The Roots of Experimental Archaeology:

Experimental archaeology has its beginnings in late 19th century England.  During this age of scientific inquiry, groundbreaking advances were being made in biology, geology, and anthropology.  Answers were no longer vested in authority.  Instead, data and evidence were increasingly used to test and support hypotheses.  These more rigorous requirements inspired the first experimental work on lithic (stone tool) artifacts.  Across the Atlantic, American anthropologists were also focusing on evidence, but they had an additional source of information; Native People.  While the motivations for studying Native lifeways were based primarily on salvage ethnography, the knowledge that was gained through ethnography was nonetheless valuable.  Ethnographic data served as a building block for American Experimental Archaeology by incorporating personal experiences within an experimental framework to provide improved inferences about the past.    The availability of ethnographic data and its used by proponents of experimentation reflects a major difference between European and American approaches. 

European experimental work truly began in Scandinavia during the 1830-1850s with the research of Jens J. Worsaae and Japetus Steenstrup. Their research sought to determine the origins of midden accumulations at archaeological sites.  Through their research, Worsaae and Steenstrup were able to eliminate the obvious possible non-human processes, leaving humans as the most likely accumulator of shell midden debris (Trigger 1989:82).  Scandinavian archaeologists were truly pioneers in the field of experimentation.  They worked with biologists and geologists to reconstruct paleoclimate.  They used explicatory experiments, and they compared archaeological and ethnographic data to find similar patterns between them (Trigger 1989:86).

Early Applications of Experimental Archaeology:

Given that Scandinavian archaeology produced the most forward thinking scientific research of the time, why wasn’t it held up as a gold standard?  The answer to this is that few of the publications were translated and none of the academic elites in England or France believed that the European peripherals (Scandinavia) could produce such high quality research (Trigger 1989:87).  The foundations of European experimental archaeology instead came out debates on the European Paleolithic and questions about the deep antiquity of “man” in this region.  At center stage of this debate was Sir John Evans, a respected scientist and archaeologist.  Who, in 1859, accompanied Charles Lyell and Joseph Prestwich to the Somme Valley to assess the discoveries of Jacques Boucher de Crèvecoeur de Perthes (Trigger 1989:92).  These discoveries helped demonstrate a deep human antiquity and led Evans to become a lithic expert.   Evans at first did not accept the handaxes as human-made until, “he had proved, though experiment, that it was possible to duplicate them using only stone tools” (Johnson 1978:337); which he did.  Evans continued his experimental research, which culminated in The Ancient Stone Implements, Weapons, and Ornaments of Great Britain, published in 1872.  In this volume, he explored the challenges of flintknapping but worked under the assumption that there was only one way to produce these stone implements.  If he could recreate them, then he assumed that he had discovered how they were made (Johnson 1978:338).  Evan’s approach to stone tool production was based on trial and error and did not benefit from ethnographic data or analogies drawn from ethnographic research. 

Ethnographic information was and continues to be a vital source of information to interpret archaeological data.  The availability of this comparative data increased dramatically in the 1800s due to policy changes by major European powers.  As mentioned previously, Scandinavian archaeologists used ethnographic information in their interpretations in the 1830s-50s, but the rest of the world lagged behind.  It wasn’t until the mid- and late-1800s that anthropologists began studying the Native cultures of North America with the same kind of vigor.  These studies led to a new source of knowledge for the ‘how things are created’ problem.  Frank Cushing was one of these anthropologists.  

Cushing is widely known in the history of Anthropology as a pioneer in participant observation with his ethnographic work among the Zuni of the American Southwest, but he is less well-known for his experimental work.  During his time with the Zuni, Cushing observed and learned their ways of making various objects, including stone tools (Coles 1979:9-10).  Unfortunately, his knapping demonstrations were largely ignored because researchers were not asking the types of questions that could be answered with this early important work (Johnson 1978:338).  Cushing, however, did apply this new-found knowledge to address one of the enduring mysteries of the 19th-century; the Moundbuilder Myth. 

Although it was widely recognized that the Native inhabitants of North America had created most of the archaeological record, researchers of the time were unable to accept that they had built the monumental earthworks of the Mississippi and Ohio valleys.  We now know that the Adena and Hopewell cultures were responsible for these large and impressive mounds.  In the 1800s, however, scholars debated several theories for the mound builder origins, such a Mesoamerican influence or even a European migration by the Danes or Vikings (Willey and Sabloff 1993:24-26).  There were also a small number of researchers who believed the mound builders were the ancestors of the Native Americans. As late as 1892, one critic pointed to the presence of finely made metal objects recovered within the mound as evidence for foreign influence.  In contrast, Cushing believed that prehistoric Native Americans constructed the mounds and that they were largely responsible for the artifacts within them.  In 1894 he used his knowledge of indigenous metal work to support his ideas.  Through experimentation, Cushing reproduced the metal artifacts using Native American methods and tools, and thereby helped to dispel the moundbuilder myth (Coles 1979:24-25). 

As research and experimentation continued to develop, experimental archaeology became a popular methodology.  As excavations became more detailed, several researchers began to rely on experimentation to replicate archaeological findings.  From the late 19th-century through the middle of the 20th-century, several high-profile re-creations, including prehistoric structures and watercraft, grabbed the media headlines.  Examples included the Roman gateway at The Lunt in England and a Neolithic longhouse at Hjerl Hede in Denmark (Coles 1979:36-37).  Boats also received a great deal of attention, most notably the Polynesian-inspired Kon-Tiki, the Egyptian-inspired Ra, and several Viking longboats (Coles 1979:53-54).  Details and publications on these examples can be found in Coles (1979) and (Stone and Planel 1999).  While these examples were captivating the public, more rigorous testing was underway at a much smaller scale.  This pioneering research would ultimately lead to some of the most significant changes in American archaeology as we know it today.

As noted earlier, the presence of Native Americans pushed American experimental archaeology down a slightly different track from Europe.  The early work by Frank Cushing certainly demonstrated the value of participant observation and Native informed replication. By observing and doing, these researchers learned first-hand the techniques and skills needed to create authentic artifacts.   In the development of lithic experimentation the scenario was somewhat more tragic.  In this case, a member of a Native tribe was actually placed in an institutional setting where he became a museum exhibit and a ‘resource’ for Anthropologists seeking knowledge about Native culture.  This individual was Ishi, the last surviving Yahi of theYana tribe from north-central California.  In 1911, Ishi was found in starving condition.  He was subsequently brought to the University of California at Berkeley by Alfred Kroeber, the noted Anthropologist, and spent the final five years of his life being observed and imitated by researchers trying to understand his seemingly mystical skills (Whittaker 1994:56-57).  Ishi’s impact on experimental archaeology is immeasurable.  His short five years at Berkley has led to nearly 90 years of new research and analysis.  It is therefore not a stretch to consider Ishi’s influence on a global scale.  Through Alfred Kroeber, Ishi’s knowledge greatly influenced Don Crabtree who, along with Francois Bordes, changed field of lithic experimentation. 

The Three Levels of Experimentation:

Coles (1979) recognizes two levels in the early use of experimental archaeology.  The first, most basic level is simulation.  Simulation, he argues, is not really experimentation because it involves the reproduction of the desired object using any technique, including modern tools.  Coles suggest that these replicated objects, such as structures and watercraft serve primarily as teaching aides (Coles 1979:36).  Today, many of these projects are not regarded as experimental, because they do not test specific hypotheses or build upon previous work (Hurcombe 2005:111).  The second level of experimentation is “replication of production”, using only the raw materials and techniques that were known to the past culture under study (Coles 1979:38).  This technique rests on an understanding that there are a finite number of ways humans can produce something within a constricted set of parameters (e.g. material types, techniques).  Replication of production explores these finite ‘ways’ to understand what ‘could have’ taken place under specific conditions, such as the fluting of projectile points using different techniques and materials (Johnson 1978:356).  A vital, but sometimes overlooked aspect of replication is the ethnographic record (Coles 1979:39).  Researchers should remember, however, that this is simply another possible means of production.  Just as behavioral replication studies provide a suite of possible scenarios, so do ethnographic accounts provide a suite of possibilities based on artifact manufacture.

As researchers refined their experiments they began moving beyond simulation and the replication of production, towards what Coles (1979) identifies as the third level of experimental archaeology.  This third level examines the function of the reconstructed object.  This is undoubtedly the most difficult because it requires a very careful documentation of potential biases (e.g. modern tool use and modern interpretation of use).  These later experiments were an early attempt to account for the setting and replicability of the experiments, as well as exploring ‘how’ artifacts were used.  Coles (1979:40) provides an excellent example of tool use relative to the effectiveness of stone and steel axes.  At face value, the steel ax is far superior to stone.  However, this assumes that stone axes are utilized in the same manner as steel axes, which is untrue.  The mechanical properties of the stone ax are completely different than steel, and the effective use of a stone ax requires knowledge of the properties that make it effective.   The adoption of this “third level” experimentation coincided with the development of the New Archaeology of the 1960s.

The Role of Processual Archaeology and the Development of Middle Range Theory

The New (Processual) Archaeology of the 1960s provided a strong empirical and scientific framework for experimentation.  Several of the main tenets of the New Archaeology provided greater structure to the experiments, such as a focus on hypothesis driven research, use of scientific methods, and systems thinking (Johnson 1999:20-26).   Use-wear experiments provide the best examples of utilizing controlled, hypothesis driven research to determine the uses of a particular tools.  These experiments took into account potential post-depositional taphonomic processes that could impact the conclusions of the research (Keeley 1980).   Refinement of processual ideas led to an empirically structured definition for experimental archaeology, which “explicitly attempts to apply experimental methods in the areas of data collection, description, interpretation, and explanation” (Ingersoll and Macdonald 1977:xii).  This scientific approach has led to experimental applications at all scales, from the individual to the regional,  and all materials, such as stone, bone, metals, and plants (Ingersoll et al 1977).  The discovery of universal laws of mechanics, such as correlations between soil pH levels and bone preservation (White and Hannus 1983), or the specific temperature requirements for heat treatment in lithic materials is paramount to this approach (Flenniken and Garrison 1975).  Significant contributions to bone modification (Hudson 1993), taphonomy (Marean and Spencer 1991, Lyman 1994), and lithics (Swanson 1975) are widely published and available here.

In the early 1980s, Middle Range Theory emerged in response to criticisms of the New Archaeology.  Before Binford’s (1981) development of Middle Range Theory, experimental archaeology was largely atheoretical.  The major paradigms of 19th and 20th centuries, Social Evolution and Culture-Historical Approaches, did not require an explanation of ‘how something was made’.   There was no explicit focus on how draw inferences of past human behavior based on archaeological remains.  In contrast, a more explicitly experimental Processual Archaeology and Middle Range Theory, in particular, offered the theoretical platform from which researchers could link the static archaeological record to dynamic living systems in the past. 

It’s Not All Science:

What continues to be unique about experimental archaeology is that it remains in the realm of “low-level” theory and is concerned primarily with establishing direct linkages between behavioral correlates and material remains.  Klejn (1977:2), nonetheless states that these low-level empirical approaches can lead to general theory of a less scientific bent.  Essentially, he is referring to the production of knowledge which is more open ended and experiential, while not limited to methods of scientific verification.   These approaches argue that experimental archaeology, while most closely associated with Processual science, is an evolving tool that can be applied to a wide range of questions, some of them Post-processual.  Several researchers, for example, have used experimentation in phenomenological reconstructions that privilege the experiences of the individual.  Through experimentation, these researchers attempt to experience the nature and meaning of objects and processes through bodily action and sensory interactions (Mathieu 2002:4).  Certainly these studies are not the norm when discussing experimental approaches, but they may prove valuable at some level.  One drawback of course is that the reconstruction of individual human experience is difficult to verify or replicate.  Moreover, the linkages between experiences of modern researchers and artisans of the past are less than secure.  If these explorations can be framed into testable hypotheses with verifiable results, then perhaps we may learn a great deal about the meaning of tool production or use in the past.

Binford, L.R. (1981) Bones: Ancient Men and Modern Myths. Academic Press, Orlando.

Coles, J.M. (1979) Experimental Archaeology. Academic Press, London.

Flenniken, J.J. and E.G. Garrison (1975) Thermally Altered Novaculite and Stone Tool Manufacturing Techniques. Journal of Field Archaeology 2(1/2):125-131.

Hudson, J. (editor) (1993) From Bones to Behavior: Ethnoarchaeological and Experimental Contributions to the Interpretation of Faunal Remains. Center forArchaeological Investigations, Occasional Paper No. 21, Southern Illinois University, Carbondale.

Hurcombe, L. (2005) Experimental Archaeology. In Archaeology: The Key Concepts, edited by C. Renfrew and P. Bahn, pp. 110-115. Routledge, London.

Ingersoll, D., J.E. Yellen and W. Macdonald (editors) (1977) Experimental Archeology. Columbia University Press, New York.

Johnson, L.L. (1978) A History of Flint-Knapping Experimentation, 1838-1976 [and Comments and Reply]. Current Anthropology 19(2):337-372.

Johnson, M. (1999) Archaeological Theory. Blackwell, Oxford.

Keeley, L.H. (1980) Experimental Determination of Stone Tool Uses. University of Chicago Press, Chicago.

Klejn, L.S. (1977) A Panorama of Theoretical Archaeology. Current Anthropology 18(1):1-42.

Lyman, R.L. (1994) Vertebrate Taphonomy. Cambridge University Press, Cambridge.

Marean, C.W. and L.M. Spencer (1991) Impact of Carnivore Ravaging on Zooarchaeological Measures of Element Abundance. American Antiquity 56(4):645-658.

Mathieu, J.R. (2002) Introduction - Experimental Archaeology: Replicating Past Objects, Behaviors, and Processes. In Experimental Archaeology: Replicating PastObjects, Behaviors, and Processes, edited by J. R. Mathieu, pp. 1-11. BAR International Series 1035, Oxford.

Stone, P.G. and P.G. Planel (1999) The Constructed Past: Experimental Archaeology, Education, and the Public One World Archaeology, 36. Routledge, London.

Swanson, E.H. (editor) (1975) Lithic Technology: Making and Using Stone Tools. Mouton, The Hague.

Trigger, B.G. (1989) A History of Archaeological Thought. Cambridge University Press, Cambridge.

White, E.M. and L.A. Hannus (1983) Chemical Weathering of Bone in Archaeological Soils. American Antiquity 48:316-322.

Whittaker, J.C. (1994) Flintknapping: Making & Understanding Stone Tools. University of Texas Press, Austin.

Willey, G.R. and J.A. Sabloff (1993) A History of American Archaeology. 3rd ed. W.H. Freeman and Co, San Francisco.