Roberto Ragno

Statistical approaches to the study of archaeological surface survey datasets are problematic, as they are often characterized by excessive zero counts and overdispersion. This paper introduces a Bayesian Zero-Inflated Negative Binomial (ZINB) model to classify survey grid units into user-defined functional areas based on artifact distributions. This approach was applied to the Roman maritime villa of Mascherone, located near the city of Siponto (northern Apulia, Italy), surveyed using a total sampling strategy. After filtering, 52 squares (20 × 20 m) containing 21 distinct artifact types were analyzed to identify three hypothesized functional areas: residential, storage, and craft. The model explicitly accounts for structural and sampling-derived zeros in the dataset while also handling overdispersion. Furthermore, it provides probabilistic classifications with quantified uncertainty for each square unit. Results indicate a residential core consistent with legacy aerial evidence, while storage and craft zones remain less certain due to limited indicators. This approach effectively addresses zero-inflation in survey datasets and offers a scalable framework for broader archaeological and landscape analyses.

This dissertation explores the subsistence methods, economic systems and environmental adaptations of Italian peasant communities in the 1st millennium CE, with a particular focus on the transitional period from the Roman Empire to the early medieval era. Existing work on agricultural production in this period has been based on literary sources and field surveys, or has focused on individual sites or regional collec-tions, while a multi-source archaeological study is absent from the discourse. This work addresses this gap by using environmental proxies to reconstruct the historical agricultural landscape through plant and animal occurrence patterns in legacy data. To this end, 190 botanical and 466 faunal assemblages from 309 sites are quantitatively analysed within a Bayesian framework, revealing a strong trend towards re-gionalisation in agricultural strategies during the early medieval period. In addition, these findings expose variations in agricultural techniques and dietary patterns across Roman settlements, shedding light on the extent to which Roman agricultural and economic frameworks persisted or changed during the early medieval transition, and the adaptive agricultural strategies adopted by farmers. The quantitative analyt-ical findings are also contextualised alongside wider historical sources, archaeological evidence, and cur-rent debate, allowing for a bottom-up understanding of the agricultural regimes in question. This work represents the first attempt to use temporally and geographically diverse bioarchaeological data to visualise the Italian agricultural landscape across the longue durée.

This dataset contains raw counts of archaeobotanical (macro-)remains from archaeological sites located in mainland Italy, dating from the 1st century BCE to the 11th century CE. The 195 carpological assemblages have been collected from a variety of published sources and reports and have been stored in a database with the relevant contextual and geographical metadata. The collection focuses on 40 plant taxa that were most commonly represented across the reviewed archaeobotanical reports, prioritising species with clear associations with agricultural activities and subsistence practices. This data may be used for statistical assessment of spatial and temporal differences in dietary patterns, agrarian strategies and crop distribution in the Italian peninsula between the Roman imperial period and the early medieval age.

The 1st millennium CE represents a period of significant change in the agricultural landscape of southern Italy. Sheep and goats are among the most common faunal remains recovered from archaeological excavations of this period, but the contribution of these animals to the agricultural economy (particularly wool production) is often discussed through textual sources. This paper synthesises caprine taxonomic abundance trends using a Bayesian multilevel modelling approach that employs a beta-binomial distribution to address the problems of overdispersion and unequal assemblage/group sizes. Our models contribute directly to the problem by suggesting a period of change in livestock management practices around the 4th and 6th centuries CE, when the region’s shift to cereal farming appears to be accompanied by an increase in sheep and goat numbers.

This study investigates the cereal farming practices of the Italian peninsula during the first millennium CE, with a particular focus on the Early Middle Ages. Using non-parametric multivariate statistics and a dimensionality reduction algorithm (PERMANOVA; nMDS), this research presents and compares 177 archaeobotanical caryopses assemblages from three areas of Italy. The results showed that differences in cereal farming practices between Northern and Southern Italy were not statistically significant during the Roman period, but became significant during the early medieval period. The research suggests that after the collapse of the Roman Empire, northern peasants had more autonomy in selecting their crops, while southern farmers were more resistant to change. These findings challenge the prevailing assumption that early medieval peasants across the entire peninsula uniformly adopted cereal crops like millets and rye. Overall, the study sheds new light on the diversity of cereal farming practices in early medieval Italy and presents a comprehensive collection of cereal macroremains from mainland Italy.