Our research is carried out within the field of Cognitive Classical Linguistics, recently established by the seminal book Embodiment in Latin Semantics (Short 2016) and inspired by applications of the Embodiment Theory in cognitive semantics (Gibbs 2005). More precisely, the approach is based on the methods and theoretical underpinnings of the Image Schema Theory, which maintains that language structuring largely depends on humanly embodied imaginative mechanisms, frequently activated by metaphorical extensions.
The challenge here is to see if principles and methodologies drawn from cognitive semantics can be fruitfully applied to an ancient language, and what they can  tell us  about the Romans’ way of conceptualizing their relation to the world. Moreover, the Latin data can tell us something about the diachronic stability of conceptual metaphors. 

What are our goals?
We explore a wide range of emotion metaphors, including fear, rage, hate, sadness, happiness, love, and jealousy metaphor, to reassess their status and implications for a cognitive-based approach to the semantics of Latin. The analysis is corpus-based and drawn on the Bibliotheca Teubneriana Latina corpus, which comprises over 1,140 works by nearly 600 authors spread over eight centuries. 

Why the semantics of feelings and emotions?
Emotions are abstract experiences par excellence, and are thus particularly prone to metaphorical conceptualization, also in terms of spatial configurations. Indeed, many studies have highlighted recurrent embodied patterns that build bridges between spatial orientations and feelings (embodiment of experience: Tyler & Evans 2003). For example, the human body is often vertically conceptualized along functionally asymmetrical opposite poles as an up/down antithesis, which offers the basis for portraying many experiential metaphors in terms of happy/positive is up, sad/negative is down, since standing tall is typical of feeling good, whereas lying down is associated with sleep (i.e., unconsciousness), illness, and death, thus being integral to negative emotions. 

What is embodiment?
Embodiment is a cognitivist theory that maintains that our basic cognitive functions are shaped by our bodily experience and its interaction with the environment, which work as a universal structural template to interpret and describe many aspects of reality. In the words of one of the theory’s father, “the core of our conceptual systems is directly ground in perception, body movement, and experience of a physical and social nature” (Lakoff 1987: xiv). In recent years, embodiment theory has become a research topic in Cognitive Linguistics and Lexical Semantics, but its impact has been explored mostly in relation to modern languages. We are therefore curious about the multiple ways in which embodiment affected the Romans’ experience and how this potentially universal cognitive key interacted with the Roman culture-specific component.

Research directions

Specifically, this project aims to explore the embodied basis of Latin experience from different perspectives: 

• synchronic, to provide functional accounts of embodied metaphors in selected authors, and also to trace interactions between literary genre, topic, intended audience, and style (e.g. Plautus’ comedies, Cicero’s speeches, Seneca’s letters); 
• diachronic, to explore the emergence and development of selected metaphors up to the 4th c. AD; 
• cross-linguistic, to corroborate evidence for conclusions drawn from a typological perspective; 
• quantitative, to offer corpus-based evidence for frequency and productivity effects; 
• benefiting from insights of Digital Humanities research, to participate to the annotation of the Latin WordNet 2.0 and the Lexicon Translaticium Latinum. 

Methods

The analysis is corpus-based and drawn on the Antiquitas and Aetas Patrum sections of the Bibliotheca Teubneriana Latina (5.7 million words). Some advantages of a corpus-driven approach to metaphors are to deal with a broader range of data, to explore their diachronic development and to avoid generalizations based on personal introspection (Deignan 2005).  To identify experiential metaphors in this corpus we adopt a target-oriented methodology, which allows the exhaustive description of the metaphorical mappings associated with particular target domain items in a data source (Stefanowitsch 2006). We look up reference dictionaries for lexical items expressing experiential states (e.g. ira, timor) and check if they are associated with metaphors in the corpus (e.g. in iram cadere, lit. ‘to fall in anger’; perfusus esse timore, lit. ‘to be poured over with fear’). We then classify them under general image schemas organized at different levels of semantic granularity and semantic complexity (e.g. ‘to become angry is to fall in a container’; ‘fear is a liquid substance’ and ‘to experience fear is to be poured over with a liquid’). Furthermore, the results of our analysis of Latin metaphors will enrich the Latin WordNet 2.0 and will be available to the scientific community.