Field Observations on Military Earthworks
by David Lowe 8/13/2007
What follows is a (rather rambling) discussion of the principal words and concepts that
members of the Civil War Fortifications Study Group use in the field to describe what
we're looking at when examining military earthworks. These terms refer to individual
structures, components of earthworks, and visible details. Not all of these terms would
be appropriate or necessary to include in a maintenance oriented database, although most
would be useful if/when the park service undertakes a systematic survey of its surviving
earthworks resources. I do guarantee that if you master this list you will not only know
what to look for in the field but also be able to hold your own in conversations with any
passing "trench nerd."
This discussion focuses on the pieces of the puzzle and does not really touch upon the
most fascinating aspect of studying military earthworks. That is why earthworks were
built where they were built and in the forms they were built. Suffice it to say that the
architecture of earthworks follows an inexorable and brutal military logic-to kill the
enemy and to keep the enemy from killing you. Self preservation provided a powerful
incentive to hone earthworks skills. At the beginning of the Civil War, earthworks were
largely the province of trained military engineers, thus we see many classic textbook
examples in the field. As the war dragged on, common officers and soldiers pitched in
with their own hard-earned battlefield expertise, and we see a great deal of innovation
and the breakdown (or improvement, some would say) of classic forms. That is why it is
difficult to simply adopt a vocabulary directly from the engineering manuals of the time.
As well, I give some consideration to the end users of my vocabulary. Most will not be
or ever desire to be experts on the subject of military earthworks and should not be
required to master what amounts to a foreign language to do their job.
These are the kinds of things that we look for and how we talk about what we find.
Military Earthworks: An excavated structure used for military purposes. Earthworks
could be prepared (built in safety in anticipation of use) or rapid (built on the battlefield
often under fire). Prepared earthworks tended to incorporate more of the classic forms
of fortification. Rapid earthworks can be extremely idiosyncratic within certain
parameters, and their forms evolved rather dramatically during the Civil War. Military
earthworks may be grouped roughly as artillery defenses, infantry defenses, and logistical
support structures. The architectural form of an earthwork-its outline on the ground-
was called its trace, as in "this redoubt has a hexagonal trace."
Embankment or Parapet. All military earthworks, being excavations, consist of ditches
and embankments with embankment size proportional to the width and depth of the ditch.
For convenience in the field, we have referred to all such embankments as "parapets," but
this has a more specialized meaning when addressing permanent fortifications (the wall
surmounting the rampart, rampart itself having limited application to earthworks). As
well, some quibble about applying parapet to earthworks that are off the firing line and
auxiliary to the defense. For a maintenance database that also contains descriptions of
permanent fortifications, I would be inclined to use the more generic "embankment" as
this is inherently an earthen structure and is unambiguous. Parapet, however, should
remain a common part of our vocabulary. The exterior of a linear work faced the enemy.
An enclosed work was designed for defense in all directions, so exterior is simply
outside. The interior is the sheltered area behind or inside the earthwork.
Ditch. Provided the earth or spoil used to build the embankment. The ditch may be
interior, exterior, or both interior and exterior to the embankment, depending on purpose,
design, availability of spoil (e.g., depth of topsoil or height of water table). An exterior
ditch was incorporated into a defensive structure as an obstacle. Only if an exterior ditch
was purposely flooded may it properly be termed a moat. Some writers have tried to
differentiate between "ditch" (exterior) and "trench" (interior) but this seems to
needlessly confuse the issue for laymen. The terms, for the most part, are
interchangeable. When discussing an exterior ditch, the scarp is the continuation of the
slope of the embankment into the ditch; the counterscarp is its opposing slope. In other
words, if you were unfortunate enough to be the attacker, when reaching the ditch you
would fall over the counterscarp to the bottom of the ditch and then face a hard climb up
the scarp before coming to grips with your enemy.
Revetment. A retaining wall to hold the rear of an embankment at a more upright angle.
Revetments were constructed from prepared or available materials, such as sandbags,
gabions (earth filled wicker baskets), hurdles (wicker screens), fascines (tied sapling
bundles), posts and planks, cut tree trunks, sod, and stone. The use of sandbags can
sometimes be confirmed by finding a stratum of alien sand on the surface of an otherwise
clay earthwork. Common surviving revetments are of stone, but, surprisingly, under rare
soil and drainage conditions, wooden revetments can survive in situ. We've run across
several examples over the years. Surviving revetments of stone or wood would require
special maintenance considerations.
Measurements. Grade, original ground surface. Relief, distance from the crest of the
embankment to the bottom of the ditch. Command, distance from grade to the crest.
Depth, distance from grade to the bottom of the ditch. Width of embankment, a cross
section at grade. Width of ditch, from base of the embankment to the rim at grade.
Length, linear extent of an embankment segment.
Artillery Defenses. Classic models of artillery defenses include redoubts (enclosed
structures), lunettes (four-sided and open at the rear), and redans (two faces like a
shallow V pointing toward the enemy). These were essentially the classic building
blocks of a system of fortifications and could be assembled in a variety of combinations
depending on intent and terrain and then connected with lines for infantry often referred
to as curtains or curtain walls. For example, a lunette when connected to another lunette
by a curtain (with modifications to the angles) is magically transformed into a bastion.
There is also a full range of irregular and idiosyncratic artillery defenses that are best
termed just batteries. The engineers (sometimes with a sniff) described these as irregular
works. Fort is bandied about in the sources but most often referred to an enclosed
redoubt of any size or configuration. For inventory (and intriguing conversation) it is a
good thing to note if an earthwork is a redoubt, redan, etc. but in terms of maintenance it
matters not a whit. You can leave that to interpretation. Enceinte is a term applied to the
interior ring of a system of fortifications, rarely to an individual artillery work.
Terreplein is the flattened area inside an artillery work upon which gun platforms are
constructed and probably comes closest to describing the interior of an enclosed
earthwork. We usually refer to the interior of an artillery work as the "floor."
Gun platform. An artillery piece required a firm, level platform from which to fire. This
was constructed by laying a floor of sleepers and planks or by placing logs side-by-side
(corduroy). The size of the platform provides evidence for the size of the gun the
platform was designed to hold. A platform for a standard field piece during the Civil
War, for example, measured about 14 x 14 feet, giving room for the crew to operate.
Sometimes, the platform's perimeter is outlined by a shallow ditch, evidently for
drainage. Often, a platform is flanked by traverses, which help define it. The platform
was designed so that a gun could fire in one of two ways: over the top of the
embankment, called barbette; or through an opening cut into the embankment, called an
embrasure. These French terms (unavoidable) could be considered attributes of the
platform. In some cases, the presence of a platform may be inferred by the presence of
an embrasure when there are no other indicators. But we use caution as there are many
"false embrasures" caused by large falling trees. Occasionally, a gun platform might be
roofed over and termed a casemate. Shooting a compass bearing through an embrasure
provides the direction of fire. These architectural details can show interlocking fields of
fire and provide insight into the overall design of an earthworks complex.
Gun ramp. Gun platforms were sometimes built up from and sometimes dug down into
the grade. A platform built above or below grade required an earthen ramp to move the
gun into position. Gun ramps, typically 10-12 wide, are a common survival.
Glacis (glay'-sus or glah-see'). An artificial slope built along the outer edge of an
exterior ditch primarily to deflect incoming ordnance up and over the embankment.
Mostly associated with more permanent fortifications but we have seen some beauties
associated with fieldworks.
Traverse. An embankment built within an earthwork at a right or oblique angle to its
front. The engineers considered this an internal work, and its purpose was to intercept
fire coming from the sides or enfilade fire. A large traverse might shelter a magazine or
bombproof in its rear, both also considered internal works. For maintenance, a traverse is
just another embankment, but the term is important for our vocabulary.
Magazine. A buried waterproof chamber or adjoining chambers constructed with planks
and roofed over with logs and earth. The magazine stored ordnance for the artillery and
varied in size according to the caliber of the guns. Each caliber of guns in the work
tended to have its own magazine or its own separate chamber in a complex magazine. A
surviving magazine most often appears as a large mound of earth (square-ish or oblong)
located behind or between gun platforms. The magazine was accessed by a door at its
rear, which can show up as a depression in the earth, perhaps protected by traverses. The
top of the mound usually exhibits a deep subsidence where the interior roof has
collapsed. Where there is no subsidence, the roof of the interior chamber may yet be
intact, though I don't believe any archeologist has looked to see. Not every artillery work
had a magazine.
Bombproof. A linear area roofed over with logs and earth designed to protect a garrison
from incoming artillery shells. Most often built into the rear of a traverse or in the side of
a hill, a bombproof consisted of an overhanging roof of logs and earth beneath which
soldiers could huddle. Sometimes, the rear of the bombproof was enclosed and provided
with doors. Identifying a bombproof in the field can be a bit problematic. I look for
evidence of collapsed earth behind a traverse (or consult a plan of the fort if lucky enough
to have one). A magazine would always be built first before work began on a
bombproof. Not every artillery work had a bombproof.
Wells. Semi-permanent fortifications often had a well or cistern if there was not a nearby
source of water. Usually, a round subsidence without a visible embankment.
Dams. Water courses running across the front of a line of earthworks could be dammed
and impounded to present a formidable obstacle to attack. We have observed several
examples of military dams, which would be, shall we say, embankments.
Covered Way. A military road leading into an earthworks complex usually wide enough
for a wagon. A covered (sometimes "covert") way often used existing ravines as an
approach to avoid incoming fire but would be protected by an embankment on the side of
the enemy when crossing an exposed position. It was not literally covered or roofed
over. A covered way entered the rear of an enclosed earthwork through a sally port.
Follow a covered way to the rear and expect to find other associated features. A military
road was often corduroyed with logs. I have seen a corduroy road from the War of 1812
with the logs intact, so who knows what might survive out there.
Infantry Defenses. The principal difference between artillery and infantry defenses lies
in the overall bulk (size, mass, volume) of the earthworks. Because an artillery battery
inevitably attracted counter-battery fire and a gun provided a larger target than a man,
more labor was lavished on artillery defenses than associated infantry defenses.
Prepared artillery and infantry defenses might both be constructed with an exterior ditch
when time and labor allowed; this method simply provided a stronger profile (crosssection),
and the engineers preferred it. An interior ditched work, though, could be
thrown up and revetted in half the time (every shovelful down was also a shovelful up).
A simple interior ditched work was often termed a rifle trench. Most rapid infantry
fortifications began with an interior ditch for speed but could be improved if occupied
under fire for any length of time with digging in front to fatten the embankment. Rapid
infantry works also have their associated traverses, bombproofs, gun platforms and
ramps, as well as numerous command pits, skirmish pits, bunkers, mess pits, slits
(latrines), and various arrangements in the rear for logistics. As with all rough groupings,
the lines between artillery and infantry and logistics get blurry when dealing with rapid
fortifications. Single guns might be spread out along the infantry line to provide close
support. These guns still needed platforms and usually exhibit exterior ditching. When
walking along a line of earthworks, the first indicator you may have of the presence of
artillery may be a sudden shift from interior ditch (infantry) to an exterior ditch. Closer
inspection reveals corroborating details.
Firing Step. A shelf behind the parapet upon which a soldier could stand to fire his
weapon, most often associated with prepared works and exterior ditching. After firing,
the soldier could step down out of sight to reload his weapon. The presence of a firing
step (called the banquette in French) is an indicator of good condition as this shelf tends
to slough off or get covered over as the embankment erodes.
Communications ditch or trench. A ditched pathway leading from a rear area to the
front line to allow safe entry, or a trench connecting two or more parallel lines of works.
In some cases, these were built to move troops from their camps and branched left and
right to direct troops to specific points along a defensive line. Communications ditches
were sometimes built in a zigzag pattern (to prevent the enemy from firing the length of
the trench) with the embankment switching sides with each zig and zag to stay between
the man in the ditch and the enemy. Communications ditches, covered ways, and
military roads form the transportation system of an earthworks complex. Again,
maintenance would focus on the embankment and associated ditch.
Pits and other "holes in the ground." For as long as a rapidly entrenched position
remained occupied, the soldiers continued digging. At Cold Harbor, for example, the
position was occupied for ten days. In an area of only 144 acres, we mapped 3  miles
of linear embankments and more than 600 pits or holes of various uses, some which now
we can only conjecture. When mapping we punted, called everything a "hole," and
recorded its dimensions. Similarly, maintenance would treat the survival and not concern
itself with interpretation. Understanding, the use of some of these humble holes in the
ground (so often ignored) can truly broaden our appreciation of the resources.
Command pits (for want of a readymade descriptor). These are excavations that reveal
command and control. For a single man, 2-3 feet deep and 3-4 feet in diameter with an
embankment to the front. At regular intervals maybe five yards behind an infantry line
one might find pits for non-commissioned officers; every 20th pit or so will be a bit
larger, a bit farther to the rear, perhaps a captain or lieutenant. Farther behind these pits
on the center of a regimental front may be a pit large enough for three or four men, this is
the colonel and his aides. On the center of a brigade front and farther back still may be a
full-blown bunker for 6-8 men that housed the brigadier general. (I define a bunker as an
excavation large enough to shelter 6 or more men, approx. 60 square feet.) The larger the
hole, the farther from the front, the higher the rank of the officer. Communications
ditches often extend from a command bunker to the front lines.
Skirmish pits. Sometimes called rifle pits and known at a later date as foxholes. A line
of skirmishers or pickets stood between the enemy and the main line of resistance. These
soldiers were stationed at intervals of 5-15 yards to provide early warning of attack and a
first line of defense. Each man dug an individual pit 2-3 feet deep, throwing an
embankment to his front. Sometimes three or four men would link their pits together to
create a slit trench. These pits can have surprising personality. As the Civil War
progressed, more fighting occurred on the skirmish line, and the skirmish pits grew in
size to accommodate a three-man firing team. Somewhere behind the skirmish line and
in front of the main line would be the picket reserve of 10-15 men gathered into their own
proportionately sized bunker, ready to reinforce the picket line if needed.
Mess pits. If occupying a position for any length of time, the soldiers tried best as they
could to make themselves comfortable. Often at regular intervals dug into the back of an
interior ditch, one finds dugouts large enough to shelter four or five men. Here is where
each small group of comrades (called a mess) cooked their coffee and ate their rations. In
a prolonged siege, such as Petersburg, these pits might be roofed over to create minibombproofs.
Slits. Long, narrow trenches that were filled in after use. I have only ever seen one
confirmed latrine, which was merely a subsidence in the ground, but trust me, they are
Supply caches. As one moves farther to the rear of any elaborate system of fortifications,
one encounters structures associated with the movement of food and ammunition to the
front. These features were sheltered by the terrain from enemy fire (in defilade) and can
be associated with a pre-existing road, military road (oft-times visible and traceable), or
covered way. Large, square or rectangular excavations (bunkers) adjacent to a road
served as supply caches where wagons offloaded before boxes were distributed. From
these large central bunkers, supplies were distributed through the transportation network
to smaller bunkers closer to the front.
Hut sites. Soldiers built huts behind the lines for winter quarters. Huts were often simple
excavations roofed over with canvas, but some could be elaborate structures of logs or
planks with a door and glass windows (looted from somewhere nearby), and even a brick
fire place. Every hut site needed at least a hearth for heating and cooking, often
improvised from slabs of field stone. Blackened hearths are a common feature. A
chimney might be improvised with stones or a couple of barrels, stacked and daubed
inside with mud. I have seen surviving barrel hoops in the debris of some hut sites (that
had recently been disturbed by the relic hunters). Hut sites were arranged side by side in
"streets" by unit, ceding the higher ground to officers' huts. Hut encampments are an
archeological gold mine, which is why they tend to be systematically looted by
commercial collectors.
Note on Obstacles. Obstacles in front of the earthworks were integral to every system of
fortifications and served the same general purpose as razor wire today-to delay an
attack and hold attackers under fire for as long as possible. Obstacles ranged from
improvised to elaborate. To make a slashing, axmen cut down the trees in front of the
works to create a field of fire, felling all toward the enemy so that attackers would have to
struggle through a maze of branches. An abattis or abatis (ah-bah-tee' or a-bat'-tus) was
a bit more formal: dig a trench, insert the trunks of smaller trees side by side leaning
toward the enemy, anchor the trees in the ground, backfill the trench, and then trim,
sharpen, and interlace the branches. Ouch. Try to get through one of those. I have only
observed one example of possible abattis evidence in the field, this at Fortress Rosecrans
in Murfreesboro, a shallow ditch in front of the parapet where the map showed the
presence of an abattis. A fraise was similar using larger tree trunks, evenly spaced, and
sharpened to points. Chevaux-de-frise (sheh-voh-de-freez') show up in many
photographs of the era: take a big log, bore four rows of holes, insert sharpened stakes,
stand them up, and chain them together. A lot of work. Damnyankee soldiers also strung
telegraph wire between stumps to create wire entanglements. (They were the only side
with an excess of telegraph wire.) Most all of this wire would have been recycled
immediately after the war.
Note on Condition
Conditions in the field range from near pristine to pitiful remnants. An embankment in
good condition will present a uniform appearance, exhibit crisp, sharp angles, and have
many visible associated details, such as well-defined gun platforms, ramps, embrasures or
a surviving firing step, or nicely squared off mess pits. The slope angle of the interior
and exterior slopes of the embankment can help quantify the condition of an earthwork or
be used to monitor condition over time-typically the steeper the slope, the better the
condition of the earthwork. Where a revetment has deteriorated in situ, the interior slope
retains a steeper angle than the exterior slope. Where the revetment had been stripped out
(common after the war) both slopes are similar.
Embankments are continually eroding from mass wasting, which is the general tendency
of the earth in an embankment to seek a lower level due to gravity. Mass wasting cannot
be completely stopped but only delayed. In my experience, there is no natural angle of
repose for an earthwork embankment, that is, no point at which the slope of an
embankment reaches equilibrium and ceases to erode. In the absence of protective
ground cover, the embankment will continue to flatten over time until a shallow trough is
all that remains. A very large earthwork will obviously take much longer to flatten out
and can absorb more damage than a simple rifle trench, but if mistreated can suffer
extensive erosional damage in a short period of time. Mistreatment usually consists of
stripping away the ground cover and providing no replacement. Water then penetrates
the surface of the embankment, percolates down through the earth, seeps out at the base,
and triggers slumping. This dislodged material is then washed down slope and begins
filling up the ditch. The worst cycle to get into (and we've seen it many places), is to cut
off the trees so "we can see the earthwork," let everything grow back up, and then cut the
trees off again so "we can see the earthwork." A few such cycles and there's not much
earthwork left to see.
We have found the best examples of earthworks in good condition in hardwood and
mixed forests that have been largely undisturbed by logging. Earthworks covered by
sufficient leaf litter suffer the smallest amount of mass wasting of any ground cover (See
Helms and Johnson at
Tree-covered earthworks of course are subject to the localized trauma of tree throw. In
my opinion, tree throw damage can be repaired but that is another matter for another
time. By extension, mulch, if properly applied and assiduously maintained, should have
the same result as leaf litter but that's a lot of labor. Tree-clearing followed immediately
by hydro-seeding with turf grasses can provide a viable cover and attempts have been
very successful. But problems can develop in some places with getting the grass to take,
or the grass can burn out on south-facing slopes during an arid summer, leaving the
embankment vulnerable. The least successful experiments in the park service that we've
seen firsthand have attempted to establish a viable native grass cover. It takes years and
erosion continues apace. I'm obviously not a purist. We've seen some invasive species
over the years that would make a darn good cover.