Carolyn MacDonald
Recent decades have witnessed a renaissance in scholarship on Marcus Terentius Varro, characterized by a renewed interest in the political dimensions of the antiquarian’s treatises. This is especially true of the De lingua Latina, a text that is inextricably bound up with the late Republican struggle to redefine Romanness (Bloomer 1997, Moatti 1997, Dench 2005, Wallace-Hadrill 2008, Zehnacker 2008, Spencer 2011 and 2015). This paper investigates how precisely Varro makes a linguistic treatise into an apt vehicle for political polemic by examining the text’s linguistic metaphors — both the metaphors Varro uses to characterize the workings of language and those he deploys to describe his own activity as a linguist. My argument, in brief, is that Varro’s metaphors work together to present etymology as a form of archaeology, which alone can unearth the vanished Roman past.
I begin by surveying the linguistic metaphors of the De lingua Latina, which demand attention all the more in contrast with the dry, technical prose that predominates in the work. These metaphors by no means all cohere in a single system of related images. However, those which appear most frequently — the word as a tree with its roots driving into multiple language fields (5.13, 5.74, 5.93, 5.123, 7.4, 7.28-9); and etymological scholarship as a form of digging or uncovering (5.93, 6.2, 6.37 7.1, 7.2) — bear an important relationship to one another. Together, these metaphors characterize the derivation of words and their development in relationship to one another as subterranean processes, operating below the readily accessible surfaces of language.
Once this is established, the bulk of the paper is devoted to teasing out the political significance of this characterization of language. Firstly, and most obviously, Varro’s metaphors of language developing underground problematize any neat demarcation of the boundaries between Latin, Greek, and Sabellic languages like Sabine (Briquel 2001). Indeed, Varro’s tree metaphor may even suggest that he intuited the cognate relationship between certain words in these languages (Ferriss-Hill 2014). This has clear implications for Varro’s understanding of Roman identity — like the Latin language, it too turns out be rooted in various foreign soils.
Secondly, and perhaps more importantly, there is a striking correspondence between Varro’s images of underground language and his representation of Rome in De Ling. 5.41-54 and 5.146-64. Just as the roots of contemporary words are metaphorically buried beneath the visible landscape, so too the events of the Roman past are occluded by the late Republican cityscape. In both cases, only the etymologist can reveal what lies hidden beneath the surface. By examining Varro’s explanations of the toponyms Capitolinus, Lacus Curtius, and a cluster of terms associated with the Comitium, I demonstrate how the De lingua Latina actively models the power of language to retain and even reshape topography. In Varro’s hands, etymological research undoes the changes wrought on buildings by time and Caesars, restoring forgotten names and rebuilding demolished structures. What is more, the turn to language and etymological excavation maps onto Rome a different kind of history — or better histories, for what Varro digs up is a plural past. In the De lingua Latina, a single word like ‘Curtius’ can yield multiple explanations and dredge up multiple memories of the city’s former lives. Roman toponyms also act as the vestiges of a plurality of peoples, who do not coalesce into a single narrative of the ‘Roman’. Varro’s linguistic metaphors thus operate in tandem with his tendentious articulation of the cityscape, to undermine the more unitary visions of Roman history and identity promoted by his contemporaries, and to uncover a past that is fragmented and plural.