Excavation Manual

2011 Revised Edition

Download PDF Manual II. Handbook of Recording Procedures Glossary


Larry G. Herr


Archaeology is a violent sport. Although we don’t wear shoulder pads and helmets, we are destructive.  [ Video: Rock Breaking ] Like most sports, archaeology can be entertaining and painful at times, but, instead of destroying opposing teams, we destroy what we most treasure, our finds, as we excavate gaping pits through ancient walls and floors. Yet if we record our finds in such a way that we preserve the maximum amount of information, we can at least reduce the theoretical violence of our activity. We must observe precisely and interpret accurately, so that others can work with our data. This manual helps us learn how to do that.

The first part of this manual, Procedures of Excavation, deals with basic day to day excavation activities and procedures for accurate observation. The second part, Handbook of Recording Procedures, describes the recording system.

How to Read the Manual

To some, reading a manual like this one can be as dry as the hills of Gilboa (which receive neither “dew nor rain” . . . with apologies to 2 Sam 1:21). You will wade through very technical instructions and, if you have never dug before, you may find your hair standing on end “like the knotted and combined locks of the fretful porpentine,” (with apologies to Shakespeare, Hamlet). But if you pursue with a “stiff upper lip” and gamely struggle with the manual before arriving in camp, then digging and recording will be much easier when you actually do it. Later, after the first week or two of excavation, if you reread the manual, you will find it much easier to understand as you combine the theory of the manual with the practical experience of digging.

Basic Terms

As with any specialized discipline, archaeology has its own terminology. A glossary is provided at the back of this manual. Words and phrases that are underlined with a red dotted line, will be found in the glossary. Certain terms are so basic to any archaeological discussion that they must be understood at the outset. These terms include: artifact, balk, contamination, elevation, field, field notebook, grid, level, locus, object, plan, section (drawing), square, and subsidiary balk. Other terms may be looked up as necessary.



The method of excavation used by the Madaba Plains Project owes much to Dame Kathleen Kenyon’s strict attention to the stratigraphy of earth layers (the relationship of earth layers to other features, like walls), which she imported to the Near East from England where Sir Mortimer Wheeler had developed it (it is sometimes called the “Wheeler Kenyon Method”). At Jericho, Lawrence Toombs and Joseph Callaway learned the method and brought it to G. E. Wright’s excavation at Balata (Shechem) (see L. Toombs’ appendix in G. E. Wright, Shechem [NY: McGrawHill, 1964]). Wright combined Kenyon’s stratigraphic method with W. F. Albright’s emphasis on pottery typology as a guide to stratigraphic interpretation and began to use specialists as an integral part of the excavation and interpretive process. Our own innovations are merely amplifications of those roots. We have also been affected by interpretive approaches and research design methods from new world archaeology.

This manual provides adequate reading if you have never excavated, and provides a complete introduction to field archaeology as done within the Madaba Plains Project. Other projects, while gather-ing much the same data, use different procedures, forms, and somewhat different terminology. In fact, reading other excavation manuals can be confusing for the beginner. But sometimes, they can provide helpful supplementary material on difficult excavation concepts. The following describe digging procedures on other projects:

Blakely, J. A., and Toombs, L. E.
     1980   The Tell elHesi Field Manual. Cambridge, MA: ASOR.

Dever, W. G., and Lance, H. D., eds.
     1978   A Manual of Field Excavation. New York: KTAV.

Joukowsky, M.
     1980   A Complete Manual of Field Archaeology: Tools and Techniques of Field Work for Archaeologists. Englewood Cliffs, NJ: Prentice Hall.

Kenyon, K.
     1966   Beginning in Archaeology. New York: Praeger.

Field Staff

Senior project director. The senior project director is responsible for the overall direction and success of the Madaba Plains Project, including professional relations with the scholarly community (e.g. American Schools of Oriental Research [Stateside] and the American Center of Oriental Research [Am-man]), relations with the local government of the Hashemite Kingdom of Jordan, and matters of dig policy and general strategy. He chairs the meetings of the project directors.

Project directors. The Madaba Plains Project is currently divided into three major research projects, each directed by a project director. The current projects include the Tall al‘Umayri excavation, the Tall Jalul excavation, and the Hisban Restoration Project and Survey. The project directors, along with the senior project director, develop the MPP’s research goals and objectives, supervise the budget, and oversee the publications of the project’s results.

Consortium director. The Madaba Plains Project is composed of a consortium of several colleges and universities. The consortium director coordinates the activities of the consortium, chairs consortium meetings, directs excavation tours, and coordinates the projects’ educational program, as well as taking part in all directors’ meetings.

Field supervisors Field supervisors are senior staff archaeologists who direct individual research projects under the sponsorship of the Madaba Plains Project. They are supervised by the project directors.   Field supervisors. Field supervisors are trained archaeologists in charge of a single field of excavation. Their responsibilities include assisting their square supervisors in outlining excavation strategies for their squares; remaining alert for locus connections between squares (and thereby, establishing a coherent stratigraphic picture for the finds in their field); organizing the work and workers in the field in an efficient manner; enforcing maintenance of neatly trimmed balks; orchestrating balk drawing; taking photographs; scheduling architects and other specialists for work in the field; ensuring that square supervisors are completing their locus sheets properly; critiquing the weekly summaries of the square supervisors; submitting a weekly interpretive summary of finds to the field director; keeping the Field director informed regarding excavation strategy and new developments; taking care of field tools and other items shared by the square supervisors (such as the Munsell Soil Color Chart and balk drawings); assisting the field director in opening and closing the field.

Square supervisor. Square supervisors may or may not have archaeological experience. They are charged with operating and recording a single square under the supervision of a field supervisor. They must be acquainted with this manual before going into the field. Their activities are many, including organizing the work of one or two volunteers and a local workman; excavating and recording properly; contributing to the strategy planning of the square; and being present for pottery readings each after-noon. The square supervisor’s work, which oversees the initial extracting and recording of raw data, is obviously fundamental to the success of the dig.

In the evening, square supervisors check their notebooks; trace top plans; and prepare for the next day’s work by making sure supplies are restocked. They may also review strategy for the next day’s work.

At the end of each week, square supervisors write a descriptive and interpretive weekly summary for their field supervisor. The following are instructions for using the Weekly Summary Form (see the sample forms at the end of this manual). After completing the identification entries, list all loci which were worked during the week in the column labeled “Locus #.” Describe them in general terms in the spaces to the right (for example, “monumental stone wall,” “hard plaster surface,” etc). In the “Interpretation” section, relate all these loci together in terms of function and stratigraphy. Do not simply copy the information from the locus sheets; try to see all the loci from a larger, more general point of view. This is the time to speculate on any pet idea you may have. Of course, always be sure to defend these speculations! Give the weekly summaries to the field supervisor no later than noon on Sunday.

Volunteers. Volunteers are assigned to specific squares. Their tasks include excavation, sifting, balk trimming, recording, and assistance in afternoon activities such as pottery washing, etc.

Support personnel/specialists. Support personnel represent the depth and variety of our excavation and turn it into a miniuniversity. Their job (in relation to their field work) is to process and analyze the finds according to their specialties, and to add data to the field records so a wholistic interpretive picture may be formed. They do much of their work in camp, though some may spend part or most of the day in the field.

Sites of Excavation

Talls. Between 1984 and 1989 the principle site of excavation for the Madaba Plains Project has been Tall al‘Umayri, a multiperiod site dating from about 2500 B.C. to 500 B.C. The Arabic term tall, formerly spelled tell, means a mound of ruined cities, and is made of strata, stacked one on top of the other, like a layered cake. In 1992 digging began at a second site, Tall Jalul, a large multiperiod site east of the modern town of Madaba. Both the ‘Umayri and Jalul excavations continue side by side. During the summer of 1997, an offseason, excavations were resumed at Tall Hisban (Hesban, Heshbon) to solve certain problems that the publication of the earlier excavations has inspired and to understand the society and lifestyles of recent cave occupation at the site.

Hinterland sites. We also excavate smaller sites within a five kilometer radius of the talls (e.g. Rujam Salim in 1987 and al Drayjat in 1989, etc.). All MPP excavations use the same digging and recording procedures. Each site receives the common name most frequently used by the local inhabitants, if known. Hinterland sites may or may not be divided into fields, and are generally excavated only one season.   Cemeteries. Because cemeteries (aka: “necropoli,” which is the plural of “necropolis”) are usually associated with settlement sites and may be located near other cemeteries, we name the cemetery/cemeteries after the nearest site, then assign a site field letter to each cemetery and a sequential number for each tomb within that cemetery. Thus “Rujam Ahmad Tomb G2” would be the second tomb in Cemetery G near Rujam Ahmad. We excavate tombs like normal hinterland sites with special adaptations.


Limits of the Square

The surveyors stake out each square following the overall grid pattern of the site before digging begins. We excavate only a 5.0 × 5.0 m area, leaving 1.0 m strips (standing balks) along the north and east sides between squares. These balks provide a vertical section in which the stratigraphy of the square remains, and can be “read.” We remove the standing balks periodically as integral parts of the square, assigning the same locus numbers to features that appear in the square.

The grid at Tall al‘Umayri is based upon a regional grid that can be applied to survey sites, as well. The squares within the grid are identified by the numbers and letters of that grid which, in order to localize the squares on the site, must contain four alphanumeric elements. For example, 8K76 refers to Square 76 within a grid of 100 squares (the number in the ten position is the NS axis while that in the one position is the EW axis); this grid is identified as 8K within yet a larger grid of 100 squares (the number is the NS axis and the letter is the EW axis). Other sites within the Madaba Plains Project simply number the squares within a field. For instance, Square B3 refers to Square 3 in Field B.

Square Preparation [ Video: Field Layout ]

The surveyors stakeout the square as illustrated and turn it over to the excavators. The supervisor and volunteers assigned to that square tie strings carefully between the stakes and make them taut and straight to define the limits of excavation. They then remove debris such as weeds up to a meter beyond or outside the square, and establish a bench mark for taking levels with the transit or theodolite. They enter the location and exact level of each bench mark onto the introduction page of the Field Notebook.

Reopening Excavated Squares

If you reopen a previously excavated square, erosion may have altered the original balk lines, making their reestablishment by the surveyors mandatory. Clean the tops of the standing balks and the balk sections. Study the final top plan, section (or balk) drawings, locus sheets, and photographs of the previous season, making a special attempt to relocate recognizable features.

Redraw the plan of the square’s features and take a new photograph prior to removing the inter-seasonal debris accumulation (cleanup). Relocate, check, and verify the previous bench mark for levels, and enter it on your new introduction pate of the Field Notebook. Then, carefully remove the inter-season debris accumulation. Although the previous excavators should have left plastic bags or some other indicator to mark the end of previous excavation, these may have disappeared. Keep looking for the emergence of in situ remains and their relation to the final top plan and photographs of the previous season. Normally, original material is slightly more compact than inter-seasonal debris. Record debris removed in this cleanup process on locus sheets, but identify it as “Cleanup A” or “Cleanup B” etc., instead of giving it a locus number. Over-clean a bit to be absolutely sure that no debris remains to contaminate the new excavation. Because some of the cleanup debris undoubtedly dates to a different occupational period than the unexcavated remains immediately below it, removing a small amount of original ancient deposit is better than excavating cleanup debris as in situ deposits (and thus confusing the data).

Square Supervisor Preparation

Before beginning, make sure you are fully acquainted with the recording procedures in this

Manual. Be sure your notebook contains the Handbook, an introduction page, several copies of each type of locus and supplementary sheets, and an ample supply of metric graph paper for top plans. We recommended that you begin each day with the following items in your work bag:

2 ballpoint pens (no fine points, please—hard to read!)
1 indelible felt pen (fine point)
2 Marshalltown trowels
10 balk tags
20 nails for balk tags
25 identification tags
10 identification tags pre-numbered for pottery pails and packaged with a rubber band
20 plastic bags (for objects, etc.)
30 m of strong, slightly stretchable nylon string
1 line level
1 3m metal measuring tape (metal is more accurate than cloth or vinyl—it doesn’t stretch)
1 plum bob
1 clipboard
4 clothes pins (to keep papers on clipboard during windy days)
2 hardlead pencils (2H is best)
3 small cardboard boxes for metal artifacts
1 ruler or scale ruler.

Whenever you enter the square it is mandatory to wear shoes with flat soles so as not to deface or destroy delicate earth surfaces. Earth surfaces are the lifeline of excavation and they must be handled with delicacy.


The Locus

When all is ready and your field supervisor gives the green light, you may begin digging. Many different things, each called a locus, will be found in every square. A locus is any feature, such as a wall, a surface, an oven, an earth layer, etc. Everything you dig belongs in some fashion to a locus. Think of a locus as a three dimensional feature. There is always length, width, and height. Every locus receives a locus number, and is described on a locus sheet (see the Handbook for details). Assign locus numbers sequentially in the order of their discovery in the square. Continue this sequence from season to season so that locus numbers will not be duplicated.

Earth Locus

The most common loci are earth layers. They are the most important units in our method of excavation because they contain datable artifacts such as pot-sherds, coins, inscriptions, etc., that can help date other associated artifacts and features. Earth loci also provide a stratigraphic context for features such as walls, etc., by connecting walls in a single stratum or phase. An earth layer is made up of a homogenous mixture of dirt and inclusions and can be separated from other layers above and below. Ideally, a single earth layer stretches at an even depth across the square with no observable irregularities or interruptions. Unfortunately, this is seldom the case! Earth layers may be thin in one place and thick in another; they may dip or slope; they may be easily defined in one place but difficult to define in another; they may be stoney in one place and less stoney in another; and they may be interrupted by other features (pits, trenches, etc.). Defining a single continuous earth locus will be one of the most common and yet most challenging and important tasks of a square supervisor.

Probing. To help us understand earth layers correctly, the “probe and peel” allows us accurate controls. In a corner of the square (often the highest), string off a 1.0 m × 1.0 m area and arbitrarily excavate it to test the homogeneity of the earth layer, being careful to note when the next layer is encountered. Keep both the main balks and the subsidiary balks of the probe straight and clean as you dig to check for the appearance of new earth layers more easily; that way you get the vertical as well as the horizontal information. Pay attention also to the dirt as you scrape it with your trowel for indications of the appearance of a new type of earth layer. Indicators will include a different color, consistence, texture, hardness, etc.

Depth of the probe. Never probe deeper than about 50 cm. Earth layers can be deeper than that, but it is best to assign a new locus number to the deeper debris, arbitrary as it may be, because earth loci can always be combined, but can never be separated into smaller units once they are excavated. If you are unsure if you should assign a new earth locus, go ahead and do so. At the least, assign a new pottery pail.

Expanding the probe—peeling. When you encounter a new earth layer in the probe, stop digging and peel (excavate) the upper layer throughout the rest of the square (this helps reduce the risk of contamination of the newly exposed earth layer by the one you are still digging). First, excavate a 1.0 m strip, then do the next 1.0 m strip, and so on until the upper layer is completely removed. Consult your field supervisor as you do this, and constantly clean and check your balks to make sure you are correctly tracing and removing the layer.

Tracing earth layers. Even the best archaeologists may lose an emerging earth layer as they merrily trace it across (or around) the square. If you do, you will be in good company! Back up and begin again, going from the known to the unknown. The earth layer may have changed slightly, or dipped, or a pit may have cut through the layer. Scrape transitional “border zones” between the two earth layers carefully with a trowel to help bring out subtle color or texture changes.

Cleanliness in tracing earth layers. As you trace earth layers, make sure you keep the area clean. This cannot be overemphasized. Do not allow large mounds of debris to build up while you dig; carefully trowel the surface of the new layer to make sure all debris is removed. If you sweep or brush the newly exposed layer, you will not only make tracing easier, but you may also see interesting new features, such as color or texture changes. These could signal the appearance of new features such as the tops of mudbrick walls. (Mudbrick is particularly difficult since “mud” reverts back to regular “dirt” through time.)

Controlled excavation. Excavate no deeper than 10 cm in a single peel, even if the layer is deeper. Always excavate in an orderly and uniform manner by setting up string guides.

Complete excavation of an earth layer. There is no crime in excavating a small part of the newly exposed lower layer along with the upper. It is, however, an archaeological sin of the most sinister order to leave some of the upper layer to be excavated and recorded as if it were part of the lower! Intrusive, later, material from an upper, incompletely removed earth layer may cause a wrong date to be assigned to the lower!

Emerging architecture and earth layers. Leave stones, bricks, or other such items in situ until you can establish their relationship to the earth layers around them. When earth layers approach a wall or installation, determine if the layers seal against (that is, “abut” or actually touch) the wall or whether they have been cut out before sealing against it. If they seal against the wall, the are later chronologically, but if the are cut before reaching the wall or installation, chances are that the cut was made by a foundation trench, and the earth layer is thus earlier than the wall.


Introduction. Surfaces are a special subgroup of earth loci. They were originally floors, streets, or courtyards, and were usually made of beaten earth. You can tell them apart from ordinary earth layers by their compactness, the presence of small bits of charcoal (and possibly seeds or manure, as well), flat lying potsherds (from being walked on), stones that rest on it, and occasionally a very thin layer of sand on the top. The sand often causes the earth above the surface to “flake off” in chunks.

Surfaces were also the resting place of many items which have since decomposed: wood, grains, textiles, etc. If the roof or walls of a structure fell onto the floor, these organic items may be preserved as a light colored stain or pattern in the dirt. The actual object disappeared long ago, but its outlines, and rarely, a three dimensional “ghost,” are sometimes preserved. (In fact, wooden furniture and musical instruments have been reconstructed by carefully filling this “ghost” shape or design left after the object itself had long since decayed!) If you suspect such items may have been trapped under a fallen wall (it probably is not grand furniture or musical instruments), so go ahead flake off the debris above the surface with a knife and blow the dust with a hand blower.

A surface is also the place to look for charred roof beams. Carbonized wood can be used for dendrochronological analysis (treering dating). Place wood samples in a plastic box (not cardboard, as the moisture in the cardboard will damage the specimen).

Thin surfaces. If a surface was used for only a short period of time, it may be so thin that it is difficult to excavate. If so, dig an arbitrary 510 cm from the top as the “surface” and switch to a new locus for the remainder of the debris below.

Removing surfaces. Excavating a surface is fraught with difficulties because the earth beneath is usually much softer than the surface itself. Remember, it is not a crime to take up some of the lower debris layer with the surface, but try to minimize it, if possible. [ Video: Advanced Digging ]

Architectural and Wall-like Installation Loci

Introduction. Walls and wall-like installations are easy to define and excavate if they are made of stone or well baked mudbricks. If three or more stones (or bricks) are found in a line, leave them standing until their relationship to other architectural features, surfaces, and earth layers can be established and recorded properly (described, drawn, and photographed).

Mudbrick. Poorly baked mudbrick presents a serious excavation challenge because it is often difficult to observe a clear difference between the bricks of the wall and those of the outfall, which builds up around the wall when its upper parts collapse and erode. Even with the most careful excavation you may discover too late that you have inadvertently excavated part of a mudbrick wall. Examine the mud-bricks carefully each morning for faint traces of mortar lines etched by the previous day’s wind. If you gently sprinkle water on the bricks, they may crack along mortar lines or along the face of the wall as they dry. Remember, regular patterns are seldom found in nature—rectangles and straight lines are often the results of human activity.

Very slight differences in color may also be apparent between the original brick and the outfall. The bricks should be slightly harder and the outfall tends to flake off from the wall if you strike it horizontally with your pick or trowel. Outfall may also contain some occupational material, such as charcoal and plaster bits, while bricks contain clear traces of straw or similar binding materials. (Some well preserved outfall may also contain straw traces, however.) The best advice on how to work with mudbrick is to keep your balks straight and clean and analyze them often.

Installation Loci

Loci in this category come in many types: ovens, drains, bins, silos, pits, etc. Treat aboveground installations like architectural loci, but handle below ground installations like pits.


The pit locus is the actual line of the pit, not the debris in its fill. Treat the fill as a separate earth locus. This is because the original digging and use of the pit may have been completely unrelated to its fill, functionally and strati-graphically. Make sure you leave a subsidiary balk as you excavate all pits, if possible. Try to determine the earth layer from which the pit was dug, because it helps to date the pit. The fill, of course, represents the end date of the pit’s use.

Identifying pits. Pits lined with stones are easy to identify, but unlined pits present a special challenge. Pit debris is usually softer, more rubbly, and often darker (because of organic matter) in color than the dirt into which it was dug. It can contain ashy debris because of burned refuse. Potsherds and stones often “float” amorphously in the fill.

Avoiding pit contamination. Because pits originally cut down into subsurface materials, they disrupt earlier loci and are later chronologically than the material into which they cut. If you do not recognize a pit, you could misdate a complete stratum. It is thus extremely important to identify pits. Always excavate them before the surrounding material. Indeed, with dirt lined pits it is often proper to excavate an arbitrary 10 cm of surrounding earth to make sure that absolutely all pit debris has been removed.

Foundation Trenches. These are special kinds of pits that were dug to provide foundations for walls. Because the top earth layer cut by the foundation trench should be the latest material before the construction of the wall, and because foundation trenches may contain potsherds from the period of the construction of the wall, they are very important stratigraphic indicators. A balk, or subsidiary balk, running up to the wall is the best way to examine a foundation trench. When you find one, excavate it like a pit.

Burials. Burials are a variety of pit and should be so excavated. When, however, you expose the skeleton (part of the pit fill, strictly speaking), use dental picks, fine brushes, and hand blowers. If the bones are fragile, paint on a silicon-based conservation fluid called PVA (polyvinylacetate). As the bones dry, the PVA will help hold them together.


Daily Excavation Habits

Beginning daily excavation. When you arrive at the site each morning, check the cleanliness of all balks; make sure no foreign debris has entered the square overnight; check if the dampness of the night has revealed new colors or textures in the dirt. Begin the first pottery pail.

Method of Digging [ Video: Basic Digging ]

We dig with a handpick and pointed mason’s trowel (Marshalltown is the trowel brand of choice). Used correctly, a good trowel can be a sensitive instrument in your hands. Wield the handpick in your strong hand, striking the dirt with light strokes to loosen it. With your weak hand use the trowel to scrape the loosened dirt away from the emerging earth layer. Use the “feel” of the trowel as it scrapes, and the “touch” of the handpick as it breaks the dirt to help trace the earth layers. Observe the pick marks to make sure you are still tracing your earth layer accurately. In very loose dirt or in semi-delicate situations, use the trowel alone.

Scrape the dirt underneath and behind you as you move across the square. Then carefully remove the buildup with a dustpan so the emerging layer is not disturbed. If you establish that the layer you are removing is consistently very thick, use a large pick and/or hoe. With extremely delicate work, such as articulated skeletons and complete pottery vessels, use fine tools. Brushing or sweeping each new layer helps bring out underlying earth features and eliminates possible sources of contamination.

Pottery [ Video: Pottery Reading ]

Pottery Pails. Broken pieces of pottery (sherds or potsherds) are found daily by the thousands. Each locus receives its own pottery pail for each day it is worked. If one pail becomes full, assign a second one with a new pail number. Using an indelible pen, fill out an identification tag for every pail immediately when you begin a new one. It is absolutely mandatory that the pail numbers are not accidentally du-plicated, because both duplicated pails are useless due to confusion on their point of origin. For this reason, square supervisors must number identification tags for pottery pails before using them in the field. Keep them together with a rubber band to avoid losing a tag. Omitting a pail number can also cause confusion. Except for reasonably whole pieces, remove all pottery at the sieve.

Pottery contamination. A contaminated pottery pail is a disaster that much of our excavation technique works to prevent. Imagine pulling a Roman potsherd from a well-sealed Iron Age locus! Careful, clean excavation and handling throughout is the best way to ensure against contamination. The following seven commandments should never be forgotten:

  1. Never lean against or sit on balks.
  2. Always throw away sherds if you don’t know where they came from.
  3. Always keep balks clean.
  4. Never use more than one pottery pail at a time.
  5. Always store pails you are not using outside the square.
  6. Never fill a pail more than 3/4 full.
  7. Always mark the identification tag “Poss[ible] Contam[ination]” and immediately start a new pail if you suspect contamination.

Mendable and complete pottery. If you find a group of sherds that may be reconstructed, place them in a separate pottery pail and mark them “Mendable.” Along with other sherds from the same locus, they will go to the Formator for mending. Place complete pots in separate pottery pails and enter them as objects on the locus sheets, as well. Put especially delicate sherds in a separate pail marked “Fragile.”

Pottery processing. At the end of the day, the pottery registrar soaks the pails in water, organizes their washing (Iron Age sherds may contain writing, so examine them for writing before brushing), places them in porous plastic baskets or trays for drying, and orchestrates the “reading” sessions with the field and square supervisors. The square supervisor enters the readings on the Pottery/Bone Readings sheets. The pottery registrar completes a pottery routing tag that accompanies the pottery in all its subsequent processing, analysis, and storage.

Balks [ Video: Balks ]

Straight balks. A square’s main balks preserve the story of its finds in section and are referred to constantly. Keep balks clean and straight. Never undercut a balk or attempt to straighten it relying on eye alone. Periodically trim them so they are straight and plumb to the balk line. Do not leave a few centi-meters of debris, but carefully excavate right up to the balk line. (Balk trimming will be illustrated in the field.)

Do not assign a locus number to the debris from balk trimming. It is arbitrarily discarded. Pottery and other finds may, however, be saved and labeled “balk trim” on their identification tags and on earth locus sheets. If an exceptional find is made, designate the locus of origin (if you know for certain), if not, designate the several loci from which it could have come with appropriate qualifying remarks.

Leave sherds, stones, and bones protruding from the balks if they are secure, since removing them could undercut the balk. Remember that balks dry out and stones fall from dry balks more easily than they do from those freshly trimmed. Therefore, if a stone is of questionable stability, remove it before it falls out on its own. An undercut balk is preferable to a cracked head!

Balk tags. To remember which earth layers in the balk section correspond to your locus sheets in the notebook, place rectangular balk tags labeled with the locus number, into the balk with nails. Write locus numbers with a black indelible felt pen, boldly enough to be visible in most photos (usually 5 cm, about 2 in, high). Insert the nails at the locus boundary and arrange them, one above the other, in as aesthetic an arrangement as possible. Tag your balks as early as you can, because locus boundaries are always more easily discerned in fresh, moist dirt, than in old, dry soil. Later, when trimming the balk, it is easy to scrape around the tag. Never remove them, because you may put them back in the wrong place!

Balk drawing. Because the balks preserve the stratigraphy of the square, each one is carefully drawn. First, trim the balks and interpret them with the help of your field supervisor. Do not outline the locus boundaries with a trowel; it may bias fresh interpretations. Next to the plans and elevations made by the architects, balk drawings (sections) are the most precise drawings made in the field.

Have on hand a hard lead pencil, an eraser, a sharpener (if necessary), and a large piece of graph paper attached to a clipboard. Some may desire a scale rule for measuring on the graph paper. Make sure the drawing reflects the stones and the tip lines of the earth layers as they appear in the balk. When completed, always check the drawings with those from neighboring squares to confirm stratigraphic connections. As you remove balks, add the north and east stubs to the drawing.

The process will be demonstrated in the field, but here is a short description:

  1. Using a transit, place an 18 inch spike horizontally into the balk corner at any   quartermeter level (i.e. 1.00, 1.25, 1.50, 1.75 m, etc).
  2. Tie one end of a firm, slightly stretchable string onto the spike, pull the string tight, and, using a line level, make sure the string is level. Place a second spike at the other end of the balk and tie the string to it so the string is level. This string becomes the datum line from which all measurements are taken. Remove the line level so the string does not sag.
  3. Attach a cloth meter tape to the two spikes with clothespins so that zero begins where the true balk line is located, regardless of where the actual line is. (This will be obvious when the process is explained in the field.) Do not let the tape touch the datum line at any point. Although it does not need to be as tight as the datum line, stretch the measuring tape tightly enough to ensure accurate measurements.
  4. Make sure the balk stamp is on a large piece of graph paper, and, using a scale of 1:25, draw heavy vertical lines for the balk edges and a light horizontal line for the datum line. Label the datum with its level at the side. Leave space in the side margins, for balk stubs which will need to be added to the drawing when balks are removed, and at the bottom for lower portions of the balk.
  5. Most drawings include stones and the tip lines of earth layers. Take measurements of all features vertically, up or down, from the datum line every 25 cm. Place light dots on the graph paper for each measurement and connect them with a lightly drawn line.
  6. Your field supervisor will check the drawing.

Balk removal. We remove balks in order to expose horizontal features which are more extensive than a single square, and to check stratigraphic connections between adjoining squares. Remove them only after they have been drawn and photographed! Excavate the north and east balks as an integral part of your square.

Remove all balks locus by locus, using the same locus numbers as in the main part of the square. Add the new data to the old locus sheets. If new loci are encountered, assign new locus numbers. If balks are removed in a season subsequent to the primary excavation of the locus, fill out new locus sheets, but assign the old locus numbers.

When the balks are gone, draw the stubs onto the appropriate balk drawings, photo-graph them and, if freestanding, remove them layer by layer. After the newly exposed architecture has been fully studied, restring the balk boundaries and begin excavating again.

Subsidiary balks. Subsidiary balks are used to document the relationship of a feature to others when the main balks cannot be used. When excavating walls, cut a subsidiary balk from the wall to one of the four main balks in order to document the wall’s stratigraphic relationship to the main balk. Use subsidiary balks when excavating pits in order to illustrate the method of fill. Also, never remove large or special artifacts without first cutting a subsidiary balk to them, because it is important to know precisely how they are associated with the surrounding loci. Subsidiary balks are also useful when excavating very complex or unclear earth layers. The balks of a probe are actually subsidiary balks.

Draw and photograph all subsidiary balks. Otherwise documentation is incomplete. Place the drawing in the notebook alongside the top plan of the feature the balk is intended to clarify.

Sifting [ Video: Sifting ]

Because we sift all debris and because the sieve is away from the square, carefully organize the transfer of dirt to the sieve and supply the sifter with pottery pails and bone bags, etc. Always maintain a good communication with the sifter, because problems of mislabeled finds can occur. To prevent contamination, never have more than one digging operation going on in a square.

Other Artifacts and Biodata

Collection methods. Collect other (nonpottery) artifacts and biodata (floral and faunal remains) in plastic bags, cardboard boxes, or plastic containers (depending on the nature of the item) and label them with an identification tag. Do not use rubber bands and adhesive tape to bind these containers, because they contain sulfur, which can speed the chemical breakdown of some objects. Organic samples collected for chronometric analysis (14C, etc), must remain absolutely uncontaminated by other organic material. See the section below on Carbon 14 for further details. Glass fragments, coin hoards, beads found together, and other such group items from the same locus may be collected in one container. Some items, such as tessarae and roof tiles, can also be collected together if they are from a single locus.

Some artifacts made of bone, unbaked clay, textiles, ivory, and wood are hygroscopic (water absorbing) and should not be allowed to dry for conservation reasons. Place them in containers which preserve their moisture (e.g. plastic bags). Others, such as metal, should be kept as dry as possible in cardboard boxes.

Do not remove objects from the dirt until they are properly recorded, which means they are precisely plotted on the top plan, levels are recorded, and, sometimes, photographs are taken. Under no circumstance should supervisors clean artifacts in the field. Dirt is often a protection, and should not be immediately removed.

Bones. Almost every earth layer contains animal bones. Like pottery, the sifter saves all examples (including fragments) in plastic bags with an attached identification tag. Back in camp they are cleaned and analyzed by the zoo archaeologist at pottery reading. Bone readings are recorded on the Pottery/Bone Readings sheet.

Carbon 14. When you excavate carbon rich material, follow these instructions for collecting Carbon 14 samples (adapted from Anita Walker, p. 19 in A Manual of Field Excavation, ed. W. G. Dever and H. D. Lance): Clean thoroughly around the area you wish to sample (an air puffer is a good tool for this and will disturb your carbon sample least). Take your trowel and clean off any dirt with water, then wave it in the air to dry. Slip the trowel down the outside edge of your sample area and pry up a minimum of 50 grams (half a sandwich bag), more if you can. Bag your sample, seal, and affix an identification tag. Contamination of your sample will occur only if your carbon sample comes in contact with other carbon. But avoid sampling from dirt full of root systems, if possible. If the carbon you want is likely to have rootlets in it, be sure to include that information on the identification tag, if applicable.

Pollen samples. For pollen samples, sterilize a knife over a match or lighter and take a fresh sample from the earth just exposed, or at a later time from the balk, and place it into a sealed container or a tightly bound plastic bag. Only a hand full is needed.

Flotation. Flotation tanks in camp retrieve seeds and microfauna. Take as many samples as you can from carbon rich earth layers in a large plastic bag as well as periodic samples from other earth loci. The samples are then floated, dried, and analysed by a palaeobotanist.

Photography [ Video: Photography ]

Preparation for a photograph. One photograph can literally be worth a thousand words. Your field supervisor makes the decision to take a photograph. Photograph all in situ objects, architecture, and installations in ways which enlighten their stratigraphic and functional relationships. Approach every photograph as if it is potentially publishable. Even daily progress photos are sometimes published. That means that the area included in the photograph must be clean! Clean all rocks so no dirt or dust adheres to them; brush out footprints in earth layers and on balk tops (another reason to wear flat shoes); and completely remove all tools and workmen.

Because photos depicting squares in several phases of excavation are very difficult to read, try to keep your square in phase as much as possible. Try to take photos in the early morning (before the sun causes shadows) or when the sun is behind a cloud. Every photo must have a scale stick placed parallel to the baseline of the photo. Include an arrow pointing north.

Notifying the photographer. Never call the photographer until the area is completely prepared. Cleaning often takes longer than you anticipate. Be sure the photographer understands exactly what the photograph is meant to record. Ensure that the level of detail you desire will be captured by the photographer. (Wide angled aerials are great for overviews, but lack detail; closeups may show detail, but exclude relationships with adjacent features.)

Daily photographs. Take daily photographs of each square first thing in the morning to document the “progress of excavation.”

Photograph analysis. A digital photograph will be taken of every photo opportunity. Important shots will also be taken with color slide film for use in slide lecture. Make sure the photographer gives you the number of the photograph and that you record it on the locus sheets for all loci included in the photo. Square supervisors will receive printouts of the digital photos for them to analyze and make sure they list the photo on all relevant locus sheets. The photographer will keep his/her own log of photos, including date, site, field, square, and photo number.

Plans and Elevations

The architect draws the plans and elevations of features such as walls and some installations at the request of the field supervisors. Enter the architect’s field sheet number onto the back of the locus sheet or simply mark it as drawn.

Afternoon Activities and Organization

Completing daily excavation. Anticipate the close of work each day by shutting down earth moving operations in time to clear out all excavated dirt, give the entire square a good brush or sweep, complete field records, collect tools and equipment, and ensure that the square is left in a clean and controlled condition.

Afternoon activities. There are many essential activities that are done in camp during the afternoon, including pottery washing, pottery reading, pottery registration, artifact and object registration, architectural drawing and inking, specialty analysis, etc. Square supervisors work on notebook entries, and discuss problems with their field supervisors as needed.

All afternoon activities are geared toward classifying, preserving, organizing, recording, and analyzing the finds. Each activity is organized and directed by specialists or registrars under the coordination of the director(s). All personnel are expected to assist in these activities as the need arises.

This is also a good time to replenish excavation supplies for the next day. Make sure you have enough object containers, balk tags, nails, etc. Always number the pottery identification tags with an indelible pen well in advance of their use in the field.


Ceramics [ Video: Pottery Reading ]

Pottery reading is perhaps the single most important aspect of finds analysis, because it provides the chronological framework for everything else. Pail by pail, all pottery is soaked, washed, and placed into plastic mesh baskets to dry. (Examine all potsherds from Iron Age II loci for inscriptions before brushing.) All square personnel (supervisor and volunteers) are responsible for washing their own pottery, although if your square has little pottery to wash (and being a “team” player), you will want to help others—your heavy days will be coming and you will appreciate their help!

To prepare pottery for “reading,” lay out the washed and dried sherds on tables with the diagnostics (rims, bases, handles, and sherds with decoration or form) in a pile separate from the non-diagnostics (generally plain body sherds). The pottery registrar, field, and square supervisors record the pottery readings on various forms and tags, including the pottery identification tag, the pottery routing tag, and the Pottery/Bone Readings sheet. At the end of the reading session, the sherds and the identification tags are given to the pottery registrar for filing and registration.

Some sherds will be saved in separate storage areas for publication. These sherds are registered and preliminary analysis is organized by the pottery registrar. This includes inking the registration numbers on the sherd; sawing them in two; photo-graphing selected sherds; and channeling them to various other stations of analysis, including the ceramic technician, and the formator who mends broken pots.


Objects are cleaned and registered by the object registrar, who describes them in detail on object forms and conserves the pieces, as well.

Drafting and Surveying

Architects and surveyors are busy on the site drawing detailed plans of the architecture we find. In camp these drawings are inked and labeled.


Specialists do not simply serve the interests of the excavation; their work also provides data for their separate disciplines. Other specialists, such as glass specialists, numismatists (coins), epigraphers, climatologists, lithicists, geologists, etc, are used as needed, when available. Some of them may never work in the field or camp, but analyze the finds in North America.

Osteology. Osteologists work on both human and animal bones. At times they may be called upon to hep excavate particularly important or delicate skeletal finds, or to render an analysis in the field. Their primary work is in camp, cleaning, labeling, sorting, and analyzing the bone materials. Because of local sensitivities, human burials are reinterred elsewhere after preliminary study.

Lithics. All flints, whether clearly tools or not, are saved for analysis by a flint specialists.

Geology. The geologist aids the observation of finds on the site as they are uncovered and helps in the analysis of surfaces and stone walls. Petrological analysis of objects is another aspect of the work.

Soils. A soil scientist helps analyze earth loci in the field and works on soil deposits in the region.

Palaeobotany. The palaeobotanist deals with the seeds recovered from flotation. A palynologist analyzes pollen fractions from earth layers. [ Video: Flotation ]

Ethnoarchaeology. The study of how the material culture of contemporary peoples in the region reflects their present social organization and cultural activities may give us clues how we can use ancient artifacts to extract similar meanings about the past.

Regional Survey. A separate team surveys the to determine ancient settlement patterns, economic structures, possible land use, available natural resources, etc. They operate under their own detailed manual.


Completion of a Square

The ultimate end of excavation in every square is arrival at indisputable bedrock or undisturbed horizons of soil, including sub-bedrock features (e.g. cisterns). Take a final photograph of all standing balks and the bottom of the square; complete all balk drawings and locus sheets; and draw a final top plan. You are now prepared to celebrate with a bedrock party.

An Unfinished Square at Season’s End

If you do not reach bedrock or undisturbed soil by the end of the season, photograph all standing balks and the floor of the square; draw an accurate top plan; complete balk drawings; and finish locus sheets as much as possible. Then nail three to five plastic sheets to the bottom of the square and place a small amount of dirt on top. Be sure to plot the plastic sheets on the last top plan. When the square is reopened, the plastic sheets will indicate the beginning of in situ remains.

In phase. If at all possible, stop unfinished squares so inter-season erosion does not destroy the remains. Attempt to end your square “in phase,” so that all the exposed remains are at the same stratigraphic level.

Completion of Camp Responsibilities

In camp, complete all recording procedures and take an inventory of all tools and supplies. Staff members are still “on the job” and are responsible to the excavation until this activity is completed.

Download PDF Manual II. Handbook of Recording Procedures Glossary

© 2024 Madaba Plains Project. All rights reserved. Any unauthorized duplication of images or content on this site is strictly prohibited.