Experimental Archaeology with Scrapers: Scrape, Scrape, Scrape

My Name is Alaina Helm, I interned with the Section of Archaeology during the summer of 2009, and I am a Junior at Oberlin College in Ohio. You may already have seen my other posts about previous projects I have worked on: lithic analysis of Kings Quarry (36Lh2) and refitting debitage from Eelskin Rockshelter (36Bu159). This post is about another project I completed this summer doing experimental archaeology on end scrapers under the direction of Dr. Kurt Carr, Senior Curator, The Section of Archaeology at The State Museum of Pennsylvania.

                Prehistoric peoples demonstrate a preference for different lithic material types during different time periods. Paleoindians (10,000 to 12,000 years before present) preferred jasper and chert for making stone tools, despite inhabiting areas in closer proximity to alternative materials such as argillite and metarhyolite. During the transitional period (2800-4300 years before present) argillite and metarhyolite were intensively used throughout the Mid-Atlantic region.  With this knowledge in mind, we wanted to perform an experiment regarding the lithic composition of end scrapers to determine if there is a reason for biases towards certain lithic materials in the archaeological record.  Prehistoric peoples demonstrated a preference towards jasper scrapers despite being in closer proximity to other sources of useable material such as argillite. To determine if there was a functional reason for obtaining different materials from farther afield, we made scrapers of several materials and underwent experimental scraping with them. The goal of the experiment was to observe variations in wear patterns and effectiveness in scraping pieces of wood by using different materials.

Scrapers made for our experiment from various materials. Each scraper was assigned an alphanumeric designation for tracking purposes.

                Before beginning our experiments, I researched the literature to see if anyone had performed and written about a similar experiment. Although numerous articles have been published about use wear on scrapers, none of the articles compared wear between various lithic materials.  Our experiment consisted of several scrapers of varying materials created for the experiment by expert flint knapper Steve Nissley. The materials used were argillite, metarhyolite, jasper, quartzite, Normanskill chert, and Onondaga chert. All scraping was done on soft wood because it is easier to acquire than hide and would more quickly produce wear because it is a harder material.

Before being used, the scraper was hafted by channel lock pliers.

                The experiment was performed by hafting an end scraper using pliers padded with softened rawhide. The tools were then used in increments of 500 scrapes with a stroke length of thirty-two centimeters. The number of scrapes were carefully counted, and stroke length and strength was kept as uniform as possible to ensure consistency. Two sets of scrapers were used; one set was used by a variety of people including museum staff and volunteers, and the other set was used by only me. Having scrapers used by several people allowed more scraping to be performed faster without limitations caused by fatigue. Because several thousand scrapes needed to be performed for the experiment, having a separate set used by only one person allowed for a controlled comparison. The scrapers were photographed from multiple angles and at multiple magnifications using a Dino-Lite digital microscope with the highest resolution images at around 200x magnification. The scrapers were also measured using digital calipers at designated reference marks drawn on the scraper for consistency. All measurements and photographs were taken before the scrapers were used and at regular intervals of scraping to ensure a consistent record of wear on each scraper.

Alaina takes measurements and photographs of the experimental scrapers.

                The high-resolution images revealed that Argillite and Metarhyolite seemed to wear down faster with more visibly rounded edges than the Normanskill and Onondaga cherts, quartzite, and jasper. The chert scrapers showed a higher level of effectiveness than the jasper and the quartzite scrapers. Effectiveness was gauged by measuring the depth of the gouge each scraper created after the same number of scrapes. The argillite and metarhyolite scrapers shallower gouges than the jasper and chert scrapers, and the jasper scraper was slightly less effective than the chert scraper. These results suggest that the reason cherts and jaspers were the preferred materials for scrapers was due to their increased effectiveness in comparison with materials that may have been easier to obtain.

 

Argillite scraper with no wear (top) and after 500 scrapes (bottom).

This experiment was an interesting way to learn about lithic wear and get hands on experience with experimental archaeology. It allowed me to experience the nuances of designing an experiment and the difficulties in separating wear in differing lithic types. I learned a lot about aspects of experimental archaeology that are often not fully appreciated without the experience to back it up. This will help inform the way I approach any similar projects in the future. For example, on paper, scraping something 1000 times does not seem to be much until you realize that the individual scraping will need breaks. It is nice to occasionally switch up activities as well to make such experiments endurable.  I hope that my time with the Section of Archaeology at the State Museum of Pennsylvania will provide me with insight and experience during the planning and preparation of future research and will help to form a foundation on which I can further add to the results of our research.

Upcoming Pennsylvania archaeology events:

September 14, 2019 - Delaware County Archaeology Day at Newlin Grist Mill in Glen Mills, PA

This festival features a full day of hands-on activities. Visitors will be able to work with professional archaeologists and assist with three different excavations. An archaeologist from The State Museum of Pennsylvania will be on hand to answer questions.

September 15, 2019 - Fort Hunter Day at Fort Hunter Mansion and Park in Harrisburg, PA

Archaeologists from The State Museum of Pennsylvania will be conducting excavations in the mansion’s back yard during the park’s annual fall festival celebrating the old-time ways of life. Since 2006 archaeologists have been documenting archaeological evidence from the past occupations at this site dating from approximately 9000 years ago to the present day.

November 9, 2019 – 2019 Workshops in Archaeology at The State Museum of Pennsylvania in Harrisburg, PA

Don’t miss your opportunity to learn about the prehistoric people of western Pennsylvania that we call the Monongahela Indians. This theme will be featured at the 2019 Workshops in Archaeology hosted by the Archaeology Section at The State Museum of Pennsylvania. The program will take place on Saturday, November 9, 2019 at the museum.

For more information, visit PAarchaeology.state.pa.us or the Hall of Anthropology and Archaeology at The State Museum of Pennsylvania .

Free-standing Keyhole Structures and the Late Woodland Period

Excavation and recovery of the burned keyhole structure.

A curious archaeological feature of the Upper Ohio and the Susquehanna Valley of Pennsylvania is the semi-subterranean free-standing keyhole structure (Smith 1976). Its general two - dimensional shape can be easily compared to the keyhole of a door lock (Figure 1). Attributed to the Late Woodland Period (circa AD. 1000 – 1550) the keyhole structure has three principal parts: the body and rock filled pit connected by a trench that is usually an upward ramped tunnel that extends to the main body (Figure 2).  One or more rows of postmolds line the outer-most edge of keyholes the exception being  the opening around the rock filled pit that remained open for entry and exit purposes. Although the above ground three - dimensional aspect of this feature type no longer exists due to the ravages of time and the elements of nature, certain clues survive in the archaeological record that provide us with a glimpse of their architecture and probable function.

Figure 1.  Comparative shapes of a keyhole feature and the keyhole of a door lock.   

Figure 2. Generalized cross-section of a Late Woodland keyhole structure.

The floor of the body is often covered with carbonized material consisting of a flattened layer of charred grass thatch overlain by burned sections of saplings that connect with the post-lined pattern of postmolds encircling the wall of the body. Occasionally these are cross - configured suggesting that the saplings were inserted into the ground, then bent inward to form an arbor or igloo - like superstructure over the body and ramp. Slabbed charred bark overlying the thatch and saplings present inside many of the keyhole structures indicate that an outer layer of bark was installed to insure a weather tight shell from wind, rain and snow. Some examples show the presence of a relict drainage trench around the inside edge of the body (Figure 3).

Figure 3. Excavation plan of a keyhole structure showing the drainage trench.

                The analysis of more than seventy Late Woodland keyhole structures from the Unglaciated Plateau of northcentral Pennsylvania (Herbstritt 1995) has yielded clues as to their function.  The long axis of these structures is oriented along a northwest, west and southwest line that is also the direction of the prevailing winds in this region. The insulated walls serve as a protective barrier against these winds. A curious modification of the free-standing keyhole structure is the keyhole compound presently known only from those in the Unglaciated Plateau. The architectural design of this unique type of keyhole feature is its incorporation of up to five keyholes joined to a common rock pit (Figure 4).

Figure 4. Artist’s conception of a keyhole compound.

Fire altered rocks found in one or more areas of the keyhole structure is another clue to their possible function as a sweat lodge. Smith (1976) notes that rocks were commonly piled inside the main body, set to one side near the wall making for easy access to and from the semi-subterranean pit. The excavation of other keyhole structures in the West Branch of the Susquehanna Valley contained abundant fire altered rock scattered along the ramp joining the main body. These rocks always show evidence of reddening on their fractured surfaces. Based on the contents of the keyhole structure identified at the Fisher Farm site (36Ce35) in Centre county, Hatch and Daugirda (1980) postulated that the feature functioned as a smoke house for the curing of food stuffs.

Figure 5. Experimental reconstruction of the keyhole structure.

Results of experimental research (Figure 5) on the, burning (Figure 6) and excavation of a reconstructed  keyhole structure (see top), suggests that such features were used as a food storage facilities (Herbstritt 1995). The interior temperature of the structure could have been controlled over extended periods of time during warmer conditions. Information from the experimental work and follow-up excavation of the reconstructed keyhole demonstrated that the archaeological evidence alone could not provide the necessary information revealed through the experimental reconstruction. All said, given the present state of knowledge it may be assumed that semi-subterranean keyhole structures were likely multi-functional as their uses probably varied over the course of the year from sweat lodges to smokeries to food storage structures.

Figure 6. Burning of the keyhole structure.

We hope that you have enjoyed this brief presentation on keyholes from the archaeological and experimental perspectives. Learning from the past through archaeological investigations or experimental archaeology is important to understanding and appreciating our cultural heritage. Please help us to preserve the past by collecting responsibly and respecting our preservation laws.  Do join us next time when we will again bring you yet another fascinating topic on This Week in Pennsylvania Archaeology. 

REFERENCES

Hatch, James W. and Joyce Daugirda

1980       The Semi-Subterranean Keyhole Structure at Fisher Farm – Feature 28. The Fisher farm Site: A Late Woodland Hamlet in Context. Edited by James W. Hatch. Occasional Papers, No.12. The Pennsylvania State University, Department of Anthropology.

Herbstritt, James T.

1995       Reliving Prehistory: The Experimental Archaeology of a Keyhole Structure. Paper presented at the 66th Annual Meeting of the Society for Pennsylvania, Inc. Morgantown.

Smith, Ira F.

1976       A Functional Interpretation of Keyhole Structures in the Northeast. Pennsylvania Archaeologist 46(1-2): 1-12.

For more information, visit PAarchaeology.state.pa.us or the Hall of Anthropology and Archaeology at The State Museum of Pennsylvania .

Houses of the Past

Artist rendition of a Monongahela Foley Farm phase village

Humans, world over, used some form of shelter to insure survival. The type of shelter is normally influenced by a combination of geographical, environmental, ecological and social conditions though other mitigating factors also play into the equation. During the Late Woodland and Late Prehistoric periods A.D. 700-1550, native peoples living along the rivers and streams of Pennsylvania employed certain types of architecture that changed little over time. Archaeology is the principal tool that scholars employ to decipher the past, and it is precisely archaeology i.e. postmolds that contribute to our understanding of Native American architecture and settlement patterns. Archaeological sites linked to these periods, yield clear evidence of the type(s) of houses and their arrangement within a settlement. Anthropologically, this is fundamentally understood as the community pattern. Observed differences in community patterning can be demonstrated for different parts of Pennsylvania.  For this discussion, we will use three major physiographic regions to illustrate the diversity of Native American architecture that scholars have identified across the landscape through archaeological studies.

Glaciated/Unglaciated Plateau (western Pennsylvania)

In this region of Pennsylvania houses were principally round-shaped structures organized around a central plaza and surrounded by a palisade. Houses were vaulted wigwams with bark covering and sometimes had a semi-subterranean appendage built onto the sidewall. This type of architecture was common to the McFate, Meade Island and related people from A.D. 1000 – A.D. 1400. Longhouses, likely constructed with vaulted roofs and sheathed with bark were used by Iroquoians on the Upper Allegheny and Lake Erie Plain. These were encircled with palisades. Monongahela and other groups in the lower Upper Ohio Valley and Allegheny Mountains built wigwams and a curious form of house having straight to slightly out-sloping side walls and a conical roof of bark. Arranged around an open plaza and protected by a palisade both types had the semi-subterranean appendage feature, others were free-standing. After A.D. 1575, a plaza centered petal-structure consisting of a large circular structure with multiple appendages began to appear on Foley Farm phase settlements.

Artist rendition of a wigwam

Experimental reconstruction of a Monongahela house with conical roof

Appalachian Mountains/Susquehanna Lowlands (central Pennsylvania)

Late Woodland house types in the central and upper Susquehanna were stereotypically vaulted longhouses. There was a tendency for this form of dwelling to increase in length over time. Early in the Late Woodland Clemson Island Early Owasco periods, these were better described as “short houses” resembling a cube in shape. The community pattern appears to have been a loosely organized one. By A.D. 1300 dwellings achieved the greater lengths of northern Iroquoian style, hence the term – “longhouse”. These longhouses were arranged in rows of eight or more to a settlement and surrounded by one or more palisades. The semi-subterranean feature associated with the houses of the Upper Ohio Valley were never attached to the side-walls of longhouses in this region of Pennsylvania.

Experimental reconstruction of a small longhouse and semi-subterranean structure

Experimental reconstruction of an Iroquoian arbor roofed longhouse (under construction). 

Blue Mountain/Great Valley/Piedmont (south-central and south-eastern Pennsylvania)

Longhouse architecture has not been identified for the Late Woodland/Late Prehistoric periods in this section of Pennsylvania. As with the lower Upper Ohio Valley and Allegheny Mountain sections wigwam-shaped houses were the preferred form of architecture in the Great Valley section during the Mason Island, Montgomery and Luray phases. However, none have been identified as having the attached appendage nor have any been linked to a palisaded settlement.  This settlement model appears to mirror the early Clemson Island habitation of central Pennsylvania where the houses were loosely organized in linear formation, near small streams. South of the Blue Mountain water gap, the community pattern of the Shenks Ferry culture evolved from unplanned to planned settlements. During the early Blue Rock phase the pattern was evidently like the Great Valley settlement pattern where houses were circular-shaped and loosely organized. By the later, Lancaster and succeeding Funk phases, Shenks Ferry houses evolved into oval-shaped short longhouses with vaulted roofs.  Well organized into a planned arrangement of one or more house rings having many houses, the settlements were fortified by one or more palisades. A large circular-shaped structure, of an unknown function, was built in the center of these later settlements.

Experimental reconstruction of a Shenks Ferry house (under construction)

As we have seen, Pennsylvanian’s Late Woodland/Late Prehistoric period houses varied from region to region. The environment and social organization of a culture were major dictates as to the type of dwelling being created. Experimental archaeologists provide us with some detail and guidance relative to how houses were constructed by people of the distant past. The primary building materials available to them would have been poles for the framework and bark for hafting the pieces together and to cover the building. These resources were harvested from trees growing in the nearby forest. Raw materials would have been manipulated with stone celts, adzes and other tools. Although houses could be built during all seasons of the year, the spring would have been better suited when trees were easily debarked and the softness of the ground made securing the poled framework into the earth an easier task. Because of their size, smaller wigwam structures took less effort while the large longhouses of 100 feet or more in length would have been a corporate task undertaken by most of a settlement’s population.

We hope that you have enjoyed reading this bi-weekly addition to This Week in Pennsylvania Archaeology and encourage you to visit us again at this web site for more fascinating information about Pennsylvania archaeology. 

For more information, visit PAarchaeology.state.pa.us or the Hall of Anthropology and Archaeology at The State Museum of Pennsylvania .

Experimental Archaeology at the Historical and Museum Commission - Building a Dugout Canoe

The Pennsylvania Historical and Museum Commission (PHMC) has carved three dugout canoes over the past decade. These have been done as public programs using replicated historic and/or prehistoric tools. The sight of people sometimes dressed in loin cloths, wood chips flying and fire attracts a lot of attention and these programs have been very popular with the public. The resulting dugouts have been included in a variety of presentations, most notably, the annual Pennsylvania Farm Show. Although the publicity is good, these projects are examples of experimental archaeology. They are being conducted to aid in the interpretation of the archaeological record.

As a simple definition, a dugout canoe or, simply, a “dugout” is a hollowed-out log used as a watercraft. It is typically made in a cycle of burning and cutting that includes repeatedly burning the log with a controlled fire and then scraping and chopping out the charred and softened wood with a variety of tools that can be as diverse as shells, wooden scraping tools and stone adzes.

The dugout is likely the earliest form of constructed watercraft in the world, and specimens in Europe have been dated to over 9,000 years old. Considering that humans voyaged to Australia at least 50,000 years ago, dugouts are probably at least that old. In North and South America, dugouts have been the main form of water travel since Native Americans arrived from Siberia over 16,000 years ago. In addition, there has been recent speculation that these early people first arrived by boat. The Northwest Coast seems to have the greatest variety of dugouts with some of these being very large and elaborately designed. In Eastern North American, dugouts are preserved in the lakes and bogs of Florida, Louisiana, North Carolina, Tennessee, Ohio and Pennsylvania. The oldest of these date to over 6,000 years before the present.

During our 2005 project, we only used tools that were available to the prehistoric inhabitants of the Commonwealth. It has frequently been assumed by archaeologists that stone adzes were a common tool in dugout construction. The stone adzes were made by grinding down basalt into the desired shape. This was time consuming but attaching them to handles was the real challenge. Several handles and one adze were broken but eventually we developed a design that worked very well. Once the dugout was completed, the wear patterns on the stone adzes were analyzed and compared to archaeological specimens. Surprisingly, the wear patterns on the experimental specimens were not the same as most of the archaeological specimens. Our conclusion was that adzes were not commonly used in dugout construction. For more information on dugouts, visit our Building a Dugout page.

For more information, visit PAarchaeology.state.pa.us or the Hall of Anthropology and Archaeology at The State Museum of Pennsylvania .

State Museum Archaeologist Assists in Death Scene Investigation during High School Forensic Science Class

Dr. Kurt Carr, Senior Curator from the State Museum of Pennsylvania, with the Pennsylvania Historical and Museum Commission, assisted Ms. Evans with a simulated Archaeology excavation. The excavation is part of a lesson on Forensic Anthropology in Ms. Evans's three Forensic Science classes at Susquehanna Township High School, Harrisburg, Pennsylvania.

Eighty high school 11th and 12th grade students worked at the death scene with Dr. Carr. The 100 square meter excavation site is an open, dirt covered, area outside the high school building. The students uncovered and mapped deer skeletal remains at the site.

The skeletal remains were placed in the site with related “artifacts”, that when uncovered and mapped, will “tell the story” of what happened in the deaths of two deer. Under the direction of Dr. Carr, the Forensics students used archaeological techniques, methods and theory in their investigation into the deaths of the two deer. The deer skeletal remains are being “recycled” from the classes’ labs on Forensic Entomology at the high school and were donated by the Pennsylvania Game Commission.


Forensic Science students set up a measurement grid for the excavation. They then walked across the surface of the site mapping the location of bones and related artifacts (potential evidence). The goal of today's activities was to decide the best locations to excavate. Based on this surface collection strategy, student teams picked the square meter of the excavation in which they wanted to search for more evidence. They returned on Friday with trowel in hand to uncover the bones buried in the ground.



On the second day of the investigation, the Forensics Science students worked in teams to excavate a single unit at the site. Under the supervision of State Museum archaeologists, the investigators mapped, bagged and labeled the artifacts found on the surface of their unit. Units are 1 square meter areas on the 100 square meter site. The units have been named using a alphanumeric system. Once the surface artifacts were removed, the investigators used trowels, brushes and dust pans to remove the soil.


After uncovering the buried artifacts, investigators were instructed to map the location of the artifact using tape measures and record this measurement to graph paper. Toward the end of the day, investigators observed a pattern starting to emerge that would "tell the story" of what happened at 36Da231. (Final Report on the Investigation at 36Da231 to come.)


For more information, visit PAarchaeology.state.pa.us or the Hall of Anthropology and Archaeology at The State Museum of Pennsylvania .