In contrast to the survey’s broad outlook, the excavation focuses on the individual site. This line of fieldwork allows the archaeologist to plumb the depths of a given site in greater detail. As one digs down through the layers at a site, there is the opportunity to document the stratigraphy of the site. In turn, the stratigraphic sequence plays a key role in working out the site’s chronology. The relative age of a given layer is determined by its relationship to the layers above and below it. Technically, this is known as the law of superposition. (As in the case of geology, the layer, or stratum, on top is younger than the one below.) The stratigraphic sequence at an archaeological site is documented by drawing the sections that are exposed by the dig. The excavation also permits the archaeologist to uncover the internal layout of the site. In the case of the Iron Age hillfort of the Heuneburg overlooking the Danube River, this layout takes the form of several rows of rectangular, timber-framed houses located inside the site’s defensive wall. To record the structural remains (hearths, houses, fences, drainages ditches, and so forth) brought to light by the excavation, plans are drawn and photographs taken during the course of the work.
Because of movies, television documentaries, and the popular press, many people have an idea about what takes place in the context of archaeological excavation. The treatment of excavation in the media—with its inevitable focus on the moment of great discovery by the lone archaeologist—often fails to give a true picture of this kind of work, however. Instead of picking up misguided ideas from the media, one needs to think about excavation in more realistic terms. To begin with, one must set aside the notion that the "dig" commonly involves moments when remarkable finds are suddenly brought to light. On most excavations, this rarely happens. Most of the work that is done is much less eventful. The excavation calls for sustained and patient work day after day, first in cleaning each new stratigraphic unit and then in documenting it. The excavation is like a marathon race: the archaeologist sets out on a long, slow run that is likely to be parsimonious in its moments of excitement.
It is worth recalling that the archaeologist is responsible for the proper recovery and documentation of all that comes to light during the course of excavation and not just those things that he or she may find of special interest. At the same time, one of the keys to the success of the excavation is teamwork. The dig is no place for the individualist. The members of the crew must have the training and experience needed to perform their respective tasks skillfully, and under the leadership of the director at the site, they have to work together as a team.
Since archaeological sites vary widely from one to the next, there is no one best approach to excavation. The methods that the archaeologist employs need to be tailored to the specific nature of the site. For example, at the small hunting camp of Pince-vent in France, which was occupied some fifteen thousand years ago, each piece of worked stone was carefully exposed in place, and its position or provenance then was plotted in three dimensions. This was done for individual pieces of bone and other classes of material culture as well. An attempt was made at total recovery in the context of an entire site.
In contrast, the excavation of the early medieval town of York in northern England had to deal with a much larger and more complex site, which produced vast quantities of cultural remains. There, the earliest archaeological layers are buried deeply in the ground, and the town as a whole can be uncovered only by means of a series of excavations over many years. In addition, York is a rescue excavation, where the archaeologist has to complete the dig by a fixed date and where the fieldwork may well be the last chance to investigate the site before it is transformed by modern urban development. In other words, much depends not just on the kind of site but on the purpose of the excavation. If research is the primary motivation for the dig, the excavator is likely to place greater emphasis on fine-grained recording and to ask a wide range of specialists to collaborate in the project. If the work is done because the site is threatened and a rescue excavation has to be undertaken with limited resources, then a more practical approach will have to be adopted.
Moreover, local environmental conditions can influence the kinds of objects that are preserved at a site and, in turn, the appropriate methods of recovery. For example, in an area with wetland conditions of preservation, such as the Somerset Levels in Southeast England, two-thirds of the finds recovered can take the form of organic materials (wood, seeds, leaves, and so forth). Accordingly, special methods must be used in the excavation and the conservation of what is found at such a site.
The main implication is that excavation is far from a routine matter. For the excavator, it is not simply a question of finding a good site and assembling the equipment (trowels, shovels, buckets, brushes, wheelbarrows, screens for sieving, notebooks, drawing boards, cameras, and equipment for surveying the site and taking elevations) and the crew. The director must make many crucial decisions at the start of an excavation, and they determine, in large measure, how successful the field-work turns out to be.
Some of the most important decisions are those in the sphere of sampling. Briefly, this is the name given to the choice of the size of the excavation units, the places where they are to be dug, and the proportion of the site’s total area to be excavated. In the simplest terms, sampling is the decision-making process concerned with choosing where and how to dig at the site. In making such plans, the archaeologist naturally is interested in achieving a good trade-off between the volume of earth to be moved and the return of information about the site from such work. Thus, the sampling strategy— whether the excavator opts for a formal design or for a more informal approach based upon previous experience—attempts to harness the goals of the excavation and the resources (manpower and funds) that are available for the project.
While there are wide differences between one excavation and another, it is possible to identify several common features or integral parts of all excavations. One of them, as previously mentioned, is a sampling strategy. At the start of the fieldwork, there has to be a clear idea of how the excavation will proceed. Since new information about the site will emerge as the excavation unfolds, it often makes good sense to think in terms of a sequential approach to sampling (that is, one where new information, as it becomes available in a stepwise sequence, is incorporated in the decision-making process). A second shared feature of all excavations is the recovery and recording of stratigraphy, or the vertical dimension of the excavation. As mentioned earlier, the identification of individual stratigraphic units and the definition of their interrelationships are at the heart of the chronology of the site. The third integral part of the excavation is the documentation of all features and structural remains exposed at the site, or the horizontal dimension of the dig. This line of evidence provides the key to the functional interpretation of the site (i.e., the kinds of activities that once took place there and their layout and spatial organization).
Still another component of every excavation is the recovery of artifacts and their processing and classification. Collectively, the set of artifacts recovered from a site is referred to as its "finds." The recovery of an artifact can come about in three ways. First, the object can be recognized during the course of digging and its position recorded before it is removed from the ground. In the second case, all of the cultural materials that come to light from a given layer and a given grid square are collected as a group. The degree of spatial resolution in the provenance of the finds obviously is lower than in the first case. In the third case, the artifact is recovered when the soil from a given unit of the excavation is passed through screens (the sieving of the excavated soil), to make sure that even objects of small size are recovered.
The processing of finds normally begins with the washing of the material. This is followed by marking of the pieces (so that each artifact is linked with its provenance in the field) and separation of the finds into different classes of material (coins, pottery, roof tiles, and so on). The next step is the preliminary classification of each kind of material. The lists of such preliminary classifications give the director an overview of the finds at different parts of the site. Later on, specialists will make more refined classifications. To keep track of every item, a well-organized system of storage (that allows easy access to the artifacts) is essential, as is a computer-based information system. As part of the process of documentation, many of the artifacts have to be drawn or photographed. A selection of these illustrations will appear in the final excavation report. Again, the details of the work on the finds—from their recovery in the field through their processing and classification to their eventual publication—vary with the nature of the site.
COMPARING THE SURVEY AND THE EXCAVATION
It is useful at this point to step back and consider some of the ways in which the survey and the excavation differ from each other. Such a comparison, along broad lines, also helps reveal how these two forms of fieldwork complement each other. The survey is, of course, far more expansive in its orientation than is the excavation. The survey is concerned with the large picture. It sets out to record in basic terms all of the forms of habitation that are found in a given region. In this approach the archaeologist is discouraged from having a strong preference for any one period or for any one type of site. Those doing surveys have to be eclectic in their interests; they must direct their attention to the rela-tionships—both in space and in time—that exist between sites. In contrast, the excavation entails a narrower focal point; it takes the individual site as its object of study. The strong suit of the excavation is that it offers much greater control over chronology. In addition, there is the opportunity to examine the internal structure of the site—something that usually is not possible for the survey. In short, the excavation is the mode of fieldwork that allows one to focus in detail on the archaeological record but at the price of the vision of a single site.
In practical terms, the survey can be done with a lower budget than the excavation. It also calls for a smaller crew than most digs. Accordingly, the survey is attractive for the young archaeologist who may have limited access to funding. To carry out a survey successfully, however, the archaeologist should have some training in the fields of environmental studies in archaeology, geomorphology, geography, and economic history. To put it another way, the archaeologist doing a survey has to be ready to address the issues of landscape archaeology. On the other hand, the person planning to direct a new excavation needs to have a somewhat different background, including a good knowledge of the period or periods to be excavated as well as several years of experience on previous excavations.
Finally, it is worth noting that the survey is far less destructive than the excavation. It is a truism that the act of excavation destroys those parts of the site where the digging is done. Damage to the archaeological record is something that all archaeologists want to avoid. In the case of a site that is not threatened, an excavation can be justified only if the fieldwork meets high standards and the results are properly published. In contrast, the survey is far less damaging. There is a clear advantage over the excavation, since the survey leaves the sites in the region essentially intact. It is worth adding that the surface of a site is already being damaged in those parts of the world where modern forms of agriculture are practiced (that is, where the surface of the land is plowed on a regular basis). Because of plowing, the artifacts on the land surface and in the plow zone are no longer in primary context at the site. The plow zone is by no means a kind place for the long-term survival and preservation of most classes of material culture. Moreover, within the plow zone, the artifacts themselves are in active circulation from one bout of plowing to the next. Controlled experiments show that less than one-tenth of the artifacts circulating in the plow zone actually make their appearance on the land surface at any one time. Thus, the collection of pieces from the land surface has only a modest impact on the full set of artifacts occurring in the plow zone.
One can begin to gain a sense of the interplay between the two lines of fieldwork by looking at some of the ways in which the survey and the excavation are mutually dependent. The aim here is to highlight the interaction between the two lines of field-work. The chronology established for a particular region is based for the most part on the results of excavations. In turn, this chronology is used whenever the survey archaeologist assigns dates to what is collected in the field. Once the survey is completed, a report is written on the results of the field-work, and the archaeologist often attempts to put forward a new synthesis of the long-term history of occupation in the region. In turn, the excavation, which makes it possible to explore the local situation in greater detail, offers one of the main ways to test whether this new interpretation is on the right track. At the same time, the survey usually leads to the discovery of promising new sites to excavate. In effect, the survey gives the excavator a wider choice in terms of good places to dig. When an excavation is carried out at one of the new sites found by the survey, the archaeologist, in preparing the report at the end of the excavation, tries to place the site in wider context by reviewing what is known about coeval sites in the region, as revealed by surveys. Thus, in terms of their interaction, there is an ongoing, two-way dialogue between the survey and the excavation.
CURRENT TRENDS IN DEVELOPMENT
It is important to emphasize that both the survey and the excavation are still in the course of development. The methods and strategies employed are not final ones that exist in a standardized form. Like modern medicine in the cure of many diseases, the search is still under way to find the most effective treatment. This aspect of the survey and the excavation can be illustrated by looking at their histories over the last few decades of the twentieth century. One observes in both cases a tendency toward intensification in the approach taken to recovery in the field. When time and money are not an issue, there is an interest in achieving a more refined grain of spatial resolution in the documentation of the archaeological record. Another major development of these years, shared by both survey and excavation, is the growing consciousness of the role played by many different factors in the formation of a site. These include the cultural factors that contributed to the form of the site at the time of its occupa-tion—for example, the structures built at the site, their use and modification over time, and what happened to the objects and building materials at the site when it was abandoned. There are also the natural factors that subsequently acted to transform the objects and features that happen to have survived there. When a survey or excavation is conducted, the archaeologist tries to think through the full range of processes that are involved in the generation of the archaeological record, including post-depositional factors.
In Europe, the approach taken to excavation changed considerably in the last half of the twentieth century. Before 1960 (following the lead of Sir Mortimer Wheeler, a distinguished British archaeologist), it was common for the excavation to be carried out by making a series of trenches at the site. Since then, under the influence of Philip Barker and his work at medieval sites in England, there has been a shift in strategy to uncovering a large, contiguous area at the site. This approach tends to be more effective when it comes to documenting the buildings and other structures at a site.
Another shift that began in the 1970s was a new emphasis on the recovery of botanical remains. To recover seeds and pieces of charcoal from the soil in a more systematic fashion, new equipment based on the principle of flotation was introduced. When soil is run through water containing a frothing agent, the seeds literally rise to the surface and can be skimmed off. Thus, the "ecofact" has taken its place alongside the artifact in field archaeology. A third significant development occurred in 1979, when Edward Harris published Principles of Archaeological Stratigraphy, which offered a new way to record and display the stratigraphic units found by an excavation. Others have begun to pay greater attention to the soils at a site—the matrix that holds the artifacts and the ecofacts and that also contains information on the processes contributing to the formation of the site.
Because of the high costs of excavation in many parts of the world (the United States, Japan, and Europe), it is essential for the archaeologist to know as much as possible about a site before digging begins. Previous knowledge invariably makes for a more efficient research design. It also gives the excavator a better chance to run a well-directed dig. The methods that are used to guide the planning for an excavation can be divided into two main groups: remote sensing and work on the ground. The former method involves acquiring images of the site’s layout or structure from the air. One technique is aerial photography, which has a long history of use in archaeology. In most cases, the photographs that are examined are ones that have been taken for other purposes, such as mapmaking. It is often more rewarding for a project to have its own series of air photographs, taken at a larger scale (that is, from a lower height and showing the more details of the site). Satellite images sometimes are used for this purpose. As the resolution of satellite imagery increases, it will become an essential tool for work of this kind. In addition, under arid conditions, radar imagery from space, which can penetrate desert sands, has proved to be productive in the detection of buried sites and buried features of the landscape.
There are various techniques of geophysical prospection that one can employ on the surface of a site. One of these techniques involves passing an electric current through the ground and then making inferences about buried structures at the site. The earth-resistance survey, as it is technically called, is based on the different patterns of electric resistivity observed on a map of the site. Another type of on-site prospection is the magnetometer survey. Different kinds of structures at a particular site, such as a burned house or a hearth, can be recognized in the form of magnetic anomalies that stand out from the normal soil at the site, which has other magnetic properties. A third method is called ground-penetrating radar, which is related to radar imagery from space. When an electromagnetic wave is propagated toward the ground, some of it penetrates the soil and then bounces back to the surface. By moving the radar instrument over ground along closely spaced lines, patterns of difference in the bounce-back values over the site can be discovered. Again, the appropriate method of prospection depends on the local conditions at the site under investigation.
There is commonly another step in fieldwork before the start of excavation. This consists of coring at the site as a means of checking on the results of one of the three surveys just described. Using either a hand auger or power-driven equipment, cores are made on a grid at the site to obtain a more tangible indication of what is buried in the ground. The overall aim of remote sensing, geophysical prospection, and coring is, of course, to learn as much as possible about the character of the site so that informed decisions can be made when the digging actually begins.
The survey also has witnessed change in the last fifty years of the twentieth century. To begin with, there are now information technologies, such as Geographic Information Systems (GIS), that facilitate the collection and display of spatial data. Formerly, work of this kind had to be done by hand. The earliest surveys in Britain consisted of field walking, where the main aim was to map the major monuments in the countryside. In the 1960s there were an increasing number of surveys in different parts of the world, and the methods used in the coverage of the ground soon became more systematic. There was an attempt to record the full range of sites, small and large, on the landscape. The growth of survey archaeology then accelerated in the 1970s—a time when many surveys were started throughout the world.
At first glance, the survey looks deceptively easy. In planning a new survey, the main challenges would appear to be selecting the region for study; choosing the approach to the coverage of the ground (e.g., the spacing between crew members in the field and the choice of sampling units, such as grid squares or transects); developing a system for mapping and recording the sites identified by the survey; and finding crew members with experience in this kind of work. By the 1980s, however, the realization had begun to emerge that the survey is a more complicated endeavor than the archaeologist had previously thought. The loss of innocence took place when some archaeologists began to repeat the coverage of the same area in different years (as a control on the quality of their fieldwork). To their surprise, they found that there was significant variability in what was observed on the land surface from one year to the next. In retrospect, we can see that most of the surveys done before 1990 were too optimistic in terms of their working assumptions about the dynamics of the plow zone and about the visibility of sites on the land surface.
As part of the trend toward the intensification of fieldwork, the survey archaeologist understandably would like to record the scatters found on the landscape at as fine a grain of spatial resolution as possible. To be more complete in the documentation of what is observed on the land surface, one wants to make sure that all light scatters and even individual pieces are mapped in the field. This line of thought has led some archaeologists to begin doing what is called the nonsite survey. The aim is to record the totality of the cultural materials on the land surface in those places covered by the survey. For very light scatters, there is a fundamental problem that arises in the case of a region where much of the land is plowed, which is the situation in most countries of Europe.
The problem has to do with the circulation of artifacts in the plow zone and the fact that only a small proportion of the pieces in a plow zone make their appearance on the surface at any one time. Indeed, this ratio typically is less than 1 in 10. This means that the pieces on the surface are the result of a stochastic process. In a given place, the random sample of material that is found on the land surface varies from one bout of plowing to the next. The stochastic character of the surface material does not represent a major problem in those places where there are large numbers of pieces in the plow zone. There, the surface sample tends to be much the same from one field season to the next. It becomes a serious methodological problem when the numbers are small. A given light scatter has a low degree of consistency from one year to the next in the number of its pieces, in the different classes of material culture that are represented, and even in its chronology. Thus, the question of how best to deal with light scatters remains basically an unresolved problem for the survey archaeologist who would like to aim for total recovery.
The issue of visibility is no less challenging. Few surveys before 1990 took the question of visibility seriously into consideration. The working assumption was that the pattern of sites (or scatters) observed on the surface at the time of the survey was the same as the pattern of sites that were once occupied in the region. There are two main factors that obscure the recognition of a site on the surface. One is connected with the burial of a site and has to do with geomorphological processes that have modified the landscape since the time the site was abandoned. This often happens on fluvial plains and in the case of a site located at the foot of a steep slope. The second factor involves the state of the vegetation or ground cover in a given field. If the field has been plowed and rained upon, there is no ground cover, and conditions are favorable for seeing artifacts on the surface. If a wheat crop is growing in the field or it is covered with grass, for example, then just the opposite will be the case. Controlled studies, which take both factors into account, reveal that places with good visibility yield many more sites than locations with poor visibility. Thus, all places on the landscape do not have the same potential for the recovery of archaeological sites.
There are three important implications for the design of the survey that follow from this realization. First, there is the need, at the start of the survey, for a good map of the geomorphological features of the region. Normally, the coverage of places where the inflation of the land has buried sites is not all that productive for the survey. Second, it is necessary to record detailed information about ground cover on a field-by-field basis during the course of the survey. Third, because they act as a filter through which to see in the field, the effects of visibility must be taken into account in the analysis of the spatial distributions of sites as well as in the interpretation of the true patterns of settlement in a region and how they have changed over time. In short, the survey is much more complex than it seemed to be in the past. It is no less demanding than the excavation.
Fig. 1. The distribution of Stentinello settlements at Acconia in southern Italy. The survey, by repeating the coverage of the Acconia area several times, obtained the dense pattern of impressed-ware Neolithic settlement observed here.
In archaeology, the recovery of new evidence in the field rests on the partnership between the survey and the excavation. While each line of fieldwork has its own methods and aims, the results produced are complementary. Neither method can stand on its own without the contribution of the other. At the same time, both lines of investigation are still far from reaching their full maturity in terms of their historical development. Thus, the dialogue between them is an open one and will continue to move in new directions in the years to come.