Stone Mound Investigations as of 2009*

             Norman E. Muller


For the past twelve years, I have been deeply involved in the study of aboriginal stone constructions. Others, such as Fred Werkheiser of Bethlehem, Pennsylvania, who first introduced me to this fascinating subject, have been at this much longer. The field has a small but enthusiastic group of researchers, most of whom regularly post photographs and commentaries of what they have found on their numerous excursions. Peter Waksman ( comes to mind as one who has been at this the longest. Others who maintain interesting websites and blogs are Doug Schwartz (, James and Mary Gage (, Tim MacSweeney (, Larry Harrop (, David Schewe (, and Kathy Klopchin (

   All this activity has been fostered by the availability of inexpensive digital cameras, and by rapid communication and posting of images via the Internet. Whereas before a smattering of sites was fairly well known to active researchers, now there are dozens, and the list keeps growing. By categorizing certain features, as Harrop has done on his website, we now know that certain types of carefully made stone structures, such as platform cairns, are widespread throughout the Northeast. Although many professional archaeologists still claim they were constructed by colonial farmers, the distribution of these visually impressive features, and the fact that they appear to have no agriculturally related function or any mention in the historical record, suggest that they have a significance we are not yet aware of, and that we need to look at them and other unusual stone constructions and accents more closely and seriously.

    There was once the strongly held belief that our Indian forebearers lived lightly on the land and left no visible permanent record of their existence. For the most part that is true, but if one takes a stroll through the rocky terrain of Rhode Island, as Larry Harrop has done repeatedly over the past five years or more,  one will encounter a bounty of impressive above-ground stone constructions—artifacts really—most of which were probably made by American Indians over the thousands of years they occupied the land. Some of these, such as split-wedged boulders, are like a signature or a sign that one is entering terrain where one is bound to find additional examples of Indian stone handiwork. Split-wedged boulders have been found from Minnesota (Figure 1) to Georgia (Figure 2) and up to New England (Figure 3), attesting to their widespread aboriginal origin.

Research in the Northeast

  Unfortunately, progress in answering the question of who constructed the mysterious stone features of the Northeast took a detour when William B. Goodwin published a rambling book in 1946 on the stone ruins supposedly built by Celtic explorers to New England.[1]  Goodwin compared the construction of some stone features in New England with Celtic ruins in Ireland, concluding that the New England examples were constructed by Celts around the eighth century. Goodwin’s book was followed up several decades later by several popular books on much the same theme by Barry Fell, a retired Harvard University marine biologist.[2]  Fell got no closer to proving his case than Goodwin, and both incurred the wrath of the archaeological community, which concluded that all of this was simply “fantastic archaeology,” meaning tales of fiction without much value.[3] James Mavor and Byron Dix in their book Manitou, published sixteen years after Fell’s last book,[4] took a different view of the stone ruins in New England and pronounced them to be American Indian, but their emphasis on celestial and astronomical alignment of walls and chambers, while appealing to more reasonable people, still failed to impress the archaeological establishment. Even now, serious and even-handed researchers are frequently lumped together with those who espouse a more radical interpretation of our ancient past. A good example of where the situation stands is found in a curious and rather unpleasant commentary tucked at the back of the newsletter Terra Firma #5, published in 2007 by the Massachusetts Department of Conservation and Recreation’s Historic Landscape Preservation Initiative. In a book titled “The Last Word: Debunking the Myth of Stone Walls, Piles and Chambers,” the anonymous author wrote the following:

“The stone walls built by New England farmers helped define property lines, divide fields, woodlots and pastures, and shape animal pens. Coincidently, the walls may match cardinal compass points or celestial phenomena—but for practical purposes rather than sacred. It was also common to construct cold cellars and pile surplus rocks within pastures for later use or sale.

“Some have suggested a Native American origin for these features. There is no archaeological evidence to support this conclusion. When historians and archaeologists have researched stone walls, piles and chambers, they have invariably demonstrated that these features are associated with the activities of European settlers and have no Native American (or other) origin. In addition, Native American advisors have been involved in a number of excavations and have confirmed these findings. However, archaeologists do find stone features on Native American sites, hearths for example. But they rely on context to make the determination where, during a controlled scientific excavation, archaeologists analyze the entire site, all artifacts associated with the feature, and its placement in the soil. This provides the cultural and geological context needed to interpret and date the entire site.

“Archaeologists also consider ethnographic and ethnohistorical information. For example, Native American oral traditions record that people did place small stones or twigs on a sacred spot as they passed by. Over time this might result in a small pile of pebbles, tiny cobbles, or sticks, but not large piles. Conversely there is a strong, documented ethnohistory of stone building traditions among the European settlers of Massachusetts. Together, archaeology and ethnohistory provide conclusive evidence that stone walls, piles and chambers are not the work of ancient cultures.” [5]

  What the author of “The Last Word” did not realize in his or her myopic view of possible Indian stone construction, is how much progress has been made not only in New England, but specifically in the South over the past forty years or so in determining whether stone piles and walls are colonial or aboriginal. A review of some of the documents discussed below could give added impetus in moving discussion away from a total denial of Indian manufacture for certain walls and cairns to something approaching open-mindedness.

  As one who has researched stone cairns, walls and other unusual man-made or altered stone features in the Northeast and beyond for more than a decade, I can say in all truthfulness that I have no solid factual evidence establishing who made or altered these features, when they were made, or what purpose they served. I have argued in several articles that by comparing certain types of features to others having similar morphology, a pattern of design can be established, one that suggests a commonality of purpose.

  Such an argument was presented in an article I wrote about the Oley Hills site in eastern Pennsylvania,[6] where the pattern and distribution of walls and cairns seemed to focus on a huge boulder or stone tor on the summit of a hill, which I concluded was the main reason for the construction of the wall and cairns. The placement and focus of these features on the landscape seemed clear and convincing, but not to some readers of the article, presumably because no hard evidence was presented supporting the main thesis.

  After the article was published and critical comments were received and digested, I concluded that however careful one is in formulating an argument based primarily on empirical evidence, if no hard evidence in the form of potsherds, charcoal for radiocarbon dating, or other cultural debris found in an archaeological context is presented supporting one’s viewpoint, one is simply left with an argument that is difficult, if not impossible, to justify. Since the point of this research was intended to advance knowledge, then it was simply not good enough. If we are to move forward in answering the age-old questions of, “Who made it?, When was it made?, and Why?” we must embrace archaeological methodology, invite archaeologists to work with us, and move toward a common goal of cooperation and understanding.

Stone Wall and Stone Mound Research in the South

For reasons that are not entirely clear, for decades archaeologists in the South have been more open to the idea that Indians were responsible for some of the stone features found there. Part of this may be due to the fact that when the region was first settled, pre-existing stone features on the landscape, such as the curious serpentine wall near the summit of Fort Mountain in northwest Georgia, were recorded (Figure 4). [7]

  Then there were mavericks such as Arthur R. Kelly, legendary archaeologist and founder of the Department of Anthropology at the University of Georgia in 1947, who had an open mind about cairns and other stone features he had come across. Kelly had long been interested in unusual stone walls and enclosures he had encountered. He encouraged Philip E. Smith, a graduate student in archaeology at Harvard University, to undertake a survey of aboriginal stone constructions in Georgia and surrounding states. Smith’s report, published in 1962, was titled “Aboriginal Stone Constructions in the Southern Piedmont,” and was based on a survey conducted over two summers in 1955-56.[8] At the end of his article, Smith concluded that the walls he studied seemed to tie together large boulders and outcrops, and in certain instances made deliberate detours to connect with boulders (Figure 5). Because of this, Smith raised the possibility that the main reason for the walls was to link “certain impressive natural phenomena such as boulders or bluffs which may have held some religious, symbolic or animistic significance to the people concerned.” [9] Smith’s seminal article has been an inspiration to many researchers, including myself.

  Finding no cultural artifacts during his two year investigation, Smith recognized the weakness of his research, in that it was ultimately based on hunches and not on hard evidence, and that he was no closer in getting to the real meaning of the structures at the end of his research than he was at the beginning. But by presenting the comparative evidence that he had accumulated, he hoped that it would serve as the “jumping-off point for future investigation.”
Hard research would come not from the examination of walls and enclosures, but from stone mounds, since they often represented burials. They also presented problems of cultural origin and date ever since they were first encountered. Serious research on stone mounds in the South, and whether or not they were aboriginal or colonial, began in 1978 with the publication of two articles: one by Fish, Fish and Jeffries,[10] and the other by Jeffries and Fish.[11] The former had a section titled “The Stone Mounds: A Need to Explore an Archaeological Mystery.” Beginning with a survey of historical references to stone mounds, the authors concluded with the statement: “Previous research concerning stone mounds has demonstrated that their nature and origin cannot be satisfactorily determined using surface appearance and location as the sole criteria.” [12]

  That most of the cairns in the South were probably colonial or later was the subject of a study by Thomas Gresham in 1989,[13] which was an indirect criticism on an archaeological survey of the Parks-Strickland cairn site in Gwinnett County, Georgia, undertaken by Patrick Garrow and David W. Chase in 1988.14 Garrow and Chase concluded that the dozens of cairns on the site were probably not historic because of five factors:
1. The mound complexes are located on the only level ground in the vicinity, which would have been most suitable for farming.

2. The mounds straddle a property line but are also perpendicular to this line, whereas historic stone piles are almost invariably parallel to property lines to minimize the amount of area wasted to stone disposal.

3. The land was never cultivated, and the nearest cultivated land is separated from the mound complexes by steep slopes.

4. The mounds were carefully planned and constructed. “There is simply no logical reason why a farmer would have expended that much labor to construct stone mounds that had no other conceivable function.”

5. Mounds were not randomly placed and occur at equal distances from each other on parts of the Strickland Mound Complex.[15]

   Because of its seemingly over reliance on observation and logic, Gresham would not accept Garrow’s conclusions,[16] and thus began a long drawn-out controversy between the two protagonists, with Gresham arguing for archaeological digging to determine the cultural origin and date of the mounds, and Garrow repeating his arguments based mainly on a visual and logical analysis of cairn sites.[17] Gresham countered with an archaeological survey of the Strickland Tract, during which he opened five cairns and found cultural artifacts from the nineteenth century in two of them. On the basis of this, he concluded that all the cairns at the site were probably historic, even though the other three did not produce any definitive diagnostic material.[18]

  Within the past seven years, a few careful excavations, several of which have occurred at a single site, have resulted in one cautionary tale, important observations, and discoveries. In 2002, Johannes Loubser and T.G. Grenier produced a report on a site in Track Rock Gap in north central Georgia.[19] This site, on the east side of the Gap, consists of dozens of terrace walls and cairns that had been earlier discovered by Carey Waldrip, a resident of a nearby town, Blairsville (Figure 6). Loubser dated one wall by removing soil underneath it and having it tested using oxidizable carbon ratio (OCR). The date obtained was ca. 1075 b.p.[20] Nearby, a 4.5m x 3m x .7m stone mound on a high promontory was excavated. Two column-shaped rocks were lying on the south side of the pile which Loubser interpreted might once have “stood upright in a monolith fashion.” After removing the top layer of stones, Loubser uncovered diagnostic artifacts underneath that confirmed the accuracy of the OCR date for the wall.

   A good example of the problems and potentialities in investigating stone mounds is found in two reports that concern a stone mound site in Greene County, Georgia. In 2004, Jerald Ledbetter, working for Southeastern Archaeological Services, Inc. conducted a survey of site 9GE2084, which consists of a concentration of rock piles in Greene County, Georgia, 14km south-southwest of Greensboro, and located on a knoll above a creek.[21] Nineteen distinct rock piles were identified, five of which were labeled A-E. Based on a horseshoe recovered from one of four shovel test pits, Ledbetter initially proposed that all the cairns at the site represented historic field clearing piles. But critical comments by several reviewers stating that his conclusion was not backed up by sufficient evidence, made him return to the site for Phase II, during which he and his staff undertook a more comprehensive excavation of the other cairns.[22]

  Excavations during Phase II in 2005 were much more comprehensive, in that an additional 37 shovel test pits were excavated across 9GE2084. Eight of the test pits yielded artifacts. A trench measuring 4m long by 1m wide was cut through Pile C, and in that trench they uncovered human teeth and bones, plus ancient pottery fragments attesting to an Indian burial. Additional cultural remains were uncovered in Pile E, but not in Piles A and F, which were thought to be historic in date. Ledbetter also employed the services of a geomorphologist, David Leigh, Ph.D., and concluded that:

“The prehistoric rock mounds were found to have a finer overall particle size and larger proportion of a fine (<2mm) organic and mineral matrix material between the stones. Pebble counts…demonstrate that the prehistoric mounds have a smaller average particle size when compared to the historic mounds…. The historic mounds contain few rocks smaller than 5cm, whereas <5cm cobbles are common in the prehistoric mounds. Leigh concluded that the smaller rock size, fine organic-rich matrix, and lack of plow zones beneath and around Mounds C and E are all convincing pieces of evidence that support the prehistoric age assignment.” [23]
Having examined the piles in greater detail than during the initial excavation, Ledbetter concluded, under the section “Revised Conclusions and Recommendations,” the following:

“Our original interpretation that site 9GE2084 was composed of an extensive scatter of agricultural rock piles is clearly in error. The site contains one rock mound of confirmed prehistoric origin (Rock Pile C) and a second of probable prehistoric construction (Rock Pile E). Excavation data from two other piles (Rock Pile A and Rock Pile F) provided data comparable to that used by other researchers as evidence of historic construction. Characteristics used to identify historic piles include size and shape (relatively small size of pile and elongated appearance suggestive of a wagon load), construction with larger rocks (Rock Piles A and F lacked the smaller rocks found in Rock Piles C and E), placement of the piles on a ground surfaces that appears to have been cultivated, absence of prehistoric cultural remains beneath the pile. Other more subjective observations of more recent construction include piles that are relatively high in proportion to width, rocks that are loosely packed with little or no soil accumulation between rocks, and little evidence of weathering, discoloration or staining resulting from long term moss or lichen cover. If correct in our current interpretations, 9GE2084 contains rock piles constructed during both prehistoric and historic times.” [24]

Because a residential development was slated for the area around the site, additional excavations were suggested to determine whether any of the other cairns were prehistoric. Johannes Loubser was the principal investigator of this Phase III excavation.[25] His observations confirmed the following:

“Overall then the message of this brief tour to other geographic areas is that a look below the surface is imperative; materials remain and behaviors are not always what they appear to be on the outside. For this reason then it seems unlikely that relying solely on surface inspections or on our own pre-conceptions would yield conclusive answers; excavation of one kind or another, together with consultation of ethnographic sources, are currently the only ways to properly assess rock piles.”

Finally, under the heading “4.1 Research Questions and Design” Loubser concludes with:

“Straightforward typologies, based on surface criteria, such as location, size, morphology and layout, have not always been conclusive or convincing [Loubser then refers to reports by Garrow (1994) and Gresham (1990) among other authors]…Only an investigative strategy that involves some kind of sub-surface sampling can address questions pertaining to the stratigraphic association, age, physical and chemical makeup, and cultural affiliation of the rock piles and associated soil deposits…Stratigraphic association can reveal a lot about the antiquity of a rock pile, particularly the vertical association between a rock pile and a plow zone. Plowing is a special perturbation that effectively homogenizes the included soils…If the bottom of a pile lies on top of or within a plow zone, then the rock pile is historic in date. If, however, the rocks penetrate well below the plow zone, especially into deposits with prehistoric artifacts or charcoal with prehistoric radiocarbon dates, then the rock pile is most probably prehistoric in date.”

Written and photographic documentation of cairns, walls and other unusual and impressive stone features, in addition to historical research and deed search, should certainly continue to be done. This is basic to any research, even archaeological. And, unfortunately, cairns and other man-made stone features that rest on bedrock or boulders cannot at this time be accurately dated by any means currently known. But for those cairns and stone piles that rest on soil, the examples above can serve as a guide to finally answering who built these perplexing stone features, and when.

*This article was previously published in the  NEARA Journal, volume 43 number 1, Summer 2009, p.17


[1]  Goodwin, William B. The Ruins of Great Ireland in New England. Boston 1946.

[2]  Fell, Barry. America B.C., NY 1976; Bronze Age America, NY 1982; Saga America, NY 1983.

[3] Williams, Stephen. Fantastic Archaeology: The Wild Side of North American Prehistory. Philadelphia 1991.

[4] Mavor, James W., Jr. and Byron M. Dix. Manitou: The Sacred Landscape of New England’s Native Civilization. Rochester, VT, 1989.

[5] The author obviously failed to mention the wall under the drip line of the Flagg Swamp Rockshelter in Marlborough, Massachusetts, that was excavated by archaeologists from Harvard University in 1980, who dated it to 4750 B.P. Unfortunately, it was in the way of a planned off ramp to a highway, and as John Hanson Mitchell relates in his marvelous book Ceremonial Time: Fifteen Thousand Years on One Square Mile (New York 1984, 64), “the construction crews returned, holes were drilled in the rock face, and five thousand years of history was dynamited into oblivion.” Another omission was the large commemorative Indian stone pile below Monument Mountain in Stockbridge, Massachusetts, that was sketched by Ezra Stiles in 1762 (the drawing is reproduced in Eva L. Butler, “The Bush or Stone Memorial Heaps of Southern New England,” Bulletin of the Archaeological Society of Connecticut (April 1946), 2-12). Although the present mound is not thought to be the one drawn by Stiles, such a large commemorative stone mound belies what the anonymous author wrote in the statement.

[6] Muller, Norman E. “Accenting the Landscape: Interpreting the Oley Hills Site,” Chapter 10 of The Archaeology of Semiotics and the Social Order of Things, George Nash and George Children, eds., BAR International Series 1833, Oxford 2008, 129-140.

[7] The earliest reference to the wall was made by George White, Statistics of the State of Georgia, Savannah, 1849, 442. This is also mentioned in P.E. Smith, “Aboriginal Stone Constructions in the Southern Piedmont,” University of Georgia Laboratory of Archaeology Series, Report No. 4, Part II, Athens, Georgia, 1962, 10.

[8] The article is found online at

[9] Smith, P.E., 34-35.

[10] Fish, Suzanne K, Paul R. Fish and Richard W. Jeffries. “An Examination of Interfluvial Settlement in the Georgia Southern Piedmont: The Georgia Power Company Plant Scherer Archaeological Survey,” University of Georgia Laboratory of Archaeology Series Report No. 15, (1978). Available online at

[11] Jeffries, Richard W. and Paul R. Fish. “Investigation of Two Stone Mound Localities, Monroe County, Georgia,” University of Georgia Laboratory of Archaeology Series Report No. 17 (1978). Available online at

[12] As in n6, 36.

[13] Gresham, Thomas H. “Historic Patterns of Rock Piling and the Rock Pile Problem,” Early Georgia 18 (1989), 1-40. Available online at

 [14] Garrow, Patrick H. and David W. Chase. Archaeological Investigations of Two Stone Mound Complexes in Gwinnett County, Georgia, 75 page report published under the auspices of Garrow & Associates, Inc., Atlanta, Georgia (1988).

[15] Ibid, 52-53.

[16] Gresham (as in n.9), 24-25.

[17] Garrow, Patrick H. Archaeological Investigations at the Headwaters of the Apalochee River: An Intensive Survey of the Dacula Tract, Gwinnett County, Georgia. Garrow & Associates, Inc., June 1990. Garrow, P.H., The Gwinnett Stone Mounds. 31 page report published under the auspices of Garrow & Associates, Inc., Atlanta, Georgia (1994). This controversy was also briefly discussed in N. Muller’s “The Cairns in Our Midst: Historic or Prehistoric,” NEARA Journal, 37 (2), Winter 2003, 7-8.

[18] Gresham, Thomas H. Archaeological Investigations of the Strickland Tract, Gwinnett County, Georgia, Report prepared for Braden & Associates, Norcross, Georgia, by Southeastern Archaeological Services, Inc., Athens, Georgia (1994), 39pp.; Gresham, T.H. Research Design for Determining if Graves are Present on the Braden Tract, Gwinnett County, Georgia. Submitted to Braden & Associates, Inc., One Meca Way, Norcross, Georgia 30093, November 21, 1994 by Southeastern Archaeological Services, Inc., Athens, Georgia.

[19] Loubser, J.H.N. and T.G. Grenier. The Archaeological Testing of Stone Features at 9Un367, New South Associates, Stone Mountain, Georgia. Report Submitted to Georgia Forest Watch, Track Rock Gap (2002)

[20] Ibid, 6. Oxidizable Carbon Ratio dating was done by Douglas Frink of the Archaeology Consulting Team of Essex, Vermont. The laboratory has been disbanded and OCR dating is no longer being done.

[21] Ledbetter, Jerald. Archaeological and Historical Investigations of the Georgia Pacific & Hardin Tracts, Greene County, Georgia, Prepared for Reynolds Plantation, Greensboro, Georgia by Southeastern Archaeological Services, Inc. (2005).

[22] Ledbetter, Jerald, K.T. Burns, T.G. Gresham, Scott Jones, D.S. Leigh, W.G. Moffat and L.D. O’Steen. Archaeological and Historical Investigations of the Georgia Pacific & Hardin Tracts, Greene County, Georgia (with Addendum). Southeastern Archaeological Services, Inc. Athens. Prepared for Reynolds Plantation, Greensboro, Georgia (2006): 371.

[23] Ibid, 370.

[24] Ibid, 371.

[25] Loubser, Johannes, S. Butler and J. Page. Data Recovery at Site 9GE2084: The 19 Rock Pile Site. Greene County, Georgia, Brodington and Associates Report submitted to Reynolds Plantation LLC, Greensboro, Georgia (2008)