The Harappan settlement pattern may offer some clues to the organization of society. The settlements fall basically into three categories. First there are a very few enormous settlements, the cities. These were at least 60 hectares in extent (Dholavira, which may actually be as much as 100 hectares) and could be as large as 250 hectares (Mohenjo-daro). Four (Dholavira, Mohenjo-daro, Harappa, and Rakhigarhi) have been investigated at least to some extent; Ganweriwala and perhaps others have been identified but require investigation to confirm their extent and nature. These cities were spaced fairly evenly throughout the Harappan realms and were separated from each other by somewhere between 280 and 600 kilometers. It is interesting to compare this arrangement with that in Sumer where the cities, which housed around 80 percent of the population, were far more numerous and were situated only about 20-60 kilometers apart, but distributed lineally since they were confined to the narrow floodplains of branches of the Euphrates. Harappan cities each served a huge domain of between 100,000 and 170,000 square kilometers, providing the full range of urban functions, including administration, social leadership, and specialist industries. Presumably, the cities also provided a religious focus, although it is only in the case of Mohenjo-daro that this is clear.
At the opposite end of the spectrum were rural settlements: farming villages, pastoralist and hunter-gatherer camps, and fishing villages. These could range in size from less than a hectare to 7 or 8 hectares. Their inhabitants were essentially primary producers of subsistence products who would also have undertaken domestic crafts such as weaving and woodworking. In addition, there were villages whose inhabitants were specialists, such as the shellfishers and shellworkers of Nageshwar. These settlements had none of the complexity of towns but were probably not self-sufficient in subsistence products like villages. They may have been occupied only seasonally by people who spent the rest of the year in primary production.
The third category, towns, is an amorphous catch-all, including a great diversity of different types of settlement. They were usually quite small, around 1 to 5 hectares, and, though some were larger, few exceeded 16 hectares. This figure does not include the suburbs that may have existed outside many towns; traces of suburban settlement have been reported outside a few and are suspected at others, but none has been properly investigated.
While towns resembled cities in that they housed officials, traders, and other occupational specialists and probably provided services for the people of their area, the majority were also specialist centers. Some, such as Balakot, were centers of craft production; others were ports, like Sutkagen-dor, or trading centers, like Shortugai. However, they shared certain features: They were all, to a greater or less extent, involved in craft production, usually of a specialist nature, making goods from local materials or materials brought there from a connected region, for distribution within the entire state and sometimes also for export; they all played some role in the internal distribution network, including storing goods and materials for onward transmission; and those on the periphery were located to fulfill a role also in the procurement of goods and materials from external regions. They were all small enough to be supported by the produce of their hinterland and probably all included primary producers, such as farmers and fishers, among their inhabitants, particularly in the suburbs from which these people could have easy access to their fields or to the rivers or coasts. The same is probably true of cities.
The settlement pattern seems to indicate that both towns and villages were tributary directly to their local domain capital (city), links between them being maintained by pastoralists and via water transport; towns also acted as funnels through which local goods were channeled toward the city.
The impression given by this settlement pattern is very different from the village-town-city hierarchy familiar from many cultures, in which villages looked to their nearest town for manufactured goods and services and local administration and in turn provided the town with subsistence products, while the city provided overall administration, specialist services, and specialist industries, whose products reached only the elite of the towns in its territory, from which it received subsistence produce. In states the apex of this food chain pyramid of replicated units was a capital to which all the rest of the settlements were politically, socially, and economically subordinate.
The Harappan state, in contrast, was more akin to a body: a single functioning entity composed of a number of different and complementary units, each with its own particular function within the whole. The head, limbs, organs, and other parts of the body equate to farming villages, pastoral camps, trade centers, hunter-gatherer camps, fishing villages, procurement centers, processing centers, and other types of Harappan settlement. As parts of the body are made up of cells, sharing the same basic biological elements but often specialized in function, so the Harappan settlements were made up of people who were occupational specialists, including officials and priests, as well as farmers, pastoralists, fishers, artisans, and traders, their composition depending on the function of the settlement. To continue the analogy, the internal distribution network was the circulation system that ensured that all the things necessary for life reached all parts of the body (state). Whether this had one head (Mohenjo-daro or Harappa), two, five or more is not yet apparent, but clearly this state was unusual for its time in how it operated.