This substantial volume deals with a corpus of over 150000 Roman pottery sherds from excavations in Lincoln, England, between 1972 and 1987. Given the quantity of pottery involved, this publication presents an exciting opportunity to list and analyse a valuable body of material from an important regional centre in Roman Britain. Both the dataset and its discussion allow for the spatial distinction between the original upper city, comprising the fortress and succeeding upper colonia, the lower city, defined by the lower colonia, and extra-mural settlement at Wigford. The data, discussed in seven core chapters dealing with different classes of fine- and coarse-ware fabrics and an extensive synthetic chapter, are generally biased towards the mid–late Roman period, with a relative paucity of material from the formative years of the fortress and the early phases of the colonia. In terms of pottery supply, Lincoln is not an exceptional site for Roman Britain, although this does not diminish the richly varied narrative of consumption that is presented here. This point is encapsulated by an impressive map of pottery sources from the wider Roman world illustrated with the various shapes and styles of vessels imported (fig. 247, p. 313). One interesting discussion among many concerns the evidence for substantial local supplies of Iron Age pottery to the fortress, which is unusual for legionary fortresses in Britain. This finding is mirrored at the site of Longthorpe to the south, also previously used by Legio IX Hispana, highlighting the capacity of Roman military units to engage in distinctive cultural practices, diverging from commonly held notions of a homogeneous military culture across the north-western provinces.

The value of any corpus depends on a number of criteria. First and foremost is the need to produce a database of finds that is as complete and transparent as possible, so that researchers may use the data to answer their own questions and make comparisons with equivalent sites. A more secondary concern is to conduct a series of analyses to aid in the interpretation of the material in question, thereby enhancing the state of knowledge of the site (in the case of Lincoln, published as preceding volumes in the same monograph series). Unfortunately—and not uncommon for Roman pottery reports and corpora from Britain—this volume only partially meets the primary expectations of an archaeological corpus. In particular, it falls short on two aspects: the quantification and the facility to reconstruct the assemblage composition of individual archaeological features and contexts.

The rationale for quantification is explained in a short methodology chapter that follows the introduction. Here, the case is made for quantification by sherd count, with mention of “experimental weighing” for some assemblages (p. 6) and a few larger, later assemblages quantified by estimated vessel equivalent (EVE) (p. 7). Such reliance on sherd count alone greatly reduces the research value of the corpus. The measure of sherd count is significantly affected by fragmentation, and thus cannot be used to undertake reliable statistical comparisons—either with other Roman sites or with different contexts from the same site. While financial constraints are cited to account for the absence of widespread quantification by weight and EVE (which are far more useful for the researcher), it should be noted that it has been recently observed that quantification by a range of methods including EVE does not create significant extra work for pottery specialists. The small burden of this task should be balanced against the much more costly exercise of re-studying the same assemblages at a later date, as is often required to incorporate such inadequately quantified sites into larger synthetic research projects (Doherty Reference Doherty, Perring and Pitts2013: 94).

In addition to the limitations of the methods of quantification used, the presentation of data in the appendices to this volume falls short of a full and transparent corpus. This is disappointing given the ease and low cost of storing these types of data on an accompanying CD or using web-based repositories. A series of large tables does provide a useful summary—with major pottery groups quantified according to fabric and form and more detailed categorisation of specialist wares, notably samian and Mancetter-Hartshill mortaria. Unfortunately, this is still a form of data reductionism since it is not possible, for example, to break down categories such as ‘dish’, ‘flagon’ and ‘beaker’ into quantities according to their constituent vessel form types, potentially obscuring relevant social information. This is a shame, as it limits the possibilities for researchers to incorporate the material in this corpus into bigger projects for comparative study.

In summary, this volume succeeds in summarising a large quantity of ceramic data from an important Roman centre in northern England, with a particularly insightful synthetic discussion that addresses spatial and chronological variations in pottery use at the site. As with a great many Roman pottery reports and corpora from Britain, however, there are important omissions that could have been rectified comparatively easily and cheaply prior to publication in order to make the data more amenable for future analyses. This is too big an issue to lay solely at the doors of the authors of this report, but the present situation in which detailed data are inconsistently recorded or excluded from publication is simply not satisfactory, and damages the core archaeological principle of preservation by record.