What patronage actually paid for
It is easy to read a list of temples funded by a ruler as a list of buildings. That understates what was involved by a considerable margin.
A temple foundation of the kind Ahilyabai Holkar settled repeatedly supported: the fabric of the building and its future repair; the priests who conducted its services; musicians, cooks, cleaners and administrators; frequently a kitchen feeding travellers and the poor; often a rest-house; and sometimes a school. All of this was funded by a permanent charge on land or revenue, which meant it continued indefinitely without further central expenditure.
In institutional terms, it is closer to founding and endowing a college than to commissioning a monument.
The regional traditions
India had no single temple architecture in the eighteenth century, and a patron building across several regions was necessarily building in several idioms.
The northern form
Temples of the northern tradition are organised around a curvilinear spire — the shikhara — rising above the sanctuary, capped by a ribbed disc and a finial, usually preceded by one or more pillared halls. This is the form of most of the temples associated with her patronage on the Gangetic plain and in central India.
The southern form
The southern tradition places its vertical emphasis differently: the sanctuary tower is stepped rather than curvilinear, and the dominant structures are frequently the gateway towers of the enclosure rather than the shrine itself. Works at sites such as Rameswaram and Kanchipuram sit within this tradition.
The riverfront ghat
The ghat belongs to neither and to both. It is not a temple type at all but a piece of civil engineering — a broad flight of stone steps built to hold against seasonal flooding and to remain usable through a large range of water levels. It is the form most characteristic of her programme.
Why hers survived
Three reasons, in descending order of importance.
The endowments
A building with a permanent maintenance fund attached outlives a building without one, and the difference compounds over centuries. This is the single largest factor.
Local idiom, local craftsmen
Her works were executed broadly in the building traditions of the regions where they stood, using local materials and local labour. Buildings that belong to their setting are maintained by the communities around them. Buildings imposed on a setting are not.
Continuous use
Almost everything she funded was built to be used — by pilgrims, travellers and residents. Structures in continuous use get repaired. Structures that become ceremonial do not.
The cost of survival
There is a complication worth stating. The same continuous use that preserved these buildings has also altered them. Temples in active worship are repaired, extended, refaced and rebuilt as need and funds allow, and the eighteenth-century fabric within a living temple can be difficult to isolate.
This is why attributions vary. Some works — the 1780 rebuilding of Kashi Vishwanath — are firmly documented. Others rest on local tradition, on later record, or on an inscription of uncertain date. A visitor standing in front of a structure associated with her name may be looking at her work, at a nineteenth-century reconstruction of it, or at both.
None of which diminishes the programme. It simply means that its scale is better measured by its geography and its logic than by any confident count of buildings.
A note on sources. This article summarises the broadly agreed account. Where sources disagree on detail, the text says so rather than choosing between them. Figures that vary widely between sources are not reproduced.