This is a revised version of a DPhil thesis prepared at the University of Oxford under the supervision of Professor Andrew Wilson. The book's main aim is to offer the first synthesis of Roman and late antique rural architecture and settlement in ancient Tripolitania (modern Libya), an area still ravaged by civil war. It forms part of the EAMENA (Endangered Archaeology in the Middle East and North Africa) project, a joint research initiative devoted to applying several remote sensing technologies to record available data from archaeological sites and landscapes which are under threat.

Despite the publication of some previous monographs on the archaeology of the Roman period in this region of North Africa — the best known being D. J. Mattingly's Tripolitania (1995) — our current knowledge about rural landscapes and living conditions in most parts of Roman Tripolitania is still scanty. In an attempt to fill these gaps, Nichole Sheldrick's book is based on a compilation of data about some 2,400 rural settlements, drawn from legacy data from several field survey projects including ULVS (UNESCO Libyan Valleys Surveys) or the Prospection des vallées du Nord de la Lybie project, as well as data obtained through her own remote survey performed by the use of Google Earth and Bing Maps satellite images. The presentation of all this up-to-date information represents one of the main achievements of this book. However, the analyses to be found in the book are greatly limited, as S. honestly admits at several points, by the sampling problems generated by the diversity in the provenance of the data and by the fact that it proved impossible for the author to carry out fieldwork. Perhaps the most important of these sampling problems is the impossibility of determining specific occupation chronologies for a large majority of the recorded settlements. This is not only due to the lack of stratigraphic excavations but also to the different methods used for documenting surface finds and the different criteria used to assign chronological ranges for the sites recorded in previous survey projects.

However, S.'s efforts to integrate all this information within a single GIS have great merit, since many of the sites discovered in previous survey projects were published at a time before GPS devices became widespread. This has enabled her to generate a remarkable collection of maps with which to illustrate large parts of the book.

The unsystematic character of the data leaves the analysis of the architecture of the rural settlements as the central focus of the book. This is reflected in the structure of the study's three main chapters. The first (ch. 4) focuses on military architecture, synthesising a wealth of information — especially from work by French and Italian colleagues — which, despite offering very little new information, serves as an excellent state-of-the-art discussion that will be especially valuable to an English-speaking audience.

Ch. 5 is devoted to different types of unfortified rural settlements. One of the main strengths of the book relates to the typological proposal established by S. In contrast to the traditional application of the term villa, an archaeological concept that is problematic because of its architectural imprecision and the economic connotations with which it is normally associated, the book applies a classification based on three categories: 1) single farmyard buildings; 2) courtyard buildings, distinguished by the presence of continuous ranges of rooms on two or more sides defining a courtyard; and 3) open complexes, formed by compounds of several yards grouped irregularly. This classification allows S. to articulate her discussion not only in terms of a strict building typology but also in economic terms related to the agrarian function of the different facilities associated with each category. While courtyard buildings are associated with evidence for luxury decorations and surplus-processing features such as olive-oil or wine presses, farmyards seem to be associated with a mixed agrarian model characterised by a small-scale capacity for agricultural production and the management of just a small number of domestic animals, probably performed by single peasant household units.

Ch. 6 focuses on fortified buildings. Based on the morphological criteria applied by S., all the sites are classified according to three main categories: 1) tower-like buildings; 2) fortified courtyard compounds; and 3) doubled or irregular fortified compounds. Besides discussion of the distribution of luxury equipment and surplus-processing features in these settlements, the main emphasis of the chapter is on the explanation of the several types of defensive fortifications of these. Even when the low resolution of Google satellite images makes it difficult to classify certain sites discovered by S. in her remote sensing survey, the detailed study of the building techniques recorded in these complexes is made possible thanks to the use of legacy data, especially the photographic files from the ULVS archive. The systematisation and analysis of these architectural elements can be considered one of the main achievements of the book.

More problematic is the discussion of the structure of rural settlement, the second main concern of the book. A particularly problematic point is the segregation of discussion of settlement patterns between the different architectural categories covered in each chapter. This segregation is partly justified by the belief that most of the unfortified buildings correspond to the Early Imperial period, while most of the fortified buildings can be dated to Late Antiquity. Given the impossibility of establishing reliable chronologies for most of the settlements, however, it would have been useful to have a comprehensive analysis of all the settlement types discussed in the book.

Despite these problems, it would be unfair not to acknowledge this work's notable contribution to our understanding of the rural world in this region of the Roman empire. Given the challenges she faced, the fact that S. decided to focus her research effort on the study of the rural settlements of Tripolitania, most of them humble and with obvious threats to their preservation, is a source of inspiration for all of us who believe in the imperative need to increase our knowledge of ancient peasant communities. Books like this one remind us that it is better to hobble along main roads than to wander through back streets.