ABSTRACT
A major strength of the moat hypothesis is that similar features are found at major settlements incorporated into Tiwanaku’s sphere of control and influence. Kolata (1993a; Kolata and Ponce 1992) points out that a moat physically similar to that at Tiwanaku surrounded the monumental cores of several other sites, most notably Khonkho Wankane and Lukurmata. At these sites as at Tiwanaku, the channels were natural features augmented by human activity. At Lukurmata, a swampy channel forms an arc around the Wila Kollu ridge that connects with the inundated floodplain to the north (Figure 3.3b). Raised fields line the edges of the channel. Perhaps even more feasibly than at Tiwanaku, the channel, by essentially bringing the lake into Lukurmata, may have facilitated travel into the site via balsa rafts and boats, serving trade as well as monumental construction on the platform it surrounds. Like the smaller moat around the Mollo Kontu platform in Tiwanaku, a small moat surrounds a similar platform at the site of Chojasivi, a few kilometers east of Lukurmata. Thus, whatever one argues regarding the specific function and meaning of the channels, they clearly were a repeating pattern in Tiwanaku sites. This suggests that, as Kolata argues, they formed an important element of Tiwanaku’s emerging cosmology. They isolated significant ceremonial constructions, including iconic mountains, as physically bounded, ontologically distinct sacred islands.
