How Archaeologists Actually Dig: Field Methods Explained
Television archaeology tends to show discovery: the brush on the bone, the gleam of gold, the director's sharp intake of breath. What it rarely shows is the hour of planning before a single trowel goes into the ground, the numbering system that makes the discovery mean something, or the weeks of post-excavation analysis that determine whether the breathless moment becomes a published result. Understanding field methods makes a site visit — or a museum display — considerably more interesting.
Laying out a grid
Before any ground is broken, a site is surveyed and a grid is established. In British practice this is typically a 5 m or 10 m Cartesian grid referenced to national grid coordinates or to a fixed site datum point. A total station — an electronic theodolite with distance measurement — is used to set out grid pegs to centimetre accuracy. The grid allows every find and every cut to be recorded by coordinates, so that the three-dimensional position of objects in the ground can later be reconstructed on paper.
Larger landscape projects use differential GPS and, increasingly, structure- from-motion photogrammetry: photographs taken from multiple angles are processed by software into a textured three-dimensional model of the ground surface or open trench. This allows measurements to be taken from the model rather than only from the physical ground, and creates a permanent archive of conditions at any given moment in the dig.
Opening and cleaning the trench
Topsoil removal is often done by machine — a tracked excavator with a toothless ditching bucket — but only after geophysical survey (magnetometry, resistivity, ground-penetrating radar) has located features below and established that nothing vulnerable lies near the surface. Machine excavation stops when the operator and site director agree that buried features are close; hand excavation takes over from that point.
Cleaning means taking a trench surface from rough-cut to the point where the edges of features — pits, post-holes, wall foundations, hearths — are clearly visible against the natural subsoil matrix. This is done with trowels, typically Marshalltown pointing trowels, worked at a low angle. Cleaning is one of the most skilled acts in excavation: the ability to read subtle colour, texture, and compaction differences between archaeological deposits and the surrounding geology takes years to develop reliably.
Context numbering and the Harris matrix
Every discrete deposit, cut, or surface is assigned a context number — a unique integer within the site number system. A pit is two contexts: the cut (the physical void dug into the ground by whoever dug the pit) and the fill or fills (the deposits that accumulated in it afterwards). A wall is a context; the surface it was built on is a context; the demolition rubble above it is another.
A context record sheet is completed for each number: shape, dimensions, colour by Munsell notation, texture, inclusions, interpretation, and relationships to adjacent contexts. Relationships are the key: what is earlier, what is later, what is contemporary? These relationships are formalised in the Harris matrix, a diagrammatic tool developed by Edward Harris in the 1970s. The matrix arranges all site contexts as nodes in a directed graph, with lines expressing stratigraphic sequence. A complex urban site can produce a matrix of thousands of nodes; reading it tells you the site's entire sequence from earliest to latest at a glance.
Single-context recording
Single-context recording, standard in British practice since the late 1970s and widely adopted internationally, means that every context is drawn and photographed independently as it is excavated, rather than drawing composite "phase plans" that combine material from several periods. The result is a comprehensive record that can be analysed at any later date to reconstruct the sequence, even if the archaeologist who dug it is unavailable. It also means that context sheets can be completed on site at the moment of excavation, when the relationships are physically visible, rather than reconstructed later.
Sieving and flotation
Not everything found in the ground is large enough to be spotted by eye. Soil from features of particular interest — hearths, pit fills, floor surfaces — is dry-sieved through meshes of typically 10 mm, 5 mm, and sometimes 2 mm to recover small finds, animal bone, and industrial residues. Wet flotation recovers even smaller material: the soil is agitated in a water tank and the light fraction — seeds, charcoal, insect remains — floats off into a sieve while the heavier fraction, including small bone fragments and fired clay, is caught separately. The resulting environmental assemblage can reconstruct diet, land use, and local ecology in ways that the structural archaeology alone never could.
Finds management
An object found in situ is left in place until its position is recorded by total station and photographed in context. It is then lifted and placed in a finds bag with a durable paper label carrying the context number, find number, date, and site code. Small finds — coins, worked bone, personal ornaments, ceramics with complete profiles — receive separate small-find numbers and more detailed records. Bulk finds such as tile fragments and animal bone go into bags by context and are counted, weighed, and sorted in the site finds hut.
Fragile objects are packed in acid-free tissue, consolidated with reversible adhesives if necessary, and prioritised for specialist examination. Metal objects are particularly vulnerable to corrosion once lifted from stable anaerobic conditions; they go directly into controlled storage and to a conservator.
The site notebook and end-of-day review
Every digger keeps a context record; the site director or trench supervisor keeps a site notebook that records decisions, interpretations, problems, and observations too discursive for a context sheet. At the end of each day, the team reviews what has been found and what the next day's strategy should be. This is when the matrix is updated, relationships are confirmed, and provisional phase interpretations are tested against the physical evidence.
From the trench to publication
Post-excavation can take longer than the dig itself. Specialists report on the pottery, animal bone, plant remains, metalwork, radiocarbon dates, and human skeletal material. The site director integrates these into a stratigraphic narrative. The result is a grey literature report deposited with the local Historic Environment Record, and ideally a peer-reviewed monograph or journal article that places the site in its regional and period context.
Many of the world's famous sites on the map are still active excavations. Visiting a working dig — even as a background visitor — makes the scale and the method legible in a way no museum display can. Some sites offer formal public open days; others run field school programmes. Either way, the trowel and the context sheet together are the basic instruments of the discipline.