should in no way be held accountable for that fact. I also benefitted from the valuable comments one anonymous reviewer provided on a draft of this manuscript.

Along the way, I have also profited from informal conversations with Dan Cassedy, Matt Jorgenson, and Paul Webb that have filled gaps in my knowledge regarding relevant projects they have been involved in around the state. Similarly, Larry Kimball and Tom Whyte provided input regarding their perspectives on point types from the North Carolina mountains. As Larry often has stated, the Coe Axiom works, except when it doesn’t. My recognition of fluted point types in the Carolinas has benefited from conversations with my longtime colleagues from South Carolina, Al Goodyear and Chris Moore. I am also grateful to another South Carolina colleague, Joe Wilkinson, for introducing me to the rare Southern Hardin type.

Yet another colleague deserves a special thanks. Steve Davis, at the Research Laboratories of Archaeology at the University of North Carolina, Chapel Hill, helped me in numerous ways. Beginning in my graduate school days, Steve tutored me in Coe’s typology. Since then, I have benefitted from innumerable conversations with him regarding chipped stone projectile points. More directly related to this project, Steve was my host for a week-long research trip to the RLA during the summer of 2018, where I examined and photographed many of the points illustrated here. Like the proverbial kid in the candy store, I was able to roam freely in the North Carolina Archaeological Collection facility examining projectile points curated from across the state. Steve facilitated my search by guiding me to shelves where artifacts were stored when I requested the need to look at points from specific sites. In addition, Steve arranged for the high-resolution color photographs of the figures in chapter 5 to be archived on the RLA’s webpage. Finally, I am also indebted to Steve for his thorough editorial and substantive comments on a draft of this manuscript. No better colleague can a scholar ask for.

The above-mentioned trip to the RLA was made possible by a Research and Creative Activity Award from East Carolina University. East Carolina University students Parker Moore and Rachel Boyd are thanked for using their computer skills to generate the figures used here.

With respect to the publication process, I appreciate the encouragement of Chris Rodning, series editor for the Archaeology of the American South: New Directions and Perspectives, to pursue this project. Wendi Schnaufer, senior acquisitions editor for the University of Alabama Press, ably guided this manuscript to publication. Susan Harris copy-edited the volume.

Finally, as always, thank you, Becky, for your support and for freely giving me the most precious resource I needed to complete this book: time.

Introduction

IN A REVIEW OF JOFFRE L. Coe’s 1964 book, The Formative Cultures of the Carolina Piedmont, Alfred Guthe (1965) notes that Coe’s volume provides an important reference for Southeastern specialists. One can hardly imagine a more prophetic statement about an archaeological work. Since its appearance over 50 years ago, this volume has been reprinted seven times. Indeed, The Formative Cultures of the Carolina Piedmont (hereafter Formative Cultures) is a rare example of an archaeological report that is as popular with the general public as it is with professional archaeologists. For the former group, the abundant and clear photos of projectile point types are of interest. For the latter, that the artifact types had chronological significance is paramount. Coe (1964:9) was able to make temporal sense of a hodgepodge of projectile point types previously recovered from the surface or shallow plow-zone excavations based on excavation of three stratified sites along the Yadkin and Roanoke Rivers in the Carolina Piedmont. Taken together, a composite stratigraphic sequence was created that allowed Coe to associate particular chipped-stone points with various sub-periods of the state’s prehistory that today we know spans some 12,000 years. This result was unprecedented and stimulated similar stratigraphic excavations around the Southeast (e.g., Broyles 1971; Chapman 1977; Claggett and Cable 1982). Today, the Formative Cultures sequence is familiar to all southeastern archaeologists (Figure I.1). In short, the sequence begins with the Hardaway complex that includes a point type marked by a concave eared base form similar to Dalton points elsewhere in the Southeast. Subsequently, a series of side-notched and corner-notched points mark the Early Archaic in the Piedmont, followed by a series of various stemmed points representing the Middle and Late Archaic. Finally, the Woodland period is marked by a series of triangular points. Fieldwork in the Mountains (e.g., Dickens 1976; Keel 1976) and Coastal Plain (e.g., South 2005) subsequently added additional types appropriate to their respective regions of the state. Still, Formative Cultures remains the primary projectile point type reference for professionals and avocationalists alike in the Southeast. For the former, Coe’s work is the framework for many other statewide point typologies or regional publications of the Southeast (e.g., Cambron and Hulse 1975; Chapman 1977; Justice 1987; McGahey 2000). For the latter, popular publications like Overstreet (2015) also rely significantly on Coe’s publication.

As important as Formative Cultures is to southeastern archaeology, my several decades of examining tens of thousands of chipped-stone points has engendered questions about several aspects of Coe’s typology. This book is my attempt to deal with these questions and is grounded in the following issues.

Image: FIGURE I.1. Formative Cultures sequence from Coe (1964:121) (Reproduced with permission of the American Philosophical Association)

FIGURE I.1. Formative Cultures sequence from Coe (1964:121) (Reproduced with permission of the American Philosophical Association)

First, over the last 50 years some ambiguities have arisen with respect to some type definitions as originally proposed in Formative Cultures. For example, difficulties exist in distinguishing between Palmer versus Kirk Corner-Notched points (e.g., Chapman 1977; Claggett and Cable 1982; Daniel 1998). Similarly, haziness arises in differentiating some Guilford Lanceolate versus Morrow Mountain Stemmed points (e.g., Drye 1998; Goodyear et al. 1979; Gunn 1993). As I discuss later, Coe tended to describe rather than define point types, which also created some typological ambiguity. Consequently, I review all the point types defined in Formative Cultures and, when appropriate, suggest clarifications to point type definitions. I have also created some new types to account for unexplained variation that I see among points.

Second, the last five decades have witnessed the seemingly random addition of several point types to Coe’s cultural-historical sequence including various fluted point types (e.g., Clovis and Redstone), bifurcate (St. Albans and LeCroy), notched (Rowan/Big Sandy), and stemmed types (Small Savannah River and Thelma). Many of these types have been adopted in a haphazard fashion without much consideration for what their inclusion means for the state’s culture-history (particularly with respect to the Archaic period). In this book, I review how these types came to be recognized in the state and evaluate their validity.

A third issue, a reevaluation of the Formative Cultures cultural-historical sequence, follows from the above two issues. Coe saw the various point types in the Formative Cultures sequence as manifestations of either cultural change or continuity. I reconsider this claim based upon new data and updated ideas regarding the notions of projectile point types and traditions and propose some significant revisions to the sequence, primarily with respect to the Paleoindian and Archaic periods.

This book, then, is the outcome of my thinking in regard to those issues and is organized as follows. Chapter 1 provides some historical context for understanding the origins of the Formative Cultures point typology. In chapter 2, I introduce the concept of communities of practice as a framework to propose several ancient point-making traditions whose constituent types reflect group identities (e.g., McElrath et al. 2009; Pauketat 2001; Sassaman and Rudolphi 2001). Chapter 3 provides a summary of the stone types commonly used to manufacture the various point types found in North Carolina. North Carolina is unique in the Southeast with respect to the occurrence of abundant knappable metavolcanic stone that was the primary tool stone utilized in and around the Carolina Piedmont. Chapter 4 is the heart of the book. In it, I review several dozen North Carolina point types. This discussion includes refining most existing type descriptions, questioning the validity of others, and proposing several new ones. Chapter 5, then, places these proposed point traditions in a revised sequence based upon the practice-theory inspired framework outlined in chapter 2. Finally, I emphasize that this book is written for both professional and avocational archaeologists. I explicitly address both audiences in chapter 6, which is revised and updated from an article (Don’t Let Ethics Get in the Way of Doing What’s Right: Three Decades of Working with Collectors in North Carolina) originally published in North Carolina Archaeology (Daniel 2016). Its premise is that under the right circumstances private collections do have scientific value, and I outline the benefits and challenges I have encountered during some three decades of using those collections. If we are to maximize those benefits and minimize challenges, both professional and avocational archaeologists need to continue to find common ground, and I present this book as a small step to further that goal.

1

Revisiting the Formative Cultures Typology

I BEGIN NOT WITH ARTIFACTS but with an abstraction—the type concept. Artifact classifications are the foundation of any archaeological report and are based upon archaeological types—the basic unit of archaeological classification. In archaeology, classification provides order on the vast numbers of material remains present in the archaeological record, primarily in the form of artifacts. It would be incredibly tedious to describe every item recovered from an archaeological site (and even more tedious to read a report including such descriptions), so archaeologists group like items into categories that become proxies for the artifacts themselves. Archaeological types are usually created based on some central tendency present in a group of objects. That is, types are usually defined based upon one or more shared characteristics (often called attributes by archaeologists) that tend to minimize differences within the type while maximizing differences between other types. The archaeological literature abounds with references regarding classification, particularly of stone artifacts (e.g., see Andrefsky 2005; Odell 2004).

For our purposes, it is relevant to point out that type is a deceptively simple term that has a history of controversy in archaeology (e.g., Ford 1954a, 1954b, 1954c; Spaulding 1953, 1954a, 1954b). In this chapter, I outline the history of the type concept as it relates to the types presented in Coe’s Formative Cultures. I then address two related questions. How were the point types in Formative Cultures created, and how were they interpreted?¹ Understanding those issues provides the background necessary to understand the typological revisions I propose in subsequent chapters.

BACKGROUND TO COE’S TYPES

The artifact typology in Formative Cultures has its roots in the simple artifact typologies of American archaeologists in the nineteenth century that focused on describing artifacts based upon form and raw material (Dunnell 1986; Willey and Sabloff 1993:38–95). In fact, chipped stone points dominated some of the earliest classifications in the United States (Dunnell 1986:154–61). This was the case for two related reasons. First, ethnographic accounts made their identification as artifacts clear. Such accounts outlined how these weapons were made and used and hence made them an intuitively appealing functional group for classifying. Second, stone points were the most common complete artifact that was ubiquitous across North America. While pottery sherds were equally common in some areas, they received much less attention because they were broken and thus less interesting. Thomas Wilson’s (1899) Smithsonian Institution publication of Arrowheads, Spearheads, and Knives of Prehistoric Times serves as an early example of this form of classification. Chipped-stone points from across the United States were classified under one of four divisions based on artifact outline (leaf shaped, triangular, stemmed, and peculiar) with various classes as subcategories primarily based on blade shape (e.g., shouldered, shouldered and barbed, beveled edges, serrated edges). In the same way that the presumed function of these artifacts made them appealing to collect, artifact shape appeared to be an intuitively satisfying way to categorize them. But such a classification served little archaeological purpose as points from around the country were classified together without any regard to time or culture area. In large part this was due to the absence of any appreciation of the time depth or cultural diversity present in Native American groups.² Such knowledge increased relatively rapidly, however, and by the early twentieth century, archaeologists in North America began to recognize both the antiquity and regional cultural adaptations of humans in the New World. In particular, archaeologists in the southeastern United States found that certain artifact styles had historical significance that allowed archaeologists to tell time—at least in a relative way (see Dunnell 1971, 1986; Willey and Sabloff 1993:164–69). It was then that archaeologists’ attention shifted from classifying stone artifacts to pottery. In part because pottery sherds were common on archaeological sites and the fact that the materials and methods of pottery making were fundamentally different from that of manufacturing chipped-stone tools, ceramic types were initially more useful than stone tool types as temporal markers.³ But while this realization was quite useful to archaeologists, it begged a larger question about which there was some debate (Ford 1954a, 1954b, 1954c; Spaulding 1953, 1954a, 1954b). Were