CHAPTER TWO
Approaching Roman Religion:
The Case for
Wissenschaftsgeschichte
C. Robert Phillips, III
A comprehensive history of the study of Roman religion does not exist, and this
despite the upsurge of contemporary interest in Wissenschaftsgeschichtefor other
areas of classical studies. It is true that there have been overviews of the
Wissenschaftsgeschichte of Roman religion (Latte 1960: 9 –17; Wissowa 1912:
10 –17), surveys of recent scholarship (e.g. Rose 1960) and the occasional discussion
of interpretive developments (Michels 1954/5; Wide 1912: 268–71), and studies
devoted either wholly or in part to re-evaluating the work of past scholars (Bendlin
2000; Scheid 1987). Then there are varia, a plethora of widely scattered observa-
tions, such as Latte’s view of animism (Weinstock 1961: 206) or Wissowa’s Janus-
like dismissal and utilization of Frazer (Wissowa 1912: 248 n. 3, 104 n. 3). Finally
there are archival materials. Some have been published, such as the correspondence
between Usener and Wilamowitz (Dietrich and Hiller 1994). Many have not seen
the light of day, such as letters from, to, and about Stefan Weinstock (Phillips 2004);
there we learn, for example, that Eduard Fraenkel’s negative report on Weinstock’s
“Triumphus” manuscript led to its rejection by Oxford University Press (Phillips 2004:
1043 – 4). The reasons, definitely not bruited about at the time, remain unknown
and we are forced to conclude from Fraenkel’s published works that while the great
Latinist knew much about Roman religion, he did not know as much about it as he
thought. But totum pro partibus, none of these items provides either a synoptic whole
or, per se, the materials for constructing one. Writ large, there exists no history of
the study of Roman religion which attends to its socio-intellectual context, nothing
along the lines of what has been done, say, for Roman history (Christ 1982).
Consequently, I begin with the axiomatic as I briefly identify reasons why this is so;
the elaboration of those reasons and their significance will ensue along with cross-
references to these axioms.