Archaeological Sediments and Soils

Analysis, Interpretation and Management

Anthony J Barham, Richard I Macphail, Anthony J Barham, Richard I Macphail

• Verlag: Routledge
• Jahr: 2016
• ISBN:
  9781315434438
• Auflage: 1
• Thema: Social Sciences
• Thema: Archaeology

A collection of papers focusing on the links between archaeology and the study of geological sediments and soils.

Part I

Recent advances in micromorphological interpretations of soils and sediments from archaeological sites

Richard I. Macphail and Paul Goldberg

Preface

If this chapter is to be of value it must be first realised that the soils and sediments that make up a substantial part of an archaeological site are often the most commonly wasted resource. Only too often sites are excavated by machine, or trowelled away without any realisation that the material surrounding the artefacts represents the essential context of the site. In reality, once the soil or sediment is removed, the artefacts are out of context and photographs or drawings are often inadequate when reconstructing this context. As we will attempt to demonstrate, thin sections and the impregnated blocks from which they are made, help preserve that context.

Introduction

The aim of this paper is to emphasise the recent philosophical and interpretive advances that soil micromorphology has made in its application to solving archaeological problems. We believe that this continuing improvement has come about by way of a number of related factors (Courty, 1992):

i) recent results from experimental and ethnoarchaeological studies have been targeted to answer specific archaeological questions;

ii) in more recent detailed studies of archaeological sites there has been increased control of the stratigraphy, stratigraphic setting and the archaeology, and an increased awareness of the number and complexity of the processes and events that make up a site stratigraphy (site formation processes);

iii) associated with the latter there has been improved feedback from, and closer collaboration with, other specialists involved in multidisciplinary projects.

After a brief historical and methodological review of soil micromorphology in archaeology we elaborate on these three factors and provide examples. We close with a discussion of soil micromorphology as a management tool for archaeological projects and give some recommendations for future development and applications based on our understanding of current work and a recent questionaire addressed to micromorphologists (Gebhardt and Macphail in prep.).

Review and methodology

A detailed history of micromorphology in archaeology and its methodology is presented in Courty et al. (1989). Slightly more recent and condensed reviews include those of Macphail et al. (1990a), Davidson et al. (1992) and Goldberg (1992). In addition, a critique of soil micromorphology as a developing scientific technique in archaeology has been presented at the Royal Society (Courty 1992).

Kubiena (1938, 1953) was the first to study soils in an undisturbed state as a new research area in pedology. It was not until the 1950s that Cornwall (1958) and Dalrymple (1958) applied the technique to solving archaeological problems. Cornwall, although concentrating on the study of Holocene pedogenesis from datable palaeosols (those buried by monuments), was also one of the first to attempt to identify the influence of anthropogenic effects on the soil landscape (reviewed in Macphail 1987; cf. Duchaufour 1958, 1982; Dimbleby 1962). From these different studies two important and separate although sometimes interlinked directions of research have evolved:

i) the investigation of paleosols by Quaternary geomorphologists and soil scientists interested in the development of regional landscapes and climatic changes (Valentine and Dalrymple 1976; Fedoroff and Goldberg 1982; Mücher and Morozova 1983, Mücher 1986; Kemp 1986; Van Vliet-Lanoē 1985; Cremaschi 1987); and

ii) the study of Holocene palaeosols focusing on both localised and regional interpretations of human influence on pedogenesis (Romans and Robertson 1975a, 1975b; Macphail 1986; Macphail et al. 1987; Goldberg 1986).

Quaternary investigations have also included the first systematic study of cave sediments by this technique (Goldberg 1979; Flood et al. 1988). More recently the subjects of Pleistocene and Holocene pedogenesis have become more closely integrated (Fedoroff et al. 1990; Courty 1990; Van Vliet-Lanoë et al. 1992) and Holocene paleosols have been accepted as important data sources for the study of global soil and climate change (Catt, Fedoroff and Yaalon, 14th ICSS, Kyoto pers. comm. 1990) especially the role of humans in pedogenesis (Arnold et al. 1990; Macphail 1990a; Gebhardt 1993).

Most important, however, is the development of soil micromorphology as a technique that has been able successfully to characterise in thin section the effects of anthropogenic ‘activities’ (‘features’ in the sense of Courty et al. 1989). The study of various archaeological materials and activities has been investigated in such a way as to both complement and advance archaeological interpretations. These include the study of mudbricks and occupation floors (Goldberg 1979b; Goldberg and Whitbread 1993; Gé et al. 1993; Matthews this volume), soil landscapes produced by clearance and agriculture (Gebhardt this volume), abandoned farm deposits (Davidson et al. 1992) and in situ stable layers (Courty et al. 1992).

Methods of sample preparation have similarly improved over the past decade. Undisturbed samples are more often dried with acetone (to improve structural preservation) before being impregnated with a resin that produces an indurated block on curing (Murphy 1986). Large thin sections (up to 13 × 6 cm; Guilloré 1985) made from slices cut from these blocks can be easily ground down to 25-30 μm while maintaining excellent optical properties. Soil thin sections in micromorphological research are routinely examined under plane polarised light, crossed polarised light, oblique incident light and ultra-violet light. Unimpregnated material and polished blocks can also be examined, representing scrutiny from the field scale down to the micron scale (SEM and related methods [see below] including microradiography; Courty et al. 1989, ch. 4;) whereas particular features in thin sections themselves can be chemically analysed in detail by microprobe (on thin sections without cover slips) or quantified by image analysis (see below).

In the past, however, the lack of simple description methods devoid of restrictive soil genetic overtones has hindered development of the technique as a whole. Kubiena’s (1938) description method for example was biased towards pedogenesis. Similarly Brewer’s (1964) scheme and its updates (Brewer 1976) clearly mix interpretation with description. These schemes have recently been challenged by the purely descriptive method of Bullock et al. (1985). Their system is international in character and more user-friendly, employing terms that are closer to plain English (Murphy et al. 1985). As a consequence its use has permitted non-specialists to learn the technique more rapidly and has made the subject more intelligible (and more palatable) to the lay reader. The improved communication between soil micromorphologists and lay readers is still an outstanding challenge however.

Moreover, this increased use has had two consequences. Firstly, previously unstudied materials from archaeological sites can be readily described on the basis of this terminology. Secondly, because of the relative clarity of the descriptive terms, volume editors are far more likely to include texts by soil micromorphologists.

Site study strategies are now also influenced by the modern use of large thin sections. These sections should be at least 50 × 75 mm in size with 130 × 70 mm being the largest practical size in common use. These large formats are more representative of the variations and heterogeneity of archaeological deposits than are standard geological size slides (22 × 40 mm). With the ability of modern manufacturing procedures to make many high quality thin sections rapidly both on-site and off-site contexts can be sampled and studied in this way (Courty et al. 1989, 40–43). Complex archaeological and geoarchaeological sequences can thus be fully characterised (Courty 1990) to satisfy the requirements of any multidisciplinary archaeological project (Bell 1990, 1992).

Finally, we stress that thin sections become a permanent archive for the study of a site’s stratigraphy that may well have been extensively modified or destroyed by excavation. For example Cornwall’s thin sections (housed at the Institute of Archaeology, University College London) from the 1950s and Dutch soil survey studies from the 1970s still provide information which can be invaluable in present-day studies (eg Macphail 1990b, Macphail et al. 1987; Exaltus and Miedema 1994). (Some four hundred soil thin sections from archaeological sites investigated by J. C. C. Romans [eg Romans and Robertson 1975a, 1983a, 1983b] are now in the collection of the Artefact Research Unit of the National Museum of Scotland Edinburgh.)

Factors leading to improved interpretations

Experimental and analogue study programme

As noted above the original scientific approach to soil micromorphological interpretations in archaeology was Kubiena’s descriptive (1938, 1953) system which was utilised by Cornwall (1958). Subsequently a large amount of data has accumulated on the soil micromorphology of natural and agricultural soils. These data have been largely reported in the proceedings of the International Working Meetings since 1961 (eg Bullock and Murphy 1983; Fedoroff, Bresson and Courty 1987; Douglas 1990), and they constitute a base for current research in palaeoagriculture, occupation deposits and the study of earthworks.

Palaeoagriculture

Both Romans and Robertson (1983a, 1983b) and Macphail et al. (1987) initially attempted to identify ancient plough soils, using both agricultural (Jongerius 1970, 1983; Kooistra 1987) and independent environmental data. Further progress in the understanding of these soils was made by characterising soil microfabrics formed by tillage in modern experiments including the use of copies of ancient tools (Gebhardt 1990, 1992, this volume; Macphail et al. 1990b).

In the investigation of ancient pastoralism, on the other hand, little useful data have been generated from analogue situations apart from an experiment on the effects of animal trampling on soils (Beckman and Smith 1974). As a result an experimental programme was initiated to calibrate some of the micromorphological observations from archaeological sites (Courty 1990; Courty et al. 1992; Wattez 1992). The examination of burned and ‘fresh’ herbivore excrements, for example, permitted the identification of cattle and other herbivore dung in occupation sediments from protohistoric and historic sites from north-west India (Courty 1990). At Arene Candide Liguria, Italy(Maggi in prep.) several metres of greyish Neolithic cave sediment appear to be either homogeneous or stratified (Plate 1). Detailed micromorphological analysis patently indicates that these stratified sediments are mainly the result of herbivore stabling (Macphail et al. 1990, Fig. 2; Courty et al. 1992) perhaps associated with over-wintering (Fig. 1, Plate 2). Soil micromorphological experiments allowed not only sheep/goat and cattle stabling episodes to be differentiated, but also revealed that the stabling diet was mainly one of dried leaves and branches gathered by tree shredding (cf. ethnoarchaeological evidence, Halstead 1992). Previously this practice had only been inferred from pollen studies until evidence of this kind of leaf hay foddering was found in macro-botanical remains at the waterlogged Neolithic site of Weier, Switzerland (Robinson and Rasmussen 1989). These identifications of wood and leaf fodder for stabling seem to be understandably at variance with the type of byre deposits found at Hirta and Cuiltrannich in Scotland (Davidson et al. 1992), where foddering materials relate to a totally different period and environment. Seasonal pastoral activity at Arene Candide (as yet not fully defined; Rowley-Conwy pers. comm. 1992), is also independently indicated by the presence of deciduous teeth shed by sheep and goats (Rowley-Conwy 1992). Fortuitously, the stabling floor of the Moel-y-gar house at Butser Iron Age Farm, Hampshire, became available for sampling in 1990; cattle for the first three years and then sheep (Allen et al. in press). Here cattle and sheep had been alternately over-wintered from 1977 to 1990 (Reynolds 1981; pers. comm.). Each autumn an animal bedding layer some 40–50 cm thick was prepared from straw and compressed grass pellets; this was renewed halfway through the winter (Reynolds pers. comm.). Loose manure and bedding was cleared out in the spring leaving a crust on the floor (Reynolds pers. comm.). It was this crusted floor layer that was investigated in 1990.

Plate 1: Archive (14th February 1940) field photograph of cave sediments at Arene Candide, Liguria, Italy, excavated by Bernabo’ Brea. On the excavated surface are features of Middle Neolithic date (layer 22). The well stratified black and white deposits above, which are truncated halfway up the vertically placed spade, are of Late Neolithic (Chassey) date, whereas the deposits above span the Copper Age to the Byzantine period. The well stratified Late Neolithic sediments are typical extensive deposits which have resulted from seasonal (overwintering?) stabling of herbivores (cattle, sheep and g...