Pliny in Space (and Time)

In their original call for papers, the organizers of this volume rightly lamented the fact that work on Pliny has tended to fall into one of two traps. "We have [either] plundered him for snippets on" various aspects of Roman civilization or "confined ourselves to analysis of individual letters." In a sense, I will commit both sins in this paper, so let me offer some pre-emptive justification. My main interest in Pliny here will indeed be to exploit him in service of something else—in this case the history of geography. I am not, however, looking for pre-existing facts for which Pliny's record is the incidental conduit. Rather, his record will itself be the salient set of facts. As for the second problem, I will primarily be reading two letters, and a hackneyed pair at that. On this point I can only ask the reader to suspend judgment on the novelty and broader significance of my reading until the end.

The two letters I have in mind are the famous descriptions of his Laurentine (2.17) and Tuscan (5.6) villas. Numerous attempts have been made to reconstruct the respective floor plans from the texts. ¹ Most have been plausible, but none demonstrably correct. The difficulty, I suggest, is that Pliny is not trying to do what scholars seem to have assumed he is attempting. In fact, his project is incommensurable with theirs in at least two important respects. In the first (and longer) section of this paper, I treat Pliny's treatment of space in the villa letters and its inextricable connection with his treatment of time. In the second section, I offer a possible extension of this analysis to an urban context, though the relative lack of evidence makes this a more speculative project. Before I get to this, however, it may be helpful if I say a few words in general about how dwellings (and spaces more generally) may be described.

Studies of this topic have often made a fundamental distinction between the route or itinerary, on the one hand, and the map on the other. Even if neither exists physically, many spatial descriptions seem to presuppose one or the other way of organizing information. There is a considerable literature in cognitive psychology on the choice of modes and the use of, for instance, landmarks and orientations within each mode. ² One version of the map/itinerary distinction has been applied to the classical world in general by Pietro Janni's 1984 monograph on The Map and the Periplous: Ancient Cartography and Hodological Space. In this work, he sets Roman spatial practice in a broader historical context; in particular, he takes a developmental view. Map-oriented thinking represents, he claims, a more advanced stage of development than route-oriented thinking. ³ Development from the itinerary to the map did not start until the late Middle Ages or even the early modern period and was not really completed until much later than that. The Romans, on this view, could have had no true maps, or even think in such terms. Maps are perhaps a better way of doing the same thing as itineraries, but they have similar aims. At the most, there is a difference in point of view. One either follows a path on the ground or looks at a map from above. A few classicists have cited Janni's work, most approvingly, but none have built on it or even given his theses extended scrutiny. ⁴

A somewhat different spin is put on the same distinction by Michel de Certeau in a section of his book The Practice of Everyday Life (1984). Generalizing on the basis of a more technical article by Linde and Labov (1975), he analyzes what they call "routes" and "maps." De Certeau, like Janni, sees a historical development and agrees that mapping in the technical sense in question does not really begin until the fifteenth century and is not fully developed until much later. For him...