The conclusions I have drawn previously from this evidence have tended to emphasise the way in which the puri has utilised this niskala landscape as a symbolic resource in a contemporary political economy dominated by tourism (MacRae 1997, 1998, 1999), namely by reconstructing, inventing (Hobsbawm and Ranger 1983) or imagining (Anderson 1991) a neo-negara, a new kingdom constructed in symbolic terms, through material practices of temple construction and ritual sponsorship. Seen through the lens of comparative Austronesian ethnography, however, it is instructive to consider the ways and extent to which this is built on foundations of appropriated local and inter-local organisation which predate and appear to exist independently of the puri and resonate with ideas more common to the Austronesian than the Indic world. ^[32]

Firstly, the idea that the landscape has, as well as its physical aspect, a subtle, inner magico-ritual aspect is common in India as well as throughout the Austronesian world. Some aspects of this landscape, however, take forms either more Indic or more Austronesian. Secondly, while sacred mountains are (literally) central to Indic sacred landscape, the primary level of spatial organisation in Bali—the linear uphill-downhill axis—corresponds directly to a well-documented Austronesian pattern as do the practices of orienting buildings and their component elements to this axis (Fox 1993: 14-15). The centric, mandala-type desa forms on the other hand, are unknown elsewhere in the Austronesian world and are in fact found in villages clearly dominated by puri of Hindu-Javanese descent. Thirdly, despite the absence of clear trans-desa levels of organisation such as the banua of the mountains, there is considerable evidence of an embryonic, residual or perhaps simply different form of inter-linkage between desa, articulated especially through temples. Fourthly, the programs of negara-building carried out by puri, while referring to aspects of Indic kingship, are based on mobilising relationships with and ideas about land, which are congruent with Austronesian ones in general and Bali Aga ones in particular.