Excavation is the most time-honored archaeological tool for understanding the processes of the human past, and it certainly represents the type of activity that most people attribute to archaeology. As a definition, excavation is simply the controlled exploration of what lies below the surface, usually carried out systematically in gridded trenches with shovel and trowel. It is often slow and tedious work which involves digging down a centimeter at a time, but can also be backbreaking, difficult toil, shoveling through meters of densely packed soil. But the purpose is the same in either case, to reveal the types of human activities that took place at a site over time. Through the process of excavation, archaeologists look backwards into time, examining an area at discrete temporal periods. Excavating a few centimeters down may reflect the material culture of the 1800s while several meters may uncover artifacts from ancient times. Researchers can use the incomplete material record to reconstruct the cultural history of the place at particular points in time.
The visible remains of the ancient past do not normally lie exposed on hilltops or in the open desert. The Parthenon in Athens and the Egyptian pyramids are the exceptions and not the norm (Figure 8.1). More commonly, archaeological sites are buried beneath the surface and may be partially or totally invisible to the eye. How then do archaeologists even locate sites given such a situation? In the olden days of classical archaeology, explorers used ancient literary references to place names as guides in locating lost cities. For example, Heinrich Schliemann relied on literary references from Homer, his own hunches, and a little luck to find the ancient cities of Troy and Mycenae. In addition, many ancient places (for example, Athens, Mycenae, Sparta) have kept their ancient names unto the present, making it clear where - in a general sense - the ancient places were.
Apart from this traditional method, however, archaeologists today use a variety of heuristic tools in locating sites. Systematic surface survey (discussed in the next section) reveals the variety of materials present across a region, provides information about what activities may have taken place there over time, and suggests the different cultural components represented. Artifact scatter patterns reflect the location of subsurface structures and are used to guide archaeological excavation. When structures are wholly invisible on the surface, researchers can employ the recent advances in archaeological technique to locate sites. Remote sensing, which includes aerial photography and satellite imagery of an area, may accentuate subtle differences in landscape that are outside the spectrum of visible light, thereby suggesting subsurface buildings and features (Figure 8.2). A dark soil stain on an aerial photograph may indicate the rich organic material of an ancient trash midden. Because the state of vegetation depends on soil fertility, abnormal crop marks may suggest underground walls, ditches, and roads. Slight nuances in shadow may point to elevation differences and ancient structures. Computers are now used to detect these subtle differences. Archaeologists can examine the pixel shapes and forms of known structures (e.g., temples) on digitized photographs and try to relate these to similar spectral emissions on the photograph of the survey area. Other projects have utilized geophysical prospecting devices to reveal more about the use of the area. A team may use resistivity meters to discover the soil's resistance to electrical current and magnetometers to detect variation in the magnetic properties of the soil. These tools reveal subsurface soil anomalies which often represent ancient features such as walls and trash middens. Techniques are now so sophisticated that if the buried structure is well preserved, a team might be able to create a reliable plan of the invisible feature. If an excavation phase follows, this will certainly help guide the placement of test trenches and excavation grid.
These advances, along with continual chance discoveries through modern construction, ensure that there is never a shortage of researchable sites and little need to "go looking" for cities, treasures, and pretty things. Additionally, modern archaeologists seldom excavate sites to find what is there. More often than not, the research questions and objectives guide the choice of what and where to excavate.
Planning and Logistics
The excavation of a site is a tremendous endeavor and the obligations to publish the results encourage careful forethought about the direction the project should head and the particular issues to be addressed. Basic logistical questions have to be considered. How many people will be involved? How much money and other resources are available to the project? At what scale will excavation be carried out? In planning an excavation project, many institutions formulate a research design, the overarching goals and plans of the project. It is principally the job of the director(s) to consult with others on the project and develop such a plan.
The size and scope of archaeological excavation and the field team depends upon the resources and funds available, ranging from a small team of volunteers to a network of paid professionals, workers, and supervisors under the guidance of a project director (Figure 8.3). At Isthmia, where excavation has been conducted continuously for half a century, the complexity of work necessitates a high level of organization. Many field archaeologists and students with various backgrounds visit the site every summer to participate in a unified effort to recover and analyze information about the past. The directors provide an overarching plan and oversee all work but it would be impossible to carry out the demands of fieldwork without the help of project coordinators, team leaders, volunteers, and specialists. A project as complicated as the Roman Bath at Isthmia, for example, requires people with technical skills in botany, geology, computers, cartography, as well as special knowledge of certain periods (e.g., Roman, Greek) and areas (e.g., architecture, pottery).
During the planning stage, it is important that the director decide what and where to excavate. Excavation is tedious, time consuming, and expensive, and it is rarely feasible or possible to expose an entire site (Figure 8.4). Nor would it be wise to dig a whole site since archaeological technique will continue to improve and future scientists will be able to do more with the data than is possible now. Because of this, most projects employ a sampling strategy to select areas within the site grid which should reflect the whole site. This can be either completely random, systematic at set intervals (e.g., one trench every 10 meters), or predicated based on the research design. In classical archaeology, this "predictive" approach is most commonly used, since it allows archaeologists to focus on areas where they think they will find information to answer their basic questions. Moreover, projects may decide to sample only a proportion of the archaeological material within each trench. But all these decisions should be made before excavation ever begins.
Archaeologists employ a great variety of equipment to carry out a scientific excavation. The tools used will depend upon the nature of the project goals, time constraints, and the manner of excavation. Most projects use shovels, picks, and trowels, but it is not unheard of to also use bulldozers and heavy equipment to find walls or strip the first layer of soil away in search of features. Heavy equipment is sometimes still used in special circumstances today but this is not the norm. Consider the typical tool list for Ohio State's excavation:
Setting up a Grid
Transit and tripod
Machete and sickle
Gas-powered weed eater
Pocket knife / razor blade
Shovels and Spades
Brushes and brooms
Buckets and Zembilia
Rulers (Straight Edges)
Munsell Color Book
Processing and Storage
Screens and Sieves
Rubber bands, tacks, tape, scissors
The Spatial Dimension
Modern archaeology is scientific and systematic in its approach to vertical and horizontal space. Since the end goal of research is always to interpret the data in meaningful ways, spatial control is essential to any excavation. The degree of spatial control varies according to the goals and resources of the individual project. An agency contracted to do salvage work before an area is bulldozed will have significantly less time for precision than archaeologists who return to the same site on a yearly basis. But in either kind of project, archaeological material is recorded in terms of its matrix, provenience, and association with other artifacts. The matrix is simply the physical (cultural or geological) medium in which artifactual material is found (i.e., normally it is the soil in which the material is located), while the provenience is the specific three-dimensional location of the material within that matrix. Two or more artifacts found together are said to be in association (Figure 8.5). The only meaningful way to interpret the archaeological record is by understanding artifacts in their spatial context.
The most common way of establishing spatial control at an archaeological site is by the imposition of 3-dimensional horizontal and vertical grid. The grid can be actually laid out, using strings or tape, or it can be more notional. All points in a grid are related to a datum, a reference point of known horizontal and vertical location. Usually, archaeologists will designate this point with a semi-permanent marker such as a wooden stake, a spike, or a metal pipe. The datum may be assigned an arbitrary grid reference value such as 0,0 and an elevation such as 100.00 meters (above sea level, often referred to as AMSL, "above mean sea level"), numbers which do not correspond to true elevation and geographical location but nonetheless are points of reference for the site. Moreover, it is possible, though more difficult, through the use of a handheld GPS unit (which tracks one's position in relation to satellites) or a topographic contour map to link the datum to an actual location and a true elevation; actual location will normally be based on longitude and latitude or, more commonly, a national or regional grid, often based on the so-called UTM system.
After the datum point has been chosen and given a value, a physical grid is often created over the entire area of examination. Using telescopic equipment such as a transit, a dumpy level, or a "Total Station" (similar to the other two, but done with infrared light beam and an internal computer), a crew shoots a baseline from the known datum point, plants stakes or nails into the ground at set intervals (e.g., every ten meters), and then triangulates other points off this baseline. The stakes are then strung to create a visible grid across the site; seen from above, the area appears as a series of squares of consistent size (e.g., 10x10m). The corners of the squares are assigned values relative to the datum. Thus, a point with the coordinate value 55N, 32.5E and an elevation of 125.78 m., lies 55 meters north, 32.5 meters east, and 25.78 meters above our hypothetical datum. The value of this system is that every point within the boundaries of the grid is known in relation to every other point and all artifacts can be recorded in spatial terms.
Sedimentation and Stratigraphy
The remains of sites today rarely know the sunlight of the surface. Thousands of years of environmental and human processes have buried ancient civilization under meters of earthen material. Ancient buildings collapsed and were covered over either by new construction or silts deposited through natural activities such as erosion and mudslides. All these immediate and long-term processes left their marks on the archaeological record through discrete layers of earth (called strata) that built up over time (Figure 8.6). The guiding principle in all scientific excavation is stratigraphy, the study and interpretation of strata (layers) in order to understand the historical processes of site formation. Excavation by strata provides not only a meaningful methodological tool for managing vertical and horizontal space but also a conceptual framework for understanding the geological, environmental, and cultural history of the site. Stratification is the long-term buildup of sequential layers of earthen material through human and geological activities. Because the processes of sedimentation change at different points in the history of a site, discrete deposits of organic and geological composite will form which vary in soil composition, color, texture, thickness, and associated cultural material. Thus, the excavation will encounter different "layers" at various elevations under the surface. It is the task of the archaeologist to distinguish these layers at the time of excavation, a problematic and difficult exercise since strata fade into each other and are rarely completely distinct.
Because each stratum resulted from specific kinds of depositional processes at work over a restricted period of time, it is possible to chronologically relate the position of one layer to another. According to the law of superposition, since sedimentary layers accumulate upward through time, the oldest layers will always be the lowest vertical levels, while the most recent layers will be the highest vertical levels. This rule is not without its exceptions. Secondary processes such as erosion, earthquakes, floods, burrowing animals, and human activities sometimes redeposit and mix strata. For example, the digging of a pit and the redepositing of the soil often create a messy situation called reverse stratification where the most recent material lies below the older material. Nonetheless, if we can assume that generally a layer was deposited more recently than the layers below it, it follows that the archaeological material found in the layer was also created more recently. In this way, strata are useful in establishing a relative chronology for a site.
Strata represent a discrete period of time and so artifacts within the layer can be used to date the entire layer. Artifacts from the same layer are assumed to represent the same period of deposition and to have entered the layer at the same time; generally, the most recent artifact dates the entire layer. At Isthmia, for example, a layer that contained 8 Classical Greek potsherds and 2 Late Roman potsherds would receive a Late Roman date (because the Roman period is chronologically later than the Greek period). Although there may be earlier material, the later artifacts still provide the date of deposit for the layer. When the artifact is an inscription or coin, a date of some precision can be assigned to the layer, and in many Mediterranean regions, pottery (which changes stylistically through time) can also be an effective means of dating the layer. So-called "scientific" (absolute) dating techniques are generally not used in classical archaeology, as they are in prehistoric archaeology, since they commonly provide less precise dates than the relative dating of artifact types. For example, certain pottery styles for some periods are restricted to 25 years, while a radiocarbon date may represent a period of one hundred to two hundred years.
In fieldwork at Isthmia, the basic stratigraphic unit and archaeological context is called a "Basket" and corresponds to a three-dimensional area of excavation. (This term comes from the fact that in the past all the material from a single layer was put in a wicker basket - and the name has continued to be used!) All soil in a Basket is considered to be part of the same depositional processes, and all material (artifacts, floral and faunal remains, soil) presumably entered the layer at the same time. When crews observe a visible change in the excavation layers while digging, the current Basket is closed and a new Basket is opened and assigned its own number. Later, if successive Baskets (such as 7 and 8) are considered to represent the same layer, they can always be combined. This ability to later combine Baskets allows careful treatment of subtle differences in soils that may or may not reflect two different layers. But of course the opposite is not true: once a Basket has been dug, it cannot later be subdivided, so care and a tendency to divide Baskets in excavation is a wise policy.
Archaeological excavation is inherently destructive because it permanently removes both artifacts and the surrounding soil matrices from their original context. Artifactual material cannot simply be put back into the ground, and what remains in the way of notes, photographs, memories, and drawings provides the only tools to "reconstruct" the trench. Therefore, responsible and accurate recording is the most essential component of any project, and excavation is meaningless without written and visual records.
Most projects use preprinted forms and notebooks to record the process of excavation. The forms provide a standard means of dealing with information about finds, features, excavation, photographs, and stratification; this in turn ensures consistency between different trench supervisors in the kinds of information collected and permits easy transformation of the data into a digitized format. The field notebooks are the principal means of recording the process of excavation. Detailed narrative includes information about the conditions of excavation such as the nature of the matrix, the personnel present, the methods employed, amount of soil removed, and the weather. More basic observations are recorded about type and quantity of artifacts found in trenches, features and their extent, faunal remains, and stratigraphic units. This is always in juncture with spatial location data (elevation, horizontal spread) so that in layer analysis the archaeologist can reconstruct when and where artifacts began to appear. At Isthmia, the trench supervisor keeps a notebook recording the process of excavation for the trench. In a typical season, the Isthmia project will fill up multiple notebooks recording the excavation of various areas under research. Ohio State University Archaeological Projects in Greece have been involved in nearly two dozen of these areas in the northeast Peloponnesos over the last two decades. At Isthmia itself, areas include (among many) the Byzantine Fortress, the Northeast Gate, and East Field; most recently, efforts have focused on the Roman Bath. In previous years notebooks were identified by the year and the initials of the excavator (e.g., 78 JMP); in more recent times the notebooks which record this research are numbered sequentially 01, 02, 03. Usually, notebooks describe the excavation of one area:
01: Northeast Gate
02: Roman Bath, Room VI, Trenches 1-3
03: Roman Bath, Room VI, Trenches 4-7
Spatial data forms the basis of the recording system, and all objects, drawings, and photographs are linked back to their primary context, a spatial location within a 3-dimensional grid. At Isthmia, the "Lot" is a concept used to link spatial data with an object, context, or record. At Isthmia a Lot is essentially a Basket that has been processed and studied in a preliminary fashion. A Lot number has three basic parts. The first part is a number corresponding to the year of excavation, shortened to the last two digits (e.g., 1967 is 67). The second number represents the notebook in which the Basket was described and this can be related to information about both the Area of excavation and the Trench within each area. The third number designates the Basket, the basic stratigraphic unit of excavation (see above), which is described in the excavation notebooks for each trench. Thus, the lot number breaks down as follows: Year - Notebook - Basket. Lot 78-JMP-005 denotes Basket 5 in Jeanne Marty Peppers' notebook of 1978. More recent notebooks, as we have seen, are numbered sequentially, and their Lots simply record the Notebook and Basket. Thus, Lot 01-005 is Basket 5 from Notebook 1, and when we look at this first notebook, we see that it records the excavations of Trench 7 in Room VI of the Roman Bath, carried out in 1990. This is a resourceful way to link any object back to a spatial context. Other excavations have systems that use different terminologies although the basic principles are the same.
The recording and excavation process is described in its entirety, in the notebooks and in weekly reports of excavators, beginning with an assessment of the area around the trench to be excavated. Previous excavation in the area (complete with citations to earlier notebooks), surface elevations, locations of datum points and plans for the trench, sampling strategies, sifter screen size, location of the backdirt pile -- all these should be noted before digging begins. At Isthmia, the surface elevation is measured at five different points of the trench prior to excavation, ensuring that the slope of the ground can be reconstructed later. Excavation itself proceeds slowly. Soil is removed with pick, shovel, and especially trowel, the hallmark tool of the archaeologist. A trowel allows excavators to remove soil from a trench a few centimeters per scraping, thus affording maximum sensitivity in determining the end of one stratum and the beginning of another. Because discrete layers are treated differently, the crew must remain constantly aware of slight differences in soil texture and color that signify a new stratified deposit. A new deposit necessitates a new Basket designation as well as careful descriptions about the layer, including the associated artifacts, estimated date, and the reason for assigning the new basket. The layer should be described in terms of soil texture, composition, hardness, color (Munsell), and associated natural material (e.g., pebbles).
After a stratum has been fully excavated, the floor and walls are scraped clean and prepared for photography and sketching (Figure 8.7). Spraying the surface with water at this point will delineate features since decayed wood and charcoal often retain water longer than the surrounding soil matrix. Photographs are taken of both sides and bottom of the excavation, and corresponding sketches made. It is a daunting but important task for archaeologists to convert what they see in the trench to a drawing format called a plan view and elevation view. The plan view is a sketch of the bottom of the excavation (at any moment) as seen from above (Figure 8.8). Plan views delineate the horizontal extent and shape of features, artifacts, and strata to each other, complete with a scale, legend, and a key for each distinct stratum and feature. It is also desirable to draw and photograph artifacts in their original context on the excavation floor (in situ) since this is the surest proof that the artifacts did not fall into the trench during the excavation process and contaminate the layer (Figure 8.9). Archaeologists triangulate or measure from known coordinate points to map any object uncovered within the confines of the trench. Stakes or nails of known elevation (usually outside of the trench) are used to ascertain elevations for the trench floor. These elevations, along with color Munsells, and plotted artifacts, will also be included on the plan view. Moreover, the side walls (scarps) are "cleaned" to create profile drawings (or "scarp" drawings). These are scale drawings of site stratification within a trench, that are best seen in vertical cross section. A scarp drawing, complete with elevations, key, and Munsell indicators, serves as a check upon the excavators' interpretations of a site's stratification.
Thorough notes are also made about the processing and sampling of archaeological material during excavation. Because processing will vary according to research goals, it is essential to record the procedures in detail in order to determine how representative are the finds. At Isthmia, for example, not all soil removed from a stratigraphic layer during excavation is sifted; some soil is preserved for later analysis while some is simply discarded. In Mediterranean archaeology, because of the size of excavations and great quantities of artifacts found, it is simply impossible to process all material. Rather, excavators decide beforehand a set percentage of earthen matter to sift (e.g., 50%, 1 out of 2 buckets) and discard the rest. Moreover, even this sieving preserves only a sample of all artifacts in a trench, as artifacts smaller than the holes of the sieve are lost. The larger artifacts recovered in the sample are placed in small cardboard boxes or bags, with attached tags that note the lot number for the basket. A small sample of earthen material is "water sieved" through a finer mesh (1/16 inch) to ascertain the environmental history of the site. The soils wash through the screen but the organic material--usually seeds, charcoal, and animal bones - float to the surface and remain to be carefully sorted out with tweezers and dental picks. As discussed in a later section, all this material is analyzed back at the excavation house.
Artifacts found during the processes of excavation are regularly sketched into the field notebook at the time of excavation or formally drawn to scale on graph paper. At Isthmia, it is not possible to draw and photograph all the finds and so artifacts that seem representative or unusual (imported artifacts, coins, etc.) are drawn and catalogued (Figure 8.10). These drawings and pictures are often published so that other archaeologists know what has been found at Isthmia and so parallels can be made at other sites.
Crews also supplement drawings with photographs of basket floors, profile walls, and in situ artifacts and features. Black and white, as well as color photos, are often better visual records than sketches and will certainly clarify plans and drawings. A project may use photo inventory sheets including information about aperture, shutter speed, and a description of the photo. At Isthmia, small contact prints are pasted directly into the notebooks next to a description of whatever was photographed. Together, plan views, profiles, photographs, and detailed descriptions permit a reasonable reconstruction of the natural deposits of sedimentary material and subsequent analysis of the processes in the archaeological site.