Critique of Archaeological Reason
7. Themes

Stratigraphy

Laerke Recht – January 2017

Back to top: Stratigraphy

Introduction

Stratigraphy in archaeology is part of the excavation process and subsequent analysis. It ultimately derives from geological methods, and builds on main the principle that top layers are superimposed on earlier ones. The layers (or ‘strata’) thus reflect a relative time sequence of depositions and are invaluable in determining the relations between artefacts and features in archaeology.

Back to top: Stratigraphy

Historical background

Stratigraphy originates in the geosciences, where its main principles were first articulated by Nicolas Steno in 1669 as a solution to understanding the presence of fossils within rock (Lucas 2012, pp. 76-77). It was not until much later than its significance was recognised in archaeology, where study of soil deposits or differences in soil did not initially make up part of the investigations.

Although stratigraphic excavation is now an integral part of archaeology’s identity, it did not start as such. Archaeology (or anthropology/prehistory/antiquarianism) started with an interest in typology and classification. Material in private collections and museums were very suitable for this kind of pursuit. At some point focus changed from artefacts to cultures, and this in turn effected a change in collection methods, eventually leading to more controlled, stratigraphic excavations (see Lucas 2001b). Famous contributors in the Anglo-Saxon world include Augustus Pitt Rivers, Mortimer Wheeler, Kathleen Kenyon and Edward Harris.

Some of the ideas of stratigraphy in archaeology were expressed at least as early as the the early 19th century. Christian Jürgensen Thomsen’s 1836 Guide to Northern Antiquities included a guide to excavation that involved recording of object association, and his successor, Jens Jacob Asmussen Worsaae emphasised stratigraphy and the inclusion of sections during excavation (1843). In these cases, we still see the focus on typology (Thomsen is often credited with “inventing” the three-age system of Stone, Bronze and Iron, which is in fact based on artefact typologies), but from the very beginning, at least some excavations had stratigraphic elements or awareness (indeed, it could be argued that if they did not, we would not define them as excavation, but rather along the lines of looting). In the first half of the 20th century, Pitt Rivers, Wheeler and Kenyon systematised excavations through use of grid trenches and development of the section drawing, and in the late 1970s, Harris invented the now commonly used Harris Matrix (see below).

Back to top: Stratigraphy

Definition

Since the concept of stratigraphy is initially derived from geology and the geosciences, it may be useful to compare definitions from those:

the science dealing with the description of all rock bodies forming the Earth's crust - sedimentary, igneous, and metamorphic - and their organization into distinctive, useful, mappable units based on their inherent properties or attributes. Stratigraphic procedures in geosciences include the description, classification, naming, and correlation of these units for the purpose of establishing their relationship in space and their succession in time. (Salvador 1994: 137, quoted by Stein 2005, p. 182).

Another definition used for both geology and archaeology is offered by Gasche and Tunca (1983, p. 334):

n.; from the latin stratum and the Greek graphia, stratigraphy is the descriptive science of strata. It deals with the form, the arrangement, the distribution, the chronological succession, the classification, and the interrelation of strata (and of all other associated bodies), in normal sequence, having regard to certain or all characteristics and particularities possessed in their contents. This definition includes the study of the origin, the composition, the environment, the age, the history, the relations with organic evolution, and a large number of other particularities of the contents of strata.

Stratigraphy as specific to archaeology (also sometimes called ‘archaeostratigraphy’) may instead be defined as:

the study and interpretation of stratified deposits. The emphasis is on superposition and succession for the purposes of dating. (Stein 2005, p. 182).

or:

the study of strata or the different layers and levels of occupation on an archaeological site and their relation to each other, and the determination of the archaeological sequence or order in which they were laid down (Joukowsky 1980, p.150).

In essence, stratigraphy as a field of study does not differ significantly in archaeology and the geosciences (contra Harris 1989, p.xii). The main difference is one of focus and the specific concern with human-made depositions in archaeology (see also McAnany & Hodder 2009). For these purposes, principles of archaeological stratigraphy have been adapted (see below). Some scholars advocate using a terminological distinction between strata which are geological without anthropic elements and those which do have anthropic elements (“ethnostratigraphy” - see Gasche & Tunca 1983). Stein further claims that in archaeology, stratigraphy is solely concerned with time, while geology uses both time and space (2005, p. 182). This does not really tally with how archaeologists work. Although geologists may compare strata across much greater space, this concern is also crucial in archaeology, where every stratum is placed both in spatial and chronological relation to other strata. The main goal may be the construction of the chronological sequence, but this is not possible without spatial dimension.

Theoretically, stratigraphy is independent of artefacts. That is, the strata can be studied in their own right and the sequence not depend on artefact content. In practice, the stratigraphy is often informed by study of the finds in them (in particular pottery and flint). It can also occur that strata are defined precisely on the content without discernible differences in soil. However, the sequence defined by stratigraphy is more important than that which can be established through artefact study. This is because the stratigraphy reflects the time of deposition of the soil and its contents, while typologies reflect time of manufacture which may not be the same as that of the deposition.

Back to top: Stratigraphy

Stratification

While stratigraphy refers to a field of study, stratification refers to the extant layers or strata on a site. Stratification exists on all sites (CAR, 4.2), but not all sites are excavated in a process that follows the stratigraphy, which is usually what is meant by stratigraphic excavation.

Back to top: Stratigraphy

Stratigraphic excavation

An important distinction must be made between stratigraphy and stratigraphic excavation. While stratigraphy is the study of stratification of a given site, stratigraphic excavation refers to the process or technique: the manner in which a site is excavated. The term stratigraphic excavation is rarely defined in publications, but generally refers to excavations where the layers are removed stratigraphically, meaning each stratigraphic unit is removed in the reverse order of their deposition. In contrast, some excavations use what is usually called arbitrary layers, where a set amount of soil is removed in strips or slices (e.g., 5 cm at a time). This is also considered stratigraphic excavation by some archaeologists:

We define stratigraphic excavation as removing artifacts and sediments from vertically discrete three-dimensional units of deposition and keeping those artifacts in sets based on their distinct vertical recovery proveniences for the purpose of measuring time. (O'Brien & Lyman 2002, p. 149).

The distinction is well described by Lucas, “it is important to recognise that while [the slice/arbitrary method] is not stratigraphic excavation, it is still excavation which is aware of stratigraphy” (2001b, p. 33). This often unstated difference in what is meant by stratigraphic excavation has led to some confusion because it has implications for what is considered the first example. As noted by Lucas, this is related to a difference in practices in North America and Europe, where the slice method has been more prominent in North America.

Essentially, the difference is one of strategy or method, or excavation process (as Harris calls it). Trenches can be laid out and physically excavated in a number of ways, but in modern excavations, the method is always what Lucas calls stratigraphically aware. This means that the stratigraphy is carefully recorded and part of the analytical process.

The discussion here largely focusses on stratigraphy and stratigraphic excavation as it has developed in the Anglo-Saxon sphere. Archaeological practices vary greatly in different countries (even within Europe), and this can also be reflected in the approach to both stratigraphy and the excavation process - for examples of these differences and their histories, see e.g., Carver 2012, Carver & Lang 2013.

Back to top: Stratigraphy

The stratigraphic revolution

This term is sometimes found in the literature. It refers to the work done by Edward Harris in the 1970s and 80s on strict stratigraphical principles tailored specifically to archaeology rather than geology (see e.g., Harris 2013). However, it has also been applied to earlier work, especially in North America, including that of A.V. Kidder and others (Lyman and O’Brien 2006, p. 205; Lucas 2001b, p. 47).

Back to top: Stratigraphy

Stratigraphic principles

Perhaps most famously, stratigraphy in archaeology was theorised and its principles meticulously laid out by Edward Harris. Developed from geology, Harris created a revised and adapted set of four principles (or “laws”, as he calls them) for stratigraphy in archaeology. They are revised because geological principles are not directly applicable to archaeological sites (contra Tunca), and this is due to the man-made element of archaeology. Stratification and artefacts made by humans are not subject to the same natural processes as those studied by geologists.

The four principles of archaeological stratigraphy as defined by Harris are:

The Law of Superposition: In a series of layers and interfacial features, as originally created, the upper units of stratification are younger and the lower are older, for each must have been deposited on, or created by the removal of, a pre-existing mass of archaeological stratification (Harris 1989, p. 30).

The Law of Original Horizontality: Any archaeological layer deposited in an unconsolidated form will tend towards a horizontal position. Strata which are found with tilted surfaces were originally deposited that way, or lie in conformity with the contours of a preexisting basin of deposition (Harris 1989, p. 31).

The Law of Original Continuity: Any archaeological deposit, as originally laid down, or any interfacial feature, as originally created, will be bounded by a basin of deposition, or may thin down to a feather-edge. Therefore, if any edge of a deposit or interfacial feature is exposed in a vertical view, a part of its original extent must have been removed by excavation or erosion, and its continuity must be sought, or its absence explained (Harris 1989, p. 32).

The Law of Stratigraphical Succession: A unit of archaeological stratification takes its place in the stratigraphic sequence of a site from its position between the undermost (or earliest) of the units which lie above it and the uppermost (or latest) of all the units which lie below it and with which the unit has a physical contact, all other superpositional relationships being redundant (Harris 1989, p. 34).

These principles provide a basis for most students of archaeology today, and are an integral part of the excavation process. Harris developed the principles from existing ones in geology. In practical terms, some of the principles from geology may be a useful supplement, as explained by Bahn (2001, p. 426, here renumbered), they are:

1. Law of cross-cutting relationships: a feature that cuts across or into a bed or stratum must be younger than that bed or stratum.

2. Included fragments: fragments, material or debris from an older bed may be incorporated in a younger bed, but not vice versa.

3. Correlation by fossil inclusions: strata may be correlated based on the sequence and uniqueness of their floral and faunal content.

1 and 2 are very useful principles when establishing the stratigraphic sequence. They are both absolute in the sense that they do not contain exceptions. 2 is further an important element for dating: a deposit cannot be older than its youngest content - that is, if a pottery sherd in a clearly delimited and closed deposit dates to Late Mittani, the deposit cannot be dated prior to this date, even if the majority of sherds belong to the Akkadian period. For 3, fossils should be substituted with cultural finds, e.g., pottery, stone tools etc. In contrast with the others, 3 is more a guideline than a universal principles. Thus, when ‘phasing’ takes place (the stratigraphic sequence being divided into major occupational or significant phases of a site), correlating contents and soil from different areas of a site may be part of the process. To a more limited extent, this can also be done between sites.

Back to top: Stratigraphy

Observation and inference

The stratigraphic principles allow us to infer deposition from emplacement. The distinction is significant because, as outlined in CAR (2.3.1; 4.2), emplacement is associated with observation (insofar as uncontaminated observation can be said to be possible), while deposition involves inference and interpretation. With the best possible observations, we can use the laws of stratigraphy to infer a sequence of deposition. In other words, we do not directly see that one feature is younger than another. But careful observations mean that we can use e.g. the law of cross-cutting relationships to infer that a feature which sharply aligns several other features vertically may indicate a pit cutting earlier strata, itself representing the latest action. If the observation of emplacement is left out (i.e. simply stating “Feature 301 is a pit”), it becomes impossible for others to evaluate the statement or see how the conclusion was attained - and this exacerbates the problem of excavation as an unrepeatable experiment.

Back to top: Stratigraphy

Harris Matrix

Harris also created a diagram system for recording stratigraphic information during excavation, which he called, and is now generally known as, the Harris Matrix. The diagram consists of boxes which each contain a stratigraphic unit (that is, a context or feature number). The relation between units is then illustrated with connecting lines. Only three types of relationship can occur: superposition, no connection or direct correlation. The matrix is an extremely useful way of schematically illustrating relations between features, both horizontally and vertically. The Harris Matrix or some version of it is now common at many archaeological excavations (for theoretical examples, see here). However, in some types of excavations and stratigraphic units, its practical use can be limited, especially due to difficulties identifying strata (see below under interfaces).

Back to top: Stratigraphy

Stratigraphic drawings: sections and plans

Back to top: Stratigraphy

Section drawings

Beside the now commonly found matrix, the section (or profile) drawing is the illustration of stratification at a given site. Section drawings appear quite early, before proper stratigraphic excavations as per the definition outlined above. Harris defines a section drawing as “a drawing of a vertical soil profile, as exhibited by cutting down through a mass of stratification. Two things are shown in a section: a vertical plane view of the strata, and the various interfaces between the bodies of the strata” (1989, p. 69). Section drawings have been a mainstay of archaeological recording for many decades, but may now be less common in certain circumstances which seem to mainly pertain to practices in the UK and is related to the use of single context recording.

On excavations, section drawings are often done either at the end of an excavation season, when a ‘complete’ section is available or perhaps before a baulk is removed, or cumulatively, adding to the drawing as the digging proceeds. The advantage of the latter is that stratification is usually clearest when first excavated, as weather and exposure can change the colour and texture of the soil. Harris identifies realistic and stylised types of section drawings. The former shows the layers using different graphic textures, without lines between the layers, while the latter has clear lines showing the limits of each layer and assigns a number, as suggested by Harris (cf. the suggested formats found in various field manuals).

A potential problem with drawings is that they show only what the excavator sees/interprets, and in the process of creating the drawing, decisions about what to include and exclude, and how to represent the elements are made. This, of course, is not a problems that only pertains to drawings, but to all archaeological recording.

Another option is to use photographs and draw explanatory notes and stratification lines directly on top - this approach is used at Mozan for both section drawings and plans (see fig. 1 below). The advantage here is a lower risk of misrepresentation due to the transition to a different medium. Better access to high resolution photography means better visibility than previously on such photographs, although boundaries may still not be as clear as in firsthand encounters (of course, elements such as hardness and texture do not really display but require verbal or graphic explanation).

J2 view
Fig 1. View photo from Tell Mozan Unit J2, with marked lines showing features both vertically and horizontally (the equivalent of section and plan).

Back to top: Stratigraphy

Plans

Plans instead show the horizontal dimension of a site. There are may different conventions and formats for drafting plans (see again various field manuals for examples). Harris identifies three types which he calls multiple feature, composite and single-layer. The multiple feature plan shows all features of a trench or site in one plan, superimposed on each other - since in most cases, all features were not in use at the same time, this kind of representation can be misleading. More commonly, ‘final’ plans show one phase of occupation (e.g. consisting of a major floor surface and associated features), what Harris calls the composite plan. The single-layer plan is the one recommended by Harris and now in general use within the so-called single context system of excavation (in particular widely used in the UK). In this kind of plan, one plan (as part of a context sheet - see detailed instructions and plans at UCLAN -> 09 Context Recording) shows only one feature. The plans can then be superimposed to create the sequence or a selected number of features which are determined to make up a phase. As with the sections, another option is to use photographs which are then edited.

Back to top: Stratigraphy

Interfaces

Harris’ comment on interfaces is here a significant aspect of recording methods. Interfaces are the ‘surfaces’ where units of stratification meet, and they are required to understand the stratigraphic sequence. The interfaces represent relations, and it is precisely relations between strata that we try to understand in stratigraphy. In both plans and section drawings, Harris strongly advocates the use of firm, uninterrupted lines to depict interfaces, and numbers to demarcate layers and their sequence. The relevance of this is well illustrated in his figure 30, p. 83: the lines reveal the sequence of stratification, even if the numbers were not present. Conversely, the drawing without interface lines cannot tell us the sequence (at least not without numbers).

The importance is thus theoretically quite clear. In practice, however, interfaces are often not so clearly defined. The exact boundary of a given feature can be difficult to identify (even for the most experienced excavator) - mud brick structures in Near Eastern archaeology are notorious for this in certain circumstances, but examples abound (see also Bollong 1994 for an example of lack of visible stratification and possible solutions). The question of interpretation and subjective observation here comes to the fore. What one person sees as a clear limit, another may see as transitional or even part of the same unit (see e.g., Gero 1996 for possible gender differentiation in observation and recording). Harris’ position does not allow for uncertainty: “If the excavator cannot define any ‘clear-cut division’ in a section, the character of the stratigraphic excavation is to be questioned” (1989, p. 78). An unwillingness to recognise ambiguities in the excavated record can lead to serious misrepresentation of the stratification and eventual overall interpretation of a site (see also Rowe 1991 for problems in defining units due to what he calls ‘mixing, filling and collecting’; problems of vertical displacement are discussed by Villa 1982). Instead, a certain level of flexibility is required which allows for situations that do not fit neatly into clearly delimited ‘boxes’. As Carver writes,

How can unclear layer boundaries be represented graphically? You can use a dashed or dotted line, but how do you draw a transition zone that is more than a line-width wide? Or what is the 3D equivalent of a dotted line in 3D computer graphics? The problem is with the way that theory so often (too often?) conflicts with reality. There are also ontological problems in defining layer boundaries. Usually we assume that they are clear, but this does not have to be the case [...] In addition, archaeologists recognise that layers are defined on the basis of pragmatic and informal decisions made in the field (2012, pp. 24-25).

As Carver here notes, in practice, solutions are usually found in the field, but the theory behind the decisions is lacking. This may also be part of the reason why the use of the Harris matrix appears to be less common in North America, where excavation by artificial layers is also more frequent - this especially seems to be due to the nature of the record, with deep units that cannot always be clearly defined. As such, a different type of excavation is deemed more practical, although stratigraphy is still essential. Another problem may appear when focussing solely on interfaces and relations between deposits: the deposits themselves and their formation processes are ignored (for a discussion of this, see Lucas 2012, pp. 83-84).

Back to top: Stratigraphy

Measuring time

Back to top: Stratigraphy

Continuity vs. discontinuity

Stratigraphy is one of the most important ways of measuring time in archaeology, but it is not the only one (see Pettitt 2005 for more details). Each method has its own advantages and disadvantages. Stratigraphy is a relative measurement of time. As we have seen, it provides a sequence of strata. We thus know the order in which depositions occurred, but stratigraphy cannot tell us when in absolute dates. Generally, it also does not tell us the length of time for each individual stratum - a single, very deep stratum does not necessary correlate with an extended length of time. Other relative ways of measuring time in archaeology include seriation and index fossils, both of which use artefacts to create an order of things (see O’Brien & Lyman 2002 for a detailed discussion of these dating methods, and Rowe 1991 for a comparison of stratigraphy and seriation as credible ways of establishing sequences).

The different ways of measuring time has implications for interpretations and the kinds of questions that can be asked. Stratigraphy is a linear, discontinuous way of measuring time. The emphasis on interfaces and divisions into discrete units promotes narratives that focus on change and disjuncture because each unit can be seen as marking a new ‘event’ rather than a continuous process. This is the equivalent of the biological concept of ‘punctuated equilibrium’. Absolute dating methods also only offer discontinuous measurements of time. As a counterpoint, seriation can demonstrate continuity because it measures gradual change in artefacts, regardless of the type of seriation employed.

So far, no one way of measuring time in archaeology can work on its own. Stratigraphy must be checked against absolute dating methods and/or the study of artefacts. Conversely, absolute dating methods have limited merit without the corresponding stratigraphy. Seriation equally requires checking against other methods, including the stratigraphy of a site.

Back to top: Stratigraphy

Micro vs. macro

Interpretation also depends on the scale at which we measure time (see e.g., Lucas 2005, 2008). In very fine resolutions of time, we can measure the small scale, while low resolution is better suited for measuring the large scale. It can be argued that the nature of the archaeological record lends itself most easily to the study of large scale processes and events, and that this is truer the further back in time we travel (Bailey 2008). In a simplification of archaeological theoretical ‘camps’, study of the macro scale has largely been associated with processual archaeology (with a focus on structures and broad processes, including questions about ‘revolutions’) and of the micro scale with postprocessual/interpretive archaeology (with a focus on individuality, agency and embodied experience). Although archaeologists widely disagree about what kinds of questions archaeology should be asking (see e.g. the variety of responses to this issue in the conversations in Rathje et al. 2013), the questions of both ‘scales’ are surely important for understanding humans and human history. Stratigraphy can be used for different time scales in theory, while in practice the resolution possible very much depends on the record of individual sites.

Back to top: Stratigraphy

References

Back to top: Stratigraphy

See also

Back to top: Stratigraphy

Field manuals

Back to top: Stratigraphy

Relevant sections in CAR

  • 5: Stratigraphic analysis.
  • 2.3.1: The three levels of trace analysis.
  • 4.2: The record: stratification and stratigraphy.

Back to top: Stratigraphy