Barbara E. Luedtke
Department of Anthropology
University of Massachusetts/ Boston

Paper delivered April 11, 1996, as part of forum on "Chaos or Congruence? Regional Lithic Data Bases in North America", at annual meeting of the Society for American Archaeology in New Orleans, LA. (Amended slightly 12/11/96)

Lithic sources have been a topic of archaeological interest in New England for over 100 years, but at this point we have not drawn together information about quarries, material characteristics, and geochemistry into a coherent data base. Furthermore, regional archaeologists have just begun to discuss a host of issues such as the structure of a regional data base, the level of detail that should be included, "quality control" for data to be entered, and how to provide access to the data base. This presentation will introduce the forum, summarize progress thus far for New England, and outline other issues that need to be addressed by those of us committed to the development of comprehensive and congruent lithic data bases.

I first became involved with lithic data bases while working on my dissertation, which involved compiling data on chert sources in the Upper Great Lakes region (Luedtke 1976). In 1987 I gave a paper at the SAA meetings in Toronto in which I called for development of a much larger lithic data base, and made some suggestions about how such a data base might be structured (Luedtke 1987). Much of this paper was incorporated into an appendix of my recent book on chert (Luedtke 1992). So I thought I knew something about this topic until I attended a conference in Maine a year ago and discovered that: a) there were already many more people than I knew about working more or less independently on lithic data bases in the Northeast alone, and b) there were many issues relevant to such data bases that I had never even considered, or at least hadn't thought through enough to realize they might be problems. At that point, I decided it would be a good idea to get an even broader perspective by starting a conversation among people from across the whole continent who were interested in this issue. Knowing there were many more people interested in such data bases than could possibly talk in any symposium, I decided to try the forum format, which is designed to include a great deal of audience participation. I invited a number of people to speak about what is being done with regard to lithic data bases in their regions, and also about their concerns with the process, just as a start for the larger conversation.

If, as I suspect, we are currently in a state of chaos with regard to the development of lithic data bases, it will certainly take us longer than one morning to achieve congruence. My goals for this forum are much more modest. First, I simply hoped to get a lot of us together in one room so we can meet each other and begin talking. Second, I hoped to get a feel for what is already being done, and what models we might have for further work. Third, I wanted to get as many of the issues out on the table as possible: what do people think are absolute necessities to discuss? What are the problems they foresee? I have some ideas of my own, which I'll mention shortly, but my experience at the Maine conference suggests to me that there will be many more. Finally, I hoped this forum might help us begin a process that will lead toward building consensus on some of the issues.

I'll start off with a brief summary of the history of lithic data bases in New England. This is a challenging region for such endeavors because stone tool makers here used an unusually wide variety of lithic materials (Calogero and Philpotts 1995). Obsidian is nonexistent and chert is scarce, so volcanic and metamorphic rocks such as quartz, rhyolite, hornfels, quartzite, argilllite, silicified shale, and basalt played a more important role here than they usually do in lithically better-endowed regions. These volcanic and metamorphic rocks outcrop widely, but at many outcrops the materials are too coarse or too fissile even for New England knappers. Artifact-quality varieties often occur only in small outcrops, sometimes dikes, sills, or contact zones only a few meters in diameter. To further complicate the issue, all of New England was glaciated during the Pleistocene, so while many bedrock quarries are small and spatially restricted, the same materials are often available in secondary deposits which are sometimes located considerable distances from their bedrock sources.

Nevertheless, New England archaeologists, like archaeologists everywhere, quickly recognized the importance of knowing the sources of lithic raw materials. The first example of a proto- data base that I am aware of is Henry Haynes' 1886 description of five New England lithic sources (Haynes 1886). William Fowler also published a summary of data on several sources (Fowler 1971) and Ron Thomas included a few local sources in his pioneering effort to develop a lithic data base (Thomas nd). Other publications that include data relevant to such a data base have different goals, such as documentation of individual quarries (e.g. Bonnichsen 1982; Bowman 1981; Bowman and Zeoli 1978; Gramly 1980; Pollock 1987; Ritchie and Gould 1985) studies of how particular materials are distributed on sites in the region (e.g. Bourque et al 1984; Strauss 1988, 1989) and studies of prehistoric exchange or interaction patterns (e.g. Calogero 1991; Curran and Grimes 1989; Lavin 1983; Spiess and Wilson 1989). There is no single publication that draws together all the information available about lithic sources in this region, and in fact much information about quarries and sources remains in people's heads, and not down on paper (Luedtke 1993). We are also accumulating increasing quantities of petrographic and geochemical data, (e.g. Calogero 1991; Lavin 1983) some published, but much still unpublished.

A somewhat different approach to lithic data bases involves the assembling of lithic type collections, which can be found at many regional museums, colleges, and universities, and which have also been formed by many individuals and avocational societies. The type collection developed by Jack Holland at the Buffalo Museum of Science is probably the most comprehensive. Each of these type collections is completely autonomous; there are no widely accepted standards for how samples should be collected, how large a sample is adequate, what should be done to ensure that the samples are representative, how samples should be labeled, etc.

In 1994, Dave Sanger of the University of Maine organized a conference on Geoarchaeological Approaches to Lithic Sourcing in the Northeast; this was the regional conference referred to above that was such an eye-opener for me. Fifteen of us, representing 11 states and provinces, came together to discuss common problems such as the rationale for lithic sourcing, the role of hand specimens, student training, interdisciplinary integration, nomenclature, and new techniques. We agreed to work toward a regional data base, and I was supposed to help develop a prototype, but this effort has not progressed very far, partly because I believe there are still many unresolved issues that need to be discussed. For example:

Structure of the data base

I haven't encountered anyone who disagrees with the concept of creating a data base that includes a wide variety of information about lithic sources including quarries, visible attributes, petrographic descriptions, and chemical composition. Modern computer programs with windows and hypertext, etc. make it easier than ever to compile such multi-faceted data. However, we will need to decide on the level of detail to be included. For example, in the case of trace element data, do we want to include every single raw value, or summary statistics only? What about "bad" values? How much data on standards and laboratory procedures must be included? We will also need to develop procedures for searching the data base. Models exist from geology (e.g. Scott 1994) but I'd still argue that it makes most sense to begin geographically, working outward from the site where the artifact in question was found. There was surprisingly little agreement at the Maine conference as to whether the data base should be distributed in print or electronic format. The latter has the advantage of being infinitely expandable and easily added to or changed, but is in some ways harder to use and could present security problems. Which leads directly to the second area of concern:

Access and Control

It may be that those of us who went to college in the 60's have never quite recovered, but my initial inclination was to create "The People's Data Base" on the Internet, open to everyone, and to which everyone could add data as they acquired it. Participants in the Maine conference quickly pointed out the flaws in this scenario, and further reflection has added more. For example if the data base is to be universally accessible, then we need to think very carefully about how we provide data on quarry locations, so that the data base does not become a looter's guide to quarry sites. If only very general locational coordinates are given, the sites will be protected but the data base will be less useful for legitimate users. On the other hand, if precise coordinates are given some people may be reluctant to provide locations of quarries they know about, and this also weakens the usefulness of the data base. Perhaps we need a separate but linked data base for quarry data.

Another access issue results from the fact that a number of CRM companies now offer lithic sourcing as one of the analytical services they can provide to archaeologists. Some of them may well consider the source and material characterization data in their files to be "proprietary information" that should not be revealed to potential competitors. Is this data permanently "off limits" to the rest of us? Are there ways we can protect the rights of people who are trying to use their expertise and the data they have generated to make a living as archaeologists?

Another problem with the concept of the "People's Data Base" is quality control; who decides what is or is not to be included? My initial inclination was to include everything and let the data speak for themselves, but this now seems like a bad idea. Experienced users may be aware that data generated by a particular technique or a particular laboratory are suspect, but new users will tend to assume that all the data are of equal validity unless they are given reason to believe otherwise. Bad data will lead to incorrect source identifications, and surely the archaeological literature is already clogged with enough of those (Calogero 1992). I reluctantly admit that there needs to be a panel of some type that decides what data to put into the data base. But how is such a panel to be chosen, and what safeguards can we develop to prevent tyranny by the Rock Czars?

Units of analysis

A third area of concern is our units of analysis. Are "sources" the same as quarries? Formations? Members? Groups? All of the above, depending on circumstances? How do we deal with intra-source variability? Many type collections include only the most characteristic "typical" varieties of a material, but any visit to a quarry shows that the range of variation in all characteristics is considerable. Furthermore, examination of assemblages shows that stone tool-making people did indeed use "atypical" varieties of lithic materials, so we cannot continue to ignore these. Finally, how do we incorporate data on secondary sources, knappable materials in stream bed, glacial, and beach deposits that may lie some distance from bedrock outcrops? If our knowledge of bedrock sources is incomplete, our knowledge of secondary sources is virtually nil. Yet these sources were indeed used, and their distribution should be included in the data base.

Terminology

A fourth area of concern is terminology. As I've pointed out elsewhere, there is no consensus within archaeology or geology as to the exact meaning of basic terms such as flint, jasper, or chalcedony (Luedtke 1992). We also need to develop standard terminology for describing visible attributes such as color, texture, luster, and diaphaneity. I know that others of you have begun to discuss terminology, and Tim Church's Sourcebook is a helpful first step (Church 1994). Furthermore, assuming we can reach consensus as to our units of analysis, as discussed above, we will then have to decide how to name them. A strong argument can be made for simply following geological nomenclature, but there is also merit in continuing to use terms that are already firmly ensconced in the literature and in archaeologist's minds. After all, if I mention Knife River Flint most of you know what I'm talking about, but that's not necessarily true if I start talking about "silicified lignite probably from the Golden Valley Formation".

In Greek mythology, Chaos was a void that existed before the creation of the universe, a place of utter disorder and confusion that nevertheless contained the potential for all creation. I'd like to hope that our current chaos also contains the seeds of eventual order, but I look forward to hearing what the rest of you think.

Bonnichsen, Robson (ed.)
1982 Archaeological Research at Munsungun Lake: 1981 Preliminary Report of Activities. Quaternary Institute, University of Maine, Orono.

Bourque, B. J., R. G. Doyle, and S. White
1984 The archaeological distribution of banded spherulitic rhyolite in Maine. Man in the Northeast 28: 111-119.

Bowman, William
1981 Sally Rock quarry. Bulletin of the Massachusetts Archaeological Society 42: 39-40.

Bowman, William, and Gerald Zeoli
1978 Discovery of a new major aboriginal lithic source. Bulletin of the Massachusetts Archaeological Society 39: 35-47.

Calogero, Barbara L. A.
1991 Macroscopic and Petrographic Identification of the Rock Types Used for Stone Tools in Central Connecticut. Ph.D. dissertation, University of Connecticut, Storrs.

1992 Lithic misidentification. Man in the Northeast 43: 87-90.

Calogero, Barbara L. A., and Anthony R. Philpotts
1995 Rocks and minerals used by tool knappers in New England. Northeast Anthropology 50: 1-17.

Church, Tim
1994 Lithic Resource Studies: a Sourcebook for Archaeologists. Special Publication #3, Lithic Technology. Tulsa, OK.

Curran, Mary Lou, and John R. Grimes
1989 Ecological implications for Paleoindian lithic procurement economy in New England. In Eastern Paleoindian Lithic Resource Use, edited by C. J. Ellis and J. C. Lothrop, pp. 41-74. Westview Press, Boulder, CO.

Fowler, William
1971 Some sources of New England flints. Bulletin of the Massachusetts Archaeological Society 32 (3-4): 23-28.

Gramly, Richard M.
1980 Prehistoric industry at the Mt. Jasper mine, northern New Hampshire. Man in the Northeast 20: 1-24.

Haynes, Henry W.
1886 Localities of quarries worked by the Indians for material for their stone implements. Proceedings of the Boston Society for Natural History 23:333-336. (Reprinted 1985 in the Bulletin of the Massachusetts Archaeological Society 46:42-43.)

Lavin, Lucianne M.
1983 Patterns of Chert Acquisition Among Woodland Groups Within the Delaware Watershed: a Lithologic Approach. Ph.D. dissertation, New York University.

Luedtke, Barbara E.
1976 Lithic Material Distributions and Interaction Patterns During the Late Woodland Period in Michigan. Ph.D. dissertation, University of Michigan, Ann Arbor.

1987 Chert source analysis: getting down to basics. Paper presented at the annual meeting of the Society for American Archaeology, Toronto.

1992 An Archaeologist's Guide to Chert and Flint. Archaeological Research Tools 7, Institute of Archaeology, University of California, Los Angeles.

1993 Lithic source analysis in New England. Bulletin of the Massachusetts Archaeological Society 54: 56-60.

Pollock, Steven G.
1987 Chert formation in on Ordovician volcanic arc. Journal of Sedimentary Petrology 57: 75-87.

Ritchie, Duncan, and Richard A. Gould
1985 Back to the source: a preliminary account of the Massachusetts Hill quarry complex. In Stone Tool Analysis, edited by M. G. Plew, J. C. Woods, and M. G. Pavesic, pp. 35-53. University of New Mexico Press, Albuquerque.

Scott, D. A.
1994 A useful mineral data base for the museum laboratory. Archaeometry 36: 171-172.

Spiess, Arthur, and Deborah Wilson
1989 Paleoindian lithic distribution in the New England - Maritimes region. In Eastern Paleoindian Lithic Resource Use, edited by C. J. Ellis and J. C. Lothrop, pp. 75-97. Westview Press, Boulder CO.

Strauss, Alan
1988 A model for the prehistoric distribution of poor to moderate grade raw materials from their sources in southeastern New England: the Attleboro red felsite example. Archaeology of Eastern North America 16: 43-54.

1989 Narragansett Basin argillite: lithology, chronology, and prehistoric tool manufacture. North American Archaeologist 10: 25-37.

Thomas, Ronald A.
nd Lithic Source Notebook. Distributed by Island Field Archaeological Museum and Research Center, Milford, DE.