There is a continuing policy debate about whether it is possible to have sustainable small settlements in outback regions of Australia, where there is low and variable primary production and a sparse and mobile population. This debate is focused largely on Aboriginal settlements, but applies equally to all desert dwellers. In this contribution, we review the sources of economic flows through settlements occupied by different communities with common livelihood sources, whether based on mining, grazing, tourism, cultural resources, welfare or services, concluding that most desert livelihoods depend directly or indirectly on temporally variable inputs. Individual remote settlements tend to be dominated by one such ‘community of livelihood’, and differ in nature according to the source of that livelihood. These create types of settlement and service aspirations which are alien to more densely settled regions.
Settlement ‘viability’ is a measure of the short-term balance between aspirations for services (technical, social, but also for livelihoods and well being) and the costs of fulfilling these aspirations, and consequently is not a simple on/off switch – the community can adjust both its aspirations and the cost factors involved in meeting them. We define a resilient settlement as one that is viable in the long term in the face of its variable inputs. Thus, we determine that the concepts of settlement viability and resiliencemust be analysed differently according to the strategy adopted by different resident communities. In particular, Aboriginal (and pastoral) communities are particularly dependent on social and natural capital, yet these are not monitored. The whole analysis emphasises the importance of taking a demand- rather than supply-driven approach to services in desert settlements. Our conclusion is that, if top-down solutions continue to be imposed without appreciating the fundamental drivers of settlement in desert regions, then those solutions will continue to be partial, and ineffective in the long term.