Amazon.com has one customer review of Adam Hardy’s earlier study, Indian Temple Architecture: Form and Transformation, the Karṇāṭa Drāviḍa Tradition, 7th to 13th Centuries (New Delhi: Abhinav, 1995), from a reader “fascinated by ancient Indian temples,” looking for “beautiful pictures with some descriptive text spattered about here and there,” who concluded from its over-many “hand-drawings of details after details” and black-and-white plates that the book “was not for me (a reader with a casual interest in temple architecture), but probably is an excellent source for the academic architect.” Hardy’s new study addresses this audience, condensing his architectural analysis, examining many more of India’s architectural traditions, and illustrating them generously not only with hand drawings and black-and-white plates but also a beautiful selection of color photographs, supplied in large part by that indefatigable photographer of Indian architecture, Gerard Foekma. I think Amazon’s reader, however, might come to a similar conclusion.

Hardy has spent his career deeply committed to training students to see and understand “the set of parts” that, in his view, is used to make up Indian temples. “The aim of this book, however, has not been to provide recipes for transforming a great architectural tradition into new architecture, but to attempt a prerequisite for that: to see the tradition clearly” (243). He has organized his argument in six parts: context and concepts, precursors, design (including chapters on plans, elevation, geometry, moldings, pillars, ceilings, and specific ornament), brief histories of temples in North and South India, and legacy (“What Next?”). He provides a brief glossary, bibliography, and index but no list of illustrations. Photographs are appropriately credited in captions but sources for many drawings re-sketched for this volume are not. Because of the volume’s pedagogical structuring, a reader taking Hardy’s course on seeing is plunged quickly into the complexities of medieval architecture, with little of a historical frame (his “brief histories” begin on 167).

Much scholarship on Indian architecture and texts has occurred in the past half century on which Hardy draws somewhat selectively. Sanskrit terminology for describing temple architecture that is scattered throughout this volume is not sufficiently defined and applied. Hardy founds both his previous volume and this one on a bipartite division of India’s temple “languages” (14), Nāgara and Drāviḍa (taken to be northern and southern respectively). He describes five modes (“shapes”) for Nāgara temples, and divides Drāviḍa temples into two regional dialects from the areas of Tamilnadu and Karnataka. Inscriptions and texts, however, suggest a separate duality for temple names: regional versus formal. Nāgara and Drāviḍa indeed seem commonly to have been used to suggest varieties typical of northern India and the far south, but other regional names also occur in temple lists: Kāliṅga, to name the distinctive temples from eastern India that Hardy discusses in chapter 21; and Vesara, which Hardy acknowledges (134) “is now generally agreed” to refer (in contemporaneous texts and inscriptions, not just modern scholarship) to temples from Karnataka from after the tenth-century, but he then chooses to dismiss it on grounds that “they are not a new mode. . . . it seems best, therefore, to abandon the term” (135). His Nāgara and Drāviḍa, however, also are not “modes” (Hardy had called them languages).

Indian texts provide a separate set of useful modal terms: latina (with vertical spines), kūṭina (with kūṭa aediculae), phāṁsanā (wedge shaped), valabhī (barrel-vaulted), and two eleventh-century developments that seem more formal than regional, bhūmija (marked by levels) and śekharī (multispired). There are shifts and slips in this terminology, but the dual nature of temple names, regional versus formal, needs to be recognized. If Nāgara temples are characteristically latina, they also take phāṁsanā, valabhī, even kūṭina, and, at a later point, śekharī and bhūmija forms.

Drāviḍa temples are typically kūṭina, but also occasionally valabhī and even phāṁsanā. “On the question of whether there are Dravida modes,” Hardy does acknowledge a Drāviḍa variety of phāṁsanā (135; he complicates this by acknowledging that “7th- to 8th-century Phamsana shrines of the Early Chalukyas were Nagara in their detailing”); yet he defines valabhī solely as a “mode of Nagara architecture with barrel-roofed superstructure” (244), and does not generally apply it to the barrel-vaulted Drāviḍa shrines he illustrates. Hardy’s confusion about Vesara is amplified by not recognizing that this regional designation also could take different forms—within what he terms its “Dravida language”—such as kūṭina (with what Ajay Sinha has called “latina logic” [Imagining Architects: Creativity in the Religious Monuments of India, Newark: University of Delaware Press, 2000]), phāṁsanā, and bhūmija (see M. A. Dhaky, The Indian Temple Forms in Karṇāṭa Inscriptions and Architecture, New Delhi: Abhinav, 1977, and Gerard Foekma, Architecture Decorated with Architecture: Later Medieval Temples of Karnātaka, 1000–1300, New Delhi: Munshiram Manoharlal, 2003).

Hardy’s many drawings are those of an architect finding forms, taking them apart, and putting them together again. He focuses much of this creative analysis on what he terms Indian architecture’s “aedicularity” (“A is for Aedicule,” 10) that manifests creation through forms that “cascade” downward: “To imagine, as is often implied, that expansion, in an upward-shrinking shape like a Hindu temple, can be upwards at the same time as outwards, is to mistake visual forces, or the trajectory of the eye or mind, for expressed movement” (38). Hardy’s architecture students have benefited from this emphasis on parts and how they fit together from base to finial, but he stretches the “aedicule” too far. His ideas about “aedicular architecture” are adapted from John Summerson’s essay “Heavenly Mansions: An Interpretation of Gothic” (in Heavenly Mansions and Other Essays on Architecture, London: Cresset Press, 1949) and inspired by the citation in Summerson’s essay of a comment by James Fergusson that “everywhere . . . in India, architectural decoration is made up of small models of large buildings” (10).

Summerson, however, called the aedicule a “little house,” emphasized its function as a “miniature temple,” and pointed out that “the aedicule . . . has been used as a subjunctive means of architectural expression . . . to harmonize architecture of a strictly human scale with architecture of a diminutive [or superhuman] scale” (Heavenly Mansions, 4; emphasis in original). India in fact has a clear name for a temple-aedicule—kūṭa—and acknowledges its importance for Drāviḍa as well as Nāgara architecture in the modal category kūṭina (Michael W. Meister, “Prāsāda as Palace: Kūṭina Origins of the Nāgara Temple,” Artibus Asiae 49, nos. 3/4 [1988–1989]: 254–280). Hardy, however, does not restrict his aedicule to Summerson’s “crowning pavilions” but instead creates a “kuta aedicule,” defined as an “aedicule crowned by a kuta” (244; illus. 11.10d), combining an elevational element with a wall unit. This can help viewers to articulate a Hindu temple’s complexity; and, certainly—in many ways not parsed by Hardy—elevations of most Indian temples grow out of their plans. Yet Hardy’s compound aedicule—combining a full-scale wall section with a “little house”—misses the very “subjunctive” nature of Summerson’s argument. By the time he has deconstructed “aedicules” of a complex śekharī temple (illus. 3.9) it can look more like a field of spindly mushrooms than a buildable monument: “downward and outward emergence is expressed as simultaneous with upward growth of the whole” (41).

Hardy conflates “little houses” in the superstructure, shrine models as niche frames, and niches with ornamental pediments, making virtually all into aedicules (miniature temples). This weakens one of his strongest contributions—analysis of the intertwining surface patterns of north Indian temples made up of gavākṣa (“cow-eye”) window motifs. He wants these always to represent the end of a barrel vault, although he illustrates their use as dormers from straight-edged peaked (phāṁsanā) roofs as well (illus. 16.2b). Although this use is referred to in other examples of this decorative motif (illus. 3.6, 10.1g–h, 16.5), Hardy deconstructs the important seventh-century transitional kūṭina Nagara temple at Rajim as having kūṭa and “Valabhi” aedicules (110, illus. 10.5), basing his analysis on an axonometric drawing from the Encyclopaedia of Indian Temple Architecture (vol. II, pt. 1, Delhi: Oxford University Press, 1988, 2) that makes clear that the central gavākṣa of this tower is projected from a phāṁsanā-like pent-roof.

Hardy calls illus. 10.2g–h “Valabhi aedicules” although they show pent-roof corners behind, and defines all gavākṣa-created pediments as “Valabhi ‘pediment’ designs” (163), suggesting that in every instance they represent barrel-vaulted structures. I think he comes closer to an essential truth when he admits “fluidity between categories” and observes that an “inherent overlap between the Phamsana and the Valabhi arises from” origins of the horseshoe arch in “both the end gable of a thatched barrel roof and the gable of a dormer window projecting out of an overhanging eave” (107).

There are many other nuggets of observation and contestation buried in Hardy’s rich and admirable book. On the origin of Vesara temples in the Karnata region, where Nāgara and Drāviḍa “languages” of architecture had been used side-by-side in the sixth–eighth centuries CE, Hardy remarks, “supposedly Vesara temples are not this kind of deliberate hybrid, but rather are reminiscent of Nagara temples because they have evolved in a similar way, through a similar way of thinking. In the process of transformation, there is no decisive stage . . . it is impossible to say when temples have ceased to be Dravida and become Vesara” (135). Yet there is a decisive moment, conveniently illustrated by Hardy (93, illus. 9.4b–c), comparing ground plans of the tenth-century latina temple at Jagat, Rajasthan, and the early eleventh-century kūṭina Kalleshwara temple at Kukkanur, Karnataka. Kukkanur’s plan emphasizes corner bastions and projects a “bhadra cluster” of offsets, as does that at Jagat. For the first time in its kūṭina superstructure, architects created a stepped (accordioned) barrel-vaulted śālā, to act as an appropriate superstructure for the offset bhadra, and carried offsets in the plan up through the superstructure, culminating in an offset dome. These architects—certainly with decisive intent—proudly placed the motif of a Nāgara kūṭa-stambha (a miniature latina-Nāgara tower poised on a pillar) on the narrow juncture wall connecting the sanctum and large assembly hall in front, signaling their iconic intention.

I find it quite stimulating to struggle with Hardy’s range of perceptions about temple forms. His book is indeed “an excellent source for the academic architect.” The format dictated by the publisher’s series, however, has not allowed for the comprehensive bibliography of recent work that students need. I might have wished more sense of how these temples were constructed, what problems architects faced in particular periods, and how temples were received. Hardy cites Summerson’s opening observation: “There is a kind of play common to nearly every child; it is to get under a piece of furniture or some extemporized shelter of his own and to exclaim that he is in a ‘house’. . . . This kind of play has much to do with the aesthetics of architecture” (10). Of the aedicule Summerson concluded: “This miniature temple used for a ceremonial, symbolic purpose may even enshrine one of man’s first purely architectural discoveries, a discovery re-enacted by every child who establishes his momentary dominion under the table” (Heavenly Mansions, 3–4). It is a complement to Hardy’s enthusiasm as an architect encountering a remarkable architecture that I find him joyfully crouching under the table still.