Water in Australia is largely considered in the separate domains of urban and rural. This split exists in the narrative of settlement, political discussions, the jealousies of pub talk, supply and management systems, and policy and institutional regimes. Rather than urban–rural differences when referring to households and water, the real split is reticulated or independent supply. The split is evident too in research. Relative to population and economic activity, far more resources are expended in rural water and related research in natural resource management (NRM) than in the urban domain, despite the fact that major urban centres contain some 85 per cent of the population, and the bulk of social and economic activity. Yet on spatial extent, ecological impact and gross share of consumptive and non-consumptive water use, the rural domain deserves attention. In national policy debates, in the NWI and the National Plan for Water Security, the greatest focus is rural — or, rather, the Murray-Darling Basin. Some researchers advocate the importance of one over the other and therefore, like the policy regimes, can ignore interactions. It would be better for researchers, policy-makers and the public to believe two things at once. Both rural and urban water (and extant, missing or proposed links between them) are important and deserving of close attention.

Nevertheless, the NWI does instruct a linking of rural and urban water management (regarding the NWI, see Hussey and Dovers 2007); and, increasingly, attention is being paid to water management in hitherto overlooked peri-urban areas. Mostly, the linkage is interpreted in terms of transfer of water, whether through trading or simple capture, from rural sources to thirsty cities, and the provisions have yet to be pursued with any vigour and consistency. But this is a future area of research and policy activity with strong behavioural dimensions that are merely hinted at below.

While the focus here is urban, we do need to maintain an integrated focus, or at least recognise the other domain as a reference point. It may be that, behaviourally and in policy terms, it is in smaller rural settlements, localities, farm households and the non-metropolitan Indigenous domain where the sharpest insights into the human dimensions of abstemiousness are to be found. In southern NSW, there are small settlements that have been beyond Stage 5 water restrictions for more than five years; in rural homesteads in drought-struck areas children are bathed in suspect water hauled manually from diminishing farm dams while generational legacies of homestead gardens are irrevocably dead; small-town tourist ventures have suffered massive turnover losses and entire communities have lost the facilities to play sport; in Indigenous settlements, water supply and quality are of third-world standard. It is not a discussion of how long one should stay in the shower, as turning on the tap is an empty gesture. Such situations are far beyond the experience and imagination of the vast bulk of urban Australians and bear reflection.

Later a simple characterisation of phases in urban water management will be presented, and it is foreshadowed that the earlier phases have not been replaced but still struggle for space amid multiple values surrounding water and the inertia and path dependency of agencies and institutions. The households in drought-struck grazing districts, technological changes aside, are close in their use of water to predecessors. Joe Powell’s (2000) symbols of two fundamental black and white Australian water dreamings still apply — the Rainbow Serpent of creation and the Water Cannon of intervention. The formal recognition of Indigenous water values in the NWI in 2004 — a great advance even if yet to be addressed seriously — shows that the oldest Australian institutions regarding water, Indigenous law and story, have not gone away (Jackson and Morrison 2007). These two symbols are non-urban; later we will consider some purely urban icons of water use.