The cover of the November–December 2014 issue of the Indian Embassy in France’s journal Nouvelles de l’Inde featured an image of an “Indian Mona Lisa” (Figure 1.1). This 2006 oil on canvas was painted by Gopal Swami Khetanchi (b. 1968), who studied art at the University of Rajasthan in Jaipur and worked as an illustrator of magazines and as an assistant art director for Bollywood cinema before returning to Jaipur as a full-time artist (Bahl and Puri Reference Bahl and Puri2011: 12). His painting certainly was a good choice to grab the attention of a French audience, based as it was on the instantly recognizable image of Leonardo da Vinci’s Mona Lisa, indisputably the most popular attraction of the Paris Louvre Museum. The famous image has of course been the butt of many similar transfigurations (documented in Maell Reference Maell2015), most recently sporting masks in the midst of the COVID-19 crisis, sparking “le plaisir du déjà-vu” (Saint-Martin Reference Saint-Martin2020). Khetanchi’s incarnation of “La Gioconda,” besides sporting an admirably slender waist, was dressed up richly in Indian style, with a diaphanous veil with golden pattern and border, red upper-garment (choli) with golden rim, several layers of pearl garlands interspersed with gold jewelry, and matching gold and pearl bracelet, earrings, arm, head, and nose ornaments. Clearly, she represented an Indianized version of Mona Lisa.

Figure 1.1 Bani Thani by Gopal Swami Khetanchi. 2006. Jaipur, India. Oil on canvas. 50.8 × 76.2 cm.

Courtesy Gopal Swami Khetanchi.

What might have escaped the non-Indian observer is the specific model for the Indian elements: These were not random but based on a mid-eighteenth-century painting of the Kishangarh school of Rajput painting. This “Lady with Veil” has been attributed to the master Nihālcand and is well-known in India as it was featured on a postage stamp issued in 1973 (Figure 1.2). Not only does the costume of Khetanchi’s Indian Mona Lisa match the lady’s, so does her delicate pose, with her hennaed right hand preciously pulling the veil, and the left one balancing two lotus stalks. While Khetanchi chose the three-quarter profile and frontal gaze of the famous portrait by the Italian master, he tilted the left eye somewhat upward to more closely approach the typical curve of the famed exaggerated Kishangarhi eye. Together with the nose ring, this created the effect of rendering Mona Lisa less melancholic as she meets the onlooker’s eyes slightly cynical, perhaps somewhat cheekily. This prompts the observer to wonder – the eternal Mona Lisa question – what lies behind this amused smile?

Figure 1.2 Indian postal stamp Radha–Kishangarh. Issued in 1973. Based on Portrait of Radha attributed to Nihālcand. Ca. 1740.

Ministry of Communications, Government of India, 1973

The magazine did not offer an explanation for the choice of this image for its cover. Its contents constituted a special issue on contemporary Indian literature, but the possible link with the lady was left up to the reader to ponder. Perhaps the connection of the location of Khetanchi’s studio in Jaipur, the locale of the eponymous Jaipur Festival of Literature featured in the issue, was assumed to be obvious. Yet the “Lady with Veil” alluded to by Khetanchi’s painting does have a deep literary connection, even if few realize it. The key lies in the official title of Khetanchi’s portrait: Bani Thani. This is the nickname of the purported real-life model, whose facial traits are believed to have inspired the eighteenth-century painting. Banī-ṭhanī was a concubine of prince Sāvant Singh of Kishangarh (1699–1764), the patron of the portrait. The prince is known to have commissioned many devotional paintings to match his own poetry in praise of Krishna and Rādhā (Pauwels Reference Pauwels2015; Reference Pauwels2017). This particular painting is often understood to be a portrait of the Goddess Rādhā as he described her in his poetry, based on his concubine’s striking features.

“Banī-ṭhanī” or “Miss Decked Out” is famous for her looks and elegant sense of style, and widely regarded as the Indian equivalent of La Gioconda, the wife of Francesco del Giocondo who commissioned the Italian Mona Lisa painting. Like her famous Italian counterpart, Banī-ṭhanī is also shrouded in mystery, and it has been questioned whether she really was the model for the portrait. The most significant difference though is that Banī-ṭhanī was herself an artist, a performer as well as an author of devotional songs, which she composed under the pseudonym Rasikbihārī. It is entirely fitting that she would be featured on the cover of a magazine dedicated to Indian literature. So far though, hardly any of her poems have been translated, or even edited. This book allows us to hear her voice, featuring many of her poems translated and edited from a newly discovered manuscript. The main pursuit of this book is the connection of the portrait with literature: It is the search for the poetess behind “Lady with Veil,” Khetanchi’s work’s eighteenth-century model.

Before we meet Rasikbihārī the author afresh, we trace what we know of the “Indian Mona Lisa” and how we know it. This first chapter launches the book’s discovery project by first unpacking the implications of this trope, so compactly visually represented in Khetanchi’s painting. It seems to imply that the Indian master’s portrait is copycat work, but that is far from the truth. At first sight, this image may seem to partake in orientalism in its classical Saidian formulation: a projection of European fantasies onto the non-European “other,” rendering “the Orient” as a mirror image for the West (1978). The image can be interpreted as lacking realism in the stylization of the sitter’s individuality into a type, as reducing “the Orient” to the domain of luxury and wealth, symbolized in the lady’s rich adornments and of sensuality and eroticism, epitomized by the perceived model being a performer from the harem. Khetanchi’s Bani Thani seems to fit the prejudice to a T. But why would a contemporary twenty-first-century Indian artist employ an orientalist construct in his art?

This chapter investigates what is behind the trope and why it is so powerful. What does the comparison with Mona Lisa mean for a contemporary Indian audience against the background of an emerging market for Indian art among the rising middle classes? A sample of the reception of Khetanchi’s painting is the starting point for exploring current perceptions about the painting and the purported real-life model (Section 1.1). Where do these discourses come from? First, the “Lady with Veil” referenced in Khetanchi’s work is placed in her own art-historical context, within an eighteenth-century portrait gallery, to sort out whether the labels now current for the portrait would have fit contemporaneous notions when it was made (Section 1.2). Then, we trace the twentieth-century art-critical evaluation that produced the theory that Banī-ṭhanī was the model. Who were the agents, colonial (Section 1.3) and postcolonial (Section 1.4), that promoted and opposed the Mona Lisa trope? How did “Lady with Veil” become known as Portrait of Radha? Why did she go on to become a nationally recognizable image symbolized by the stamp issued in 1973 and end up as a cyber-orientalist tourist icon promoting her beauty and romance with the prince, while neglecting her authorship (Section 1.5)? This first chapter unravels the orientalist trope to prepare the way for an exploration of the artist behind the pretty cover of Nouvelles de l’Inde’s special issue on Indian literature. This act of “unveiling” sets us up for the quest to discover the woman author implied in the painting in the rest of the book

1.1 Returning the Gaze: The Orientalist “Indian Mona Lisa” Trope Subverted

The reception in India of Khetanchi’s canvas brings to the fore some important questions at the heart of Indian cultural appreciation today. A short BBC Hindi article from 2012 featuring the painting posed in its title the question that was posed to several respondents: “What if the Mona Lisa were Indian?” (Agar Monā Lisā bhāratīya hotī to?)Footnote ¹ This counterfactual title implies Khetanchi’s painting throws a gauntlet, raises a challenge to the Western masterpiece in its reference to the eighteenth-century Indian one. All interviewees agreed that by fusing both, Khetanchi had forged the best of Western and Eastern beauty ideals. The move of dressing Mona Lisa up in Indian garb was read as challenging the notion that Western views of beauty are universal. Thus, the artist Ekeshvar Haṭvāl indicated that Khetanchi’s interpretation confers equal status to both Mona Lisas, and art critic Īshvar Māthur stressed that the Western and Indian beauty ideals each have their own place. This was confirmed by Khetanchi, the artist himself, who called his own work a samāgam or “confluence” of both types, contrasting Western voluptuousness (mā̃saktā) and intoxication (mādaktā) with Indian subtlety (nazākat) and refinement (nafāsat), the latter expressed by Persianate words, evoking sophisticated Mughal culture.

If the appeal of the image lies in the juxtaposition of competing perceptions of East and West, these are not just ideals of beauty, as in intoxicating flesh-and-blood physicality versus more refined subtlety, but of womanhood itself. The interviewees confidently confirm India’s competing not just on equal footing in the East–West beauty contest, but with a sense of superiority. Perhaps originally da Vinci intended to portray the demure and loyal wife of his patron, as exemplified by the clasped hands and slight smile conform etiquette manuals of the time,Footnote ² yet to Khetanchi’s eye she becomes associated with voluptuousness and licentiousness. The lady in the Kishangarhi painting, on the other hand, is decorous and refined. La Gioconda has a delicate, hardly perceptible translucent head cover, but her gaze boldly looks back at the spectator, while her Indian counterpart coyly turns away, drawing the border of her veil presumably to cover her face. Khetanchi’s painting fuses this with a hint of cheekiness, which subtly undermines the stereotype of Indian women as demure. The sophisticated, yet decorous Indianized lady sports an ironic “last laugh.”

In his other artwork, Khetanchi has followed a similar pattern of reworking classical Western portraits of women, often nudes, transforming them into elegantly dressed Rajasthani beauties. Several of those canvases were on view in his 2008 London show “A Tribute to the Masters.”Footnote ³ Rather than dismissing this title as a form of flattery, a gimmick to break into an international art market, one could explore its appeal to middle-class NRI (Non Resident Indian) audiences. Similarly, the Indian interviewees for the BBC article praised Khetanchi’s Mona Lisa painting as an example of “fusion” art (using the English term in the Hindi), and saw it as astutely tapping into what is popular with a twenty-first century audience in India and beyond. The article signals an awareness of the rise of a new market for Indian art in an “India Shining” environment. University of Heidelberg popular culture professor Christiane Brosius characterizes this audience as engaged in a sustained effort to “Indianise modernity and cosmopolitanise Indianness” (2010: 328), which the painting illustrates perfectly. This is not unlike what has been observed for the popular culture of Bollywood cinema that constantly strives to outdo the West at its own game, but with a twist confirming in the end: phir bhī dil hai hindustānī “at heart I remain Indian.” At the same time that popular culture is preoccupied with the glamorous cosmopolitanism of the Indian abroad, it makes it a point to affirm its basically Indian emotional core (Kaur Reference Kaur, Kaur and Sinha2005: e.g., 310–11). This hybrid identity is characterized by an ambivalent relationship between the pull of the cosmopolitan and the call of the (lost) homeland. Yet, compared to the other paintings in the London show, there is more to the “Indian Mona Lisa” than market appeal.

How Bani Thani stands out amongst Khetanchi’s reworkings of Western “Masters’ classics” becomes clear in comparison with its twin, a slightly bigger canvas (61 × 91.4 cm) called Devashree, or Splendour of the Gods (Maell Reference Maell2015: 89). This painting represented simply a Mona Lisa in an Indian outfit, whereas Bani Thani referenced at the same time a well-known Indian master’s classic. The instant recognition of the Western work of art is repurposed to draw attention to the internationally lesser-known Indian one with which the Italian renaissance painting is fused. The kind of mimesis that this artwork performs then, is not simply that of a copycat or plaisir du déjà-vu. It does not take the form of straightforward imitation, or appropriation. It is closer to subversions of high-culture classics, to what inspired Marcel Duchamp’s irreverent 1919 L.H.O.O.Q readymade,Footnote ⁴ than to the self-advertising that treats her as a popular cultural icon, like Andy Warholl’s Thirty are Better than One (1963), or French street artist “Invader’s” Rubik Mona Lisa (Sassoon Reference Sassoon2001: 251–6).

The postcolonial context infuses Mona Lisa’s mimesis with an important dimension of contestation.Footnote ⁵ As is clear from the reception of Khetanchi’s painting, this form of mimicry includes a mockery that allows the artist not just to challenge but to upend Western concepts of art, beauty, and womanhood. In contrast to other Mona Lisas in ethnic dress, such as Alyssa’s in Tunisian costume of 1967 (Sassoon Reference Sassoon2001: 254), Khetanchi’s draws attention to an actual corresponding Indian masterpiece. In doing so, the Indian painter highlights the existence of a rich and in the West little-known art tradition that can rival the best of the Western canon itself. In the process, he shows up the smug obliviousness, the pretense of Western cosmopolitanism in the face of its own ignorance of global alternatives. Simply put, what appears to the undiscerning glance as mimicry, admiration, and imitation, has a deep bottom of critique and reverse mockery rooted in pride in its own heritage. The interviewees’ interpretations of Khetanchi’s work demonstrate how the perceived orientalist trope rather signifies the reverse: The Western imperialist gaze is cheekily returned.

There remains one more aspect of Khetanchi’s painting to unpack, again something less readable to the Western viewer, but brought out by the BBC article referenced here. Khetanchi gave his painting the title Bani Thani. Widely interpreted as the name of a real-life model, the woman whose features ended up defining the Kishangarhi type, this actually exposed Khetanchi to criticism. It was foremost in the mind of the writer and at least one of the interviewees of the BBC article. The Jaipur-based sculptor and Padmashrī award winner, Arjun Prajapati, critically suggested that Khetanchi would have done better to leave the famous Kishangarh portrait alone, as the original was the result of a unique collaboration between the artist and his patron, Prince Sāvant Singh of Kishangarh. Prajapati felt it was based on the prince’s loving descriptions of the features of his ladylove, which the painter then went on to capture on the canvas. Implying the lady in question was in purdah, he saw Khetanchi’s endeavor as disrespectful, and since the original depicted a Goddess, as an irreverent secularizing move.Footnote ⁶ The author of the BBC article chimed in, keeping the historical Banī-ṭhanī behind purdah by imagining her not as a flesh-and-blood woman, but as the heroine of Sāvant Singh’s dream that took form on canvas. Such discomfort with worldly women out of purdah spoke through Khetanchi’s own word choice, as his attribution of voluptuousness (mā̃saktā) to the Western Mona Lisa. An old orientalist discourse on art surfaces here, a colonial view that contrasts Western realism foregrounding materiality (read hedonism) with Eastern idealism stressing spirituality (restraint). Traditional Indian art is essentialized as symbolic and religious, without a direct link with observed reality (Guha-Thakurta Reference Guha-Thakurta2004). All this lays bare some discomfort with the term “Indian Mona Lisa,” which seems to be a contradiction in terms: a real-life naturalistic model for an Indian spiritual ideal type.

Whence came these persistent romantic mystifications? Why does the identification of Banī-ṭhanī as the Indian Mona Lisa seemingly compulsively return to the orientalist construct? Why do none of the commentators refer to her contributions to literature? There seems to be a strong discourse at work that keeps hijacking the conversation. Could this be because the portrait, and for that matter the Kishangarhi school of Rajput painting, was discovered by a British colonial agent who started writing about her in a certain vein? To grapple with those isssues, at this point let us telescope the Khetanchi image into its source of inspiration, the Kishangarh “Lady with Veil.” Before reconstructing the history of the art-historical perceptions that came to be projected onto it in twentieth-century discourse, let us first address its place in eighteenth-century portraiture.

1.2 “Lady with Veil” in the Portrait Gallery of Mughal and Rajput Women

What about the Indian Mona Lisa portrait, “Lady with Veil,” itself? The painting is not inscribed, so we do not actually know whether it was intended to portray the Goddess Rādhā, even less whether it has the features of the Kishangarhi concubine Banī-ṭhanī. Before tracing how it came to be associated with both these identifiers, situating the painting in its historical context helps determine whether either designation would have been likely at the time. We can build on the pioneering archival work of Dr. Faiyāz ‘Alī Khān of Kishangarh in the middle of the previous century and more recently on the art-historical work by Dr. Navina Haidar, Curator at the Metropolitan Museum and Oxford-trained specialist on Kishangarh art, who has written on this very portrait (Haidar Reference Haidar and Dehejia2004). At the outset, it is important to note that the Kishangarh school of painting from early in the eighteenth century developed in close exchange with the Mughal atelier, with artists trained in Delhi moving to the small Rajasthani principality (Haidar Reference Haidar and Dehejia2004; Reference Haidar, Beach, Fischer and Goswamy2011a; Reference Haidar, Beach, Fischer and Goswamy2011b).

“Lady with Veil” itself has mimetic elements of Mughal and European portraits, though no direct connection with the Italian Mona Lisa has been made.Footnote ⁷ It definitely is not a copycat work, but in many ways, it conforms to the type of the single, nonnarrative bust portrayals of idealized beauties designed to be admired in albums that were popular by the eighteenth century both in Mughal and Rajput painting. While actual inscribed portraits of palace women were rare, bust portraits of this type were relatively common.

In Mughal art, the seventeenth century had seen a marked increase in portraiture of women,Footnote ⁸ Abu’l Hasan’s full-length portrait of Nūr Jahān in male outfit with a gun being perhaps the most remarkable. Attributed to the same painter was a bust portrait of a Mughal lady, meant to be worn as a jewel, possibly also Nūr Jahān’s.Footnote ⁹ The first dated (1628) oval bust portrait of an imperial lady was actually the empress’ niece, Mumtāz Mahal. Painted on a mirror case by the Mughal atelier’s master painter ‘Abīd, it was likely inspired by a European allegorical model (Seyller Reference Seyller2010b: 145–53).Footnote ¹⁰ This portrait was intended for her husband Shāh Jahān, for private hands only, as was the 1630–3 album that Dara Shikoh gifted to his wife Nadira Banu Begum with its full-length portrayals of women from the imperial zanānā. This proved trend-setting and an explosion of portraits of ladies occurred in its wake.Footnote ¹¹Window or jharokhā-style profile portraits of palace ladies became popular after 1668, when Aurangzeb abandoned the practice of imperial appearance to his subjects at the palace window (jharokhā), thereby renouncing the previous reservation that only the emperor could be thus portrayed (Losty and Roy Reference Losty and Roy2012: 141–3). A few such bust portraits have been demonstrably modeled after European originals.Footnote ¹² The vogue of unidentified women’s bust portraits (often with revealed breast) spread to the provinces and became particularly popular in the eighteenth century.

The features of the women in these Mughal portraits often strike viewers as generic and stylized, so it has been suggested that these eighteenth-century paintings were essentially ornamental and not intended as portraits at all.Footnote ¹³ The idealizing nature of the portraits is often explained by the painters’ lack of access to ladies in purdah, though it has been established that women painters were active in zanānās.Footnote ¹⁴ Still, art historians often designate the more life-like looking depictions as portraits of more accessible “courtesans” or “harem attendants,” especially when holding a wine cup with hennaed hands.Footnote ¹⁵ Less ambiguously, other portraits depict musicians with their instruments, or clapping their hands, making clear their status as performers.Footnote ¹⁶ A few such paintings are inscribed, sometimes identifying the lady by ethnicity, such as Muhammad Afzal’s circa 1740 Gujarātī Woman (chahrā-e Gujarātin; Losty and Roy Reference Losty and Roy2012: 183, fig. 124). It has been suggested that in the case of palace women behind purdah, the real identity of the ladies was purposely hidden under generic ascriptions to powerful queens who did appear in public, such as Sultānā Chand Bībī, Bijapur regent for her minor son and enemy of Akbar, and most famously Nūr Jahān, Jahāngīr’s consort.Footnote ¹⁷ Notwithstanding the above-mentioned contemporaneous portraits of the powerful empress, it is the posthumous idealized depictions from the eighteenth century that became popular.Footnote ¹⁸ Both courtesan and queen are combined in the case of Lālkuṃvar, the performer-turned-empress married to the emperor Jahandar Shāh who was deposed in 1713 (Figure 1.3). For many portraits, whether idealized or real-life, a common characteristic is that the ladies are depicted with one hand delicately raised towards the face, often holding a flower or a cup. Our “Lady with Veil,” with her hand raised toward the face, preciously holding her veil, certainly does not look out of place in the gallery of such Mughal lady portraits.

Figure 1.3 Portrait of Lal Kunwar. Eighteenth century. Mughal, India. Color-wash drawing. 32.6 × 21.5 cm.

The Walters Art Museum, Baltimore (Walters manuscript leaf W.712).

Since Rajput portraiture generally developed in close interaction with its Mughal counterpart (Desai Reference Desai, Schomer, Erdman, Lodrick and Rudolph1994: 313), it is to be expected that depictions of Rajput ladies partake in the same characteristics. Some Rajput jharokhā-style lady portraits are clearly copies of Mughal examples, such as a Kishangarhi mid-eighteenth-century idealized copy of the famous Nūr Jahān portrait, which seems a likely candidate of being intermediary towards “Lady with Veil” (Figure 1.4).Footnote ¹⁹

Figure 1.4 Idealized Portrait of the Mughal Empress Nur Jahan (1577–1645). Ca. 1725–50. Kishangarh, Rajasthan. Opaque watercolor and gold on paper. 29.52 × 21.59 cm.

Los Angeles County Museum of Art, Gift of Diandra and Michael Douglas (M.81.271.7). Photo courtesy LACMA.

Like their Mughal counterparts, Rajput portraits of women also rarely identify the model (Aitken Reference Aitken2002: 449–50).Footnote ²⁰ A few portraits were inscribed with a generic noble woman’s name, for instance “Jodhabāī.”Footnote ²¹ The absence of identifications of the sitter can be understood with reference to the male-exclusive spheres of power within which portraits circulated, namely the genealogical record and gift-giving to ensure allegiance (Aitken Reference Aitken2002: 254–6).Footnote ²² Here, the question whether painters had access to palace women to sketch a portrait based on observation looms large. It seems self-evident that women in the Rajput zanānā could not be observed, so it is not surprising that popular press articles on Banī-ṭhanī, like the BBC piece quoted above (Agar Mona Lisa Bhāratiya hoti to?) take for granted the impossibility of direct observation for the Rajput harem portraits. One respondent assumed that Sāvant Singh himself either described in words or actually sketched a portrait of his mistress for his favorite painter Nihālcand to work with. Possibly, he had in mind an existing sketch “portrait of a yogini” that was drawn by the crown prince himself (Pauwels Reference Pauwels2015: 201–3). However, that drawing was not inscribed with reference to Banī-ṭhanī, nor actually do the features of this “yogini” show the extraordinary profile the singer is famous for. The point is rather that there may actually not have been any need for such intermediation by the prince. As a court singer, Banī-ṭhanī was not behind strict purdah.

As in the Mughal case, Rajput portraits of musicians are plentiful.Footnote ²³ A few such Rajput portraits of performers carry identifications, often with the epithet of bhagtan or “courtesan,” which confirms that they were sometimes based on living models.Footnote ²⁴ All this indicates that the assumption that individual women were not depicted needs to be nuanced. Rather, as Aitken perceptively suggests, portraits existed but were often not inscribed with the women’s names and thus the contextual information of who was the model was lost over time (Aitken Reference Aitken2002: 256, 273–4).Footnote ²⁵ In the case of Kishangarh, there are examples both in full and three-quarter profile (see Figures 1.5 and 2.2 in this book). While not inscribed, one of them certainly shows the characteristics associated with Banī-ṭhanī, in particular the elongated eye and brow, and the prominent nose (Figure 1.5). However, does that make the image a portrait of Banī-ṭhanī? It does not take much to see it as a forerunner of the more stylized and exaggerated features of “Lady with Veil,” where she holds a lotus instead of a musical instrument. However, the conundrum is whether Banī-ṭhanī was the model for or modeled after the distinctive Kishangarh style, an issue to which we will return (Section 3.1).

Figure 1.5 A Lady Singing. Ca. 1740–5. Kishangarh, Rajasthan. Painting on paper. 48.2 × 35.2 cm.

Collection of Howard Hodgkin, loan to Ashmolean Museum, Oxford (LI118.31).

Scholars have posited that “most depictions of women of the Rajput courts were generic and not portraits of actual people. In place of specific character traits, artists highlighted the feminine sophistication, beauty and mesmeric behavior of the women” (Mishra Reference Mishra2018: 137).Footnote ²⁶ Molly Aitken has sought to nuance this common assumption, citing examples that show it was at least not inconceivable painters had access to palace women (2002: 149–51), though she hastens to specify that the women were portrayed not individually but conventionally, with the stylistic face of the local school of painting (272–5).Footnote ²⁷ Columbia University art historian Vidya Dehejia has plausibly suggested that this development of formulaic portrayal was used “in a manner akin to the use of a flag or insignia, as a feature that individualized and distinguished them from adjoining princely courts… most prominently the female form became the hallmark of a state” (1997a: 362). This applies well to Kishangarh, which is the example Dehejia cites. In that light, the Kishangarh court’s insistence that “Lady with Veil” is not a portrait after a real-life woman is well-justified.

Ignoring the model’s actual physiognomy is all the more pertinent for women portrayed within more elaborate court or harem scenes where their husbands or patrons are central. The aesthetic preference for uniformity and generically stylized features for everyone but the king sidelines specifics of individual appearance, not just for the ladies, but for courtiers and attendants in general. However, within strongly hierarchical Rajput society, one suspects that the participants in the activities portrayed would have been very sensitive to who was represented where in the hierarchy, and which ladies were singled out for attention, even if the names were not recorded. This would seem obvious in particular for hunting scenes that depict ladies’ shooting exploits. (For an example from Bundi ca. 1760, see Bautze and Angelroth Reference Bautze and Angelroth2013: 116–7.) In the Kishangarhi context, there are several portraits that inscribe courtiers by name just above their depiction or have keys on the back (examples in Pauwels Reference Pauwels2015: 70–71, plates 2 and 3). Awareness of the identity of the portrayed then seems to have been keen, even when the portrayal was not individualized.

The conundrum of verisimilitude is tied up with the determination of intention, that is, whether specific individuals are intended to be portrayed in a historical location, or conventional scenes with generic characters. In the case of both Mughal and Rajput painting, some “portraits” actually depict ideal-type heroines, rāgamāla series, scenes from Krishna’s life, or generic conventional harem scenes, like ladies bathing, holding birds, making music, or carrying water pots (panihārin). A Kishangarhi example is a well-executed full-length portrait of a charming water carrier by the lake with her pot put down, waiting for her beloved, presumably the horseman in the background. This has been attributed to Bhavānīdās, the painter who moved to Kishangarh from the Mughal atelier.Footnote ²⁸ Strikingly, like the Mughal jharokhā portraits, she too has one hand raised delicately to the height of her face. This portrait, dated around 1725, can be related to a similar image that is attributed to Bhavānīdās’ son Dalcand and slightly later, ascribed to his maturity phase, 1730–40 by McInerney (Reference McInerney, Beach, Fischer and Goswamy2011: 574–7, fig. 13). In this case the lady is holding a flower in her hand raised to shoulder height, like the portrait of Lālkuṃvar (Figure 1.3). Perhaps the garland that the lady is holding in her other hand hints at her intent to garland her “groom” upon arrival. Here, too, there is a horseman in the background; clearly it is intended to echo the father’s painting. One suspects there is more going on here than generic depictions of ideal types, but the meaning behind that is now obscure. Both full-length portraits may well be taken to foreshadow the more stylized later bust portrait “Lady with Veil.”

Set portrayals sometimes illustrate so-called Rīti or mannerist poetry, which features a catalogue of hero (nāyaka) and heroine (nāyikā) types. This genre is found in classical Hindi literature; most famous perhaps are Keśavdās’ Rasik-priyā (The Connoisseur’s Darling) and Bihārī’s Satsaī (Seven-Hundred Poems), two of the most-illustrated classics of Hindi literature. The nāyaka-nāyikā genre has long antecedents in Sanskrit literary categorizations but is also discussed in Persianate-inspired Hindavi literature, following the exposé by Abū’l Fazl in ‘Ain-i Akbari.Footnote ²⁹ Thus, what looks like a portrait may be an illustration of a nāyikā subtype, for instance, of the virahinī, the lady pining in the absence of the lover.

“Lady with Veil” has been identified by some as an example of a nāyikā of the type vāsaka-sajjā, “All dressed up (sajjā), awaiting her lover in her room (vāsaka)” (Dickinson Reference Dhavan and Pauwels1950: 35). This category of heroine is typically portrayed in painting as eagerly awaiting her lover with the decorated but empty bed ready nearby (vāsaka-śayyā; Coomaraswamy Reference Coomaraswamy1916: 51–2). The heroine depicted in the “Lady with Veil” painting certainly is beautifully dressed up, but there is no hint of the waiting bed that is the usual giveaway in pictures of this type of heroine. Neither is her demeanor expressive of the mood of anxious anticipation due to the lover’s delayed arrival. The coquettish smile and flirtatious gesture of adjusting her veil to reveal her beauty rather suggest that the lover has arrived, and we witness the play of seduction. So perhaps she does not fully conform to the traditional trope, but there may be another classical Hindi poem underlying the illustration. More on this will follow, as we discuss the art-historical journey of the portrait (Section 1.4).

To complicate matters, there is evidence that some of the seemingly abstract and conventional Rīti literature itself was in fact more or less obliquely directed to specific “courtesans” or palace ladies. One famous example from the late-sixteenth, early-seventeenth century is the aforementioned poet Keśavdās, who in his Kavi-priyā’s The Poet’s Darling frequently references his “disciple,” Pravīnrāy, the courtesan (pātur) at the Orchha court (Dehejia Reference Dehejia2013: 10–11). Distinctions between depictions of ideal-type and of real-life woman could become blurred, as when generic harem scenes are inscribed with the names of pāsbān or dancers from Mewar and Marwar, and when a Rāginī personified is identified on the painting as a Jaisalmer princess (Aitken Reference Aitken2002: 272–3).Footnote ³⁰ Such would have enhanced the piquancy of the poems and paintings for the insiders, while remaining largely unrevealed to outsiders. This is not limited to the ladies. Court panegyrics, composed for specific rulers, depicted them, too, as ideal “heroes” or nāyakas.Footnote ³¹

This ambiguity and conflation is further deepened, as standardized depictions in apparent Rīti style are often identified as the Goddess Rādhā, thus imbuing the poetry with a devotional (bhakti) aspect.Footnote ³² That explains why “Lady with Veil” is deemed not just an ordinary portrait but reckoned to fall in the category of the idealized portrait of a deity, Portrait of Radha. Parallel depictions of Krishna with the features of rulers are well documented. Vidya Dehejia sees such conflation of God and “hero” or nāyaka (2009: 159–99) as a logical outcome in a cultural universe that has “routinely blurred the boundaries between sacred and profane,” and that sees gods as “the prototype for all human lovers” (161). Yet she also notices how the portrayal of women as “heroines” or nāyikā (and as consorts of the divine) is different from that of the rulers. The features of the heroines tend not to be differentiated from those of other women members of the court but identified only by contextual placements with their divinized partners. The “Lady with Veil” case is exceptional as, at least here, the nāyikā stands on her own.

How likely is it that an artist would depict the Goddess based on an individual woman’s features? There is evidence of portraits of historical royal women idealized as deities in Indian art. In sculpture, most famous is the example of the tenth-century Chola Queen Sembiyan Mahādevī portrayed as Pārvatī (Dehejia Reference Dehejia1998; bronze in the Freer Gallery of Art, Smithsonian Institution). Dehejia provides the ritual context when the image was carried in procession on the occasion of festival celebrations. She perceptively connects this with images of Tamil saints-devotees, who are carried in processions that elicit narratives about their devotion (1998: 43–5). Similarly, the case of Sāvant Singh and Banī-ṭhanī is seen “to allude to their alleged special relationship” with Rādhā and Krishna (Crill Reference Crill, Crill and Jariwala2010: 37–8). In other words, the god-portrait might be intended to highlight the sitter’s devotion.

This can also be understood as an extension of portrayals of rulers partaking in the realm of the divine, as was popular at the time in the small principalities of Bundi and Kota. Possibly there is a link with the Vallabhan school of nearby Nathdwara that came into its own with depictions of the deity of Śrī Nātha Jī worshiped by his priests and visiting royalty. Those were called manoratha or “vow” as they commemorated the pilgrim’s visit often undertaken in the context of a special vow.Footnote ³³ Next to the patron might be portrayed his family, including the ladies (Ambalal Reference Ambalal1987: 80). This is not limited to the Vallabhan sectarian milieus; kings are often shown with their own favorite or state deities, who are sometimes pictured in an anthropomorphic way, making it hard to distinguish between them and human personalities, conflating king and God.Footnote ³⁴ Not much work has been done on non Vallabhan traditions as of yet, but notable is a drawing of Savāī Jai Singh II and the Caitanya Sampradāya image of Govindadeva Jī that he installed in his new capital Jaipur (Tillotson and Venkateswaran Reference Tillotson and Venkateswaran2016: 68). This image also features a woman fanning the deity, perhaps intended to represent a palace lady. However, it might be just as well a mythological figure, as she resembles the Goddess Rādhā and sakhī Lalitā in the painting next to the deity itself. On the basis of this confusing evidence, at least it is fair to say that Rajput paintings routinely show several levels of conflation of divine and royal realms.

Rulers often project themselves into the divine world, conforming to the common devotional technique of participation in dramatized performances of the deities’ lives. Mewar ruler Jagat Singh II (r. 1734–51) famously had himself portrayed assisting in such musical plays (e.g., Tillotson Reference Tillotson1987: 6–7, fig. 4; Khera Reference Khera2020: 90, fig. 3.1). We know that Sāvant Singh took part in Rās-līlās, or Krishna devotional plays (Pauwels Reference Pauwels2017: 91–105). There is evidence from Kishangarh paintings from the time of Sāvant Singh’s father, Rāj Singh, where the royal family is portrayed as attendants participating in celebrating Krishna’s marriage ceremonies (Haidar Reference Haidar, Beach, Fischer and Goswamy2011a: 543–4; Pauwels Reference Pauwels2015: 152–6). Nihālcand goes a significant step further in the identification of the devotee-prince with Krishna himself but it is not unparalleled; other kings also had themselves pictured as God incarnate.Footnote ³⁵ In the case of Sāvant Singh, perhaps the male devotee’s promotion can be understood in the light of his devotional name “Nāgarīdās,” or “Servant of Rādhā.” Isn’t Krishna himself Rādhā’s greatest servant? Still, portraying a concubine as Rādhā? We should be careful not to introduce unwarranted Western art-historical tropes of Madonnas modeled after real women, including mistresses of painters and their patrons.Footnote ³⁶

To sum up, after weighing all the pros and cons, our conclusion has to be nuanced. According to contemporaneous practice, we can reasonably assume that “Lady with Veil” illustrates a description of a heroine according to classical poetic conventions, and such nāyikās would frequently be identified as Rādhā. The interpretation of “Lady with Veil” as Portrait of Radha then seems quite apt. The unofficial title “Banī-ṭhanī” though is more dubitable, and whether the portrait was actually inspired by the beautiful concubine’s features remains up in the air. The identification is not impossible: There are examples of Rajput portraits of performers at court and stylized images may be intended to portray actual court ladies. At the same time, this case is not established as fact, as is sometimes assumed. What we can say with confidence is that the portrait was conceived within a realm of beauty ideals, imagery, and poetics that set Mughal and Rajput, classical Hindavi and Hindi literature into rich dialogue. In literature, as in painting, there is a generic conflation between portrait, depiction of the ideal, and of the divine.

1.3 Colonial Construct? An Aesthete’s Discovery of Portrait of Radha

How has “Lady with Veil” been interpreted subsequently over time? Khetanchi’s “Indian Mona Lisa” painting accomplishes visually the identification of Banī-ṭhanī as Rādhā with that of La Gioconda as Mona Lisa. This is the apotheosis of a development that has been in the making for about eight decades, forged in the crucible of colonial and nationalist discourses since the “discovery” of Kishangarhi art by the international art world. The Kishangarh school of painting was brought to the attention of the Western and Indian art connnoisseurs by Eric Charles Dickinson (1893–1951), a poet and short story writer who had been professor of English at Government College in Lahore since 1928 (Singh Reference Singh1952). In 1943, on an educational delegation to Mayo College in Ajmer, Dickinson’s party made a side trip to Kishangarh, where the young Rājā Sumer Singh (r. 1939–71), himself a student at the college, still reigned under a Council of Regency. Seven years later, Dickinson himself described the incident with dramatical flair in an article published in the prestigious Indian art Journal Mārg:

The orientalist trope of the thrill of discovery has been a mainstay in popularizing Kishangarhi painting ever since. In his insistence on seeing the Rajput artworks, which he preferred above the “tedious” Mughal ones, Dickinson undoubtedly was inspired by the influential art historian Anand Kent Coomaraswamy (1877–1947), who in his path-breaking book Rajput Painting published in 1916, had raved about the Rajput paintings’ exceptional emotional depth in comparison to Mughal ones:

Dickinson felt the discovery of his own “magic world” was cemented through establishing a connection between the images and the texts that were written on the reverse:

He returned to Kishangarh with the express purpose to explore further the link of the paintings with texts, and found out about the collaboration of Sāvant Singh, alias Nāgarīdās, the poet-patron, and the painter Nihālcand.Footnote ³⁷ He describes his reaction in rhapsodic terms:

In stressing the interface between paintings and vernacular literature, Dickinson again took his cue from Coomaraswamy. As Keeper of Indian Art at the Boston Museum of Fine Arts, the latter would frequently assess newly acquired Rajput paintings:

For Coomaraswamy, the link between painting and literature was directly related to the idealism and spirituality of Indian art.Footnote ³⁸ Similarly, Dickinson was prone to see Rajput art at its finest as religiously inspired “Hindu art.” Granting that the paintings were stylistically aware “of Moghul technical innovation and linear purity they yet, in inspiration, remain faithful to the Rajput ethos.” He even spoke of a “Hindu art renaissance flourishing … following the decline of the Moghul pre-eminence” (1950: 37). Coomaraswamy had devoted a full chapter of his landmark study, Rājput Painting, to Krishna Līlā, with a special appendix on the cult of Śrī Nātha Jī (1916: 2:26–41). Dickinson learned about the Kishangarh court’s view that the religious significance of Nāgarīdās’ poetry was connected with this same influential Krishna devotional movement, whose main temple was that of Śrī Nātha Jī. The Vallabha Sampradāya, named after the sixteenth-century philosopher Vallabhācārya, had a non-ascetic approach to the divine that made it attractive for wealthy householders, including rich Gujarati merchants as well as Rajput rulers. This system became known as Puṣṭi-mārga or “The Path of (God’s) Sustaining Grace,” which Dickinson seems to have understood as “The Way of Pleasure,” the title he gave to his 1950 article. Hence, he felt there was a religious justification behind the “frankly hedonistic appeal of the paintings” (1950: 34).

Dickinson provided the caption Portrait of Radha for the painting that later would become the basis for the Radha–Kishangarh stamp, and eventually Khetanchi’s fusion canvas (1950: 35). He was the first to cast it as the archetype for the Kishangarhi facial traits and attribute it to Nihālcand. He reckoned it was:

To support his point, he cited a lush poem in English, presumably a translation of a Hindi work by Nāgarīdās, but he did not indicate the original source.Footnote ³⁹

Dickinson’s was an aesthete’s perspective,Footnote ⁴⁰ and he classified the image according to Indian aesthetics:

This analysis employed the terminology of Indian aesthetics. Again following Coomaraswamy (Reference Coomaraswamy1916: 2:42–54), he classified according to theories of Śr̥ṅgāra rasa or the “erotic sentiment” and the typology of heroines (nāyikā). This is sophisticated, even if one might quibble with his choice of the particular type of the vāsaka-sajjā “all dressed up and awaiting her lover on the bed” (as discussed in Section 1.2).

Dickinson spoke of the Rādhā in the portrait as a “tender maiden,” a coy virgin on her wedding night.Footnote ⁴¹ These musings evoke bridal mysticism, which made sense in the context of the preoccupations in contemporaneous art-historical writing about India. Again one can compare with Coomaraswamy’s views that glorified the purity of the Indian bride in “rhetorical pamphlets on The Oriental View of Woman (1910) and Sati: A Vindication of the Hindu Woman (1913), where the act of Sati (also spelled suttee) was glorified as ‘Eternal Love,’ representing the most sacrosanct image of Indian womanhood” (Chattopadhyay and Thakurta Reference Chattopadhyay, Thakurta and Bagchi1995: 164). The famous art historians, Mohinder Singh and Doris Schreier Randhawa, writing in 1980, deemed Dickinson’s evaluation of Kishangarhi art as subscribing to “the romantic cult of innocent womanhood” (1980: ix). There was a palpable tension in his attempt at framing the image within Indian aesthetic categories that presumed a sexually experienced heroine, and at the same time the urge to represent the essence of Rajput art as pure spiritual love, foregrounding an innocent one.

Was Dickinson the one who came up with the Mona Lisa trope? Certainly, comparison with European art was typical for the art-historical discourse of the time. It made sense in a context where scholars of non-Western art felt the need to advocate for their subject in terms familiar and appealing to a largely Western audience. This had become even more pertinent against the background of political assertion of the struggle for independence, when the need arose of “acknowledging South Asia’s arts as fine arts, worthy to rival the European canon” (Aitken Reference Aitken2016: 10). Dickinson, too, invoked Western parallels in his writings. He did not mention da Vinci, though he compared “Lady with Veil” with the celebrated profile portraits by fifteenth-century Italian masters (1950: 35).Footnote ⁴² He also brought up other profile-oriented painting styles, such as the Minoan cupbearers of Knossos and the art revolution under the 14th-century BCE Egyptian pharaoh, Akhetaton. Perhaps the latter betrays that Dickinson saw himself as tracing the footsteps of Egyptian archeologists, such as the German Egyptologist Ludwig Borchardt, who in 1912 discovered in Amarna the bust of Akhetaton’s Queen Nefertiti. Akhetaton’s love for art, the “realism” of the bust modeled after the queen’s features, and the pair’s shared religious inspiration would parallel the Kishangarhi case. It must be said though that Dickinson did not mention Nefertiti at all in his published writings on Rādhā’s Kishangarhi portrait.

In his efforts to promote Kishangarh to the top of world art, Dickinson kept racking his brain for Western parallels. Here he parted company with Coomaraswamy who, in his enthusiasm for the religious aspect of Rajput painting drew mainly parallels with medieval European art (Mitter Reference Mitter1977: 279). Instead, Dickinson eventually settled on Jean-Antoine Watteau’s roughly contemporaneous (1717) Embarkation for Cythera (the island where Venus was born) (L’Embarquement pour Cythère) (Dickinson Reference Dhavan and Pauwels1950: 33).Footnote ⁴³ This painting, submitted to the Académie Royale in Painting and Sculpture had inaugurated a new genre, that of the Fête galante or “arcadic revelry.” While he did not elaborate, Dickinson here brought up an interesting comparison. The nostalgia for the pastoral innocence of mythic Arcadia during this post-Louis XIV Régence period of decadence had a lot in common with that for the Krishnaite pastoral of Sāvant Singh’s time. The late Indologist Alan Entwistle identified several parallels between French arcadian and Kishangarh’s Vaishnava paintings, at the heart of which was the celebration of unencumbered love in bucolic scenes through dramatic staging. Addressing interpretations of such fantasies as escapism, Entwistle perceptively pointed out the tensions at work when sophisticated courtiers fetishize the countryside’s charms, sometimes as idealized childhood experiences (1991). While Dickinson did not spell it out as elaborately, he too was on to something more profound than the shared theme of savvy courtiers frolicking in the countryside.

In sum, the British teacher at Lahore college, Eric Dickinson, was instrumental in bringing Kishangarh paintings to the attention of the Western (and Indian) art world. He enthusiastically described them in the style of contemporaneous art-historical treatment of Indian art, for which Coomaraswamy had set the tone.Footnote ⁴⁴ This discourse partook in colonial practices of comparisons with European art and orientalist tropes, including fascination with the timeless spirituality of the Orient. Following Coomaraswamy’s spiritualization of the feminine in Rajput paintings, Dickinson characterized “Lady with Veil” as Portrait of Radha, and attributed the Kishangarhi “Hindu renaissance” to sectarian religious inspiration. In relating Kishangarh’s art to the best of the West, Dickinson went beyond Coomaraswamy’s parallels with classical or medieval art, comparing with the “arcadic revelry” paintings of Watteau for the Versailles nobility. Departing from “Coomaraswamian anonymity” (Ehnbohm Reference Ehnbohm and Barnett2002: 181), Dickinson attributed the prototype of the Kishangarhi facial type to the master painter Nihālcand who worked with the patron-poet Nāgarīdās. Introducing historical concerns, he ascribed the development of this distinctive style to rivalry with nearby ateliers of Jaipur and Jodhpur.Footnote ⁴⁵ This concern with historicity too differed from Coomaraswamy’s orientalist notions that tended to be essentialist and idealist (as evaluated by Mitter Reference Mitter1977: 279–86; see also Singh Reference Singh2013: 257–60). If Dickinson’s surmise is right, it is ironic that what started as a symbol of regional pride of Kishangarh over Jaipur, became in the twenty-first century a symbol of national Indian pride asserting superiority over Western ideals of beauty, art, and womanhood in the hands of the Jaipur painter Khetanchi. Nowhere in his single-authored articles though, does Dickinson make explicitly a parallel with the Mona Lisa, thus he was not the originator of that “colonial construct.”

1.4 The Search for the Model behind “Lady with Veil”

If the “Indian Mona Lisa” trope did not strictly speaking come from the pen of the colonial discoverer of the art, whence came the now common association apotheosized in Khetanchi’s contemporary double-portrait? Two further steps were taken: first, the identification of Banī-ṭhanī’s features as the basis for the Kishangarh type, and second, the comparison of this Kishangarhi model, concubine of Sāvant Singh who was the patron of the painting, with La Gioconda, the wife of Francesco del Giocondo who commisssioned da Vinci’s Mona Lisa. Dickinson’s probing questions about the possibility of a real-life model for the striking features of the Kishangarhi face functioned as a catalyst for the crystallization of the identification, but there were other agents who played a definitive role in creating the myth. First comes to mind the man who saw Dickinson’s work posthumously through to publication and added his own insights, the lawyer and art connoisseur, Karl Jamshed Khandalavala, about whom more below. But there were other important players who paved the way. Several early Indian contributors to the construction of this trope have been neglected while attention was focused on Dickinson’s discovery.

Dickinson himself ignored an article on Nāgarīdās that was published as early as 1897 in the Journal of the Asiatic Society of Bengal.Footnote ⁴⁶ The author was Pandit Mohanlāl Vishnulāl Pandia, formerly the minister of the state of Pratapgarh in Rajasthan. He had already mentioned Banī-ṭhanī in connection with Nāgarīdās (1897: 66–7). He even provided a sample of her poetry under the pen name Rasikbihārī, correcting the common perception at that time that the poet by that name was a man. Few commentators mention this article, even though it appeared in English in a prestigious journal. Even fewer mention Pandia’s acknowledged source: the Hindi work of Bābū Rādhākrishnadās, the first director of the organization for the promotion of the Devanāgarī script or Nāgarī Pracāriṇī Sabhā of Varanasi (NPS, established in 1893).Footnote ⁴⁷ He had given a presentation on Nāgarīdās in 1894, which was published by the Khaḍag Vilās Press, but subsequently rewritten with input of the Kishangarh court’s chronicler Kavi Jaylāl for the foreword of the first lithograph edition of Nāgarīdās’s works published in 1898 (Gauḍ Reference Gauḍ1898: 1–30; see Section 5.4). These publications focused on Kishangarhi literature and perhaps that is the reason they did not make it onto the art historians’ radar. While the focus here is on the paintings, we will have occasion to return to the work of the Hindi scholars (especially in Chapter 5).

The giant on whose shoulders Dickinson was standing (and all of us writing on Kishangarh are indebted to), was another savant-courtier, this one at Kishangarh itself, by the name of Faiyāz ‘Alī Khān (1911–2001). By virtue of his position at court, Khān had been studying its archives and collecting materials on the topic of Kishangarhi paintings and the patron Sāvant Singh. When Dickinson visited, he was in the process of writing a dissertation in Hindi on Nāgarīdās at the University of Rajasthan, Jaipur.Footnote ⁴⁸ Later he penned another dissertation at the same University, but this time for the English Department, on “The Kishangarh School of Painting” (1986).Footnote ⁴⁹ The unassuming yet erudite Khān was close to the Kishangarh ruler Sumer Singh and, while a Muslim, served in many ways as the court’s spokesperson for its Vallabhan sectarian interpretation.

Dr. Faiyāz ‘Alī Khān also advised the scholar Kiśorīlāl Gupta, who edited the works of Nāgarīdās for the prestigious Nāgarī Pracāriṇī Sabhā of Varanasi, which came out in two volumes in 1965. Gupta reports he had been working on Nāgarīdās for over a decade, mainly based on the 1898 lithograph published by the Kishangarh court.Footnote ⁵⁰ He had remained unaware that its Vallabhan discourse had been challenged by a rival Nimbārkan claim on Nāgarīdās till his book was already in press; still he made a hurried trip to Kishangarh to seek clarity about Nāgarīdās’ sectarian allegiance. Unable to make it to the Nimbārkan monastery in nearby Salemabad, he visited the temple in the Kishangarh fort and met briefly with Dr. Khān, who confirmed the Vallabhan claim. Khān himself had already finished the manuscript of his own edition, but it was not yet typed up (Gupta Reference Gupta1965: 1.1–4). Consequently, Gupta did not modify the Vallabhan interpretation of his own edition, which was immediately challenged by the scholar Vrajvallabh Śaraṇ of Vrindaban’s Śrī Sarveśvar Press, who had been publishing articles propagating his own Nimbārkan sectarian viewpoint. Śaraṇ’s rival edition was published the following year, in 1966. In turn, Khān responded with a defense of the court’s Vallabhan stance in his own long delayed edition published from New Delhi by Kendrīya Hindī Nideśālay in 1974.Footnote ⁵¹ All this careful work on the Hindi literature has been neglected to a large extent in the art-historical publications in English, which ended up foregrounding Dickinson’s pioneering role and largely ignored the challenge to the court’s Vallabhan stance.Footnote ⁵²

Given his academic degree in English, it is not surprising that Khān made comparisons with Western art movements, just as Dickinson had done. He too characterized Indian art as essentially spiritual, foregrounding Rajput art’s “mysticism” (rahasyavād) based on bhakti literature (2015: 288–93). Khān described Nāgarīdās’ literary output and the matching paintings as part of an Indian version of Romanticism (2015: 310). He worked hard in his Hindi thesis to connect Nāgarīdās’ nature descriptions with the “subjectivity” of the Romantic movement, all the while taking care to distinguish them from the stylistic Rīti-like nature descriptions that he deemed more “classicist” (especially 2015: 334–5). Instead of Coomraswamy, he preferred to quote the British Principal of the Government School of Arts in Calcutta E.B. Havell. In a 1975 article, Khān brought up the Mona Lisa comparison that had been left unarticulated by Dickinson. Khān reminisced that back in 1943. Dickinson would send him queries regarding the real model for the Kishangarhi school’s distinctive facial features. One of the English teacher’s suggestions was that she might have been the dancing girl portrayed in a painting Moonlit Music Concert of Sardar Singh by Amarcand (see Pauwels Reference Pauwels2015: plate 2) but it became clear that the painting was too late to have been formative in the school’s development. Khān commented, “The idea of tracing the original of Radha, however, obsessed Professor Dickinson” (1975: 84). In July 1944, Dickinson sent Khān a questionnaire that was further probing as to the model for the Rādhā portrait, so Khān stated, “It is just possible that he [Dickinson] wished to establish a Mona Lisa parallel in the field of Rajput painting” (1975: 84).Footnote ⁵³

Dickinson did not live to publish his findings in book format. It was left to the new director of the Lalit Kalā Akademi (established in 1954), the prolific Parsi art connoisseur Karl Jamshed Khandalavala, to publish and elaborate on Dickinson’s draft. In his capacity as editor of the prestigious Lalit Kalā Series, he edited and expanded Dickinson’s work and published it as a posthumously, coauthored volume on Kishangarh paintings in 1959.Footnote ⁵⁴ Khandalavala went further than Dickinson had in the earlier paper of 1950. He reverently put the God Krishna down as Sāvant Singh’s first love but posited also a more mundane love in the prince’s life (Reference Dickinson and Khandalavala1959: 8).

Khandalavala was the one who identified the model for the Kishangarhi Rādhā as one of Sāvant Singh’s concubines, known as Bani-ṭhani. He believed she was the woman shown in a painting published in the 1959 volume as The Poet-Prince and Bani-Thani (plate 2, p. 23).Footnote ⁵⁵ Khandalavala actually attributed this identification to a suggestion by Khān, though the latter would later change his position.Footnote ⁵⁶ Connecting the identification of the lady in the painting as Banī-ṭhanī with Dickinson’s theory of the real-life model for the Kishangarh type, Khandalavala speculated, “If Nāgarīdās was the creator of this type, then who was the model who inspired him? Surely, it would not require much imagination to conclude that it was Bani Thani” (1959: 9). He was careful to qualify in a footnote, “The truth in all probability is that the Kishangarh type is an inspired idealization, based on a living model, skillfully employed to alter the existing female types already in vogue” (9–10). In the writeup of the “Lady with Veil” (published as plate 4 in the volume), adding to Dickinson’s comparison with the quatro-cento artists, Khandalavala brings in the comparison with the Mona Lisa, “What a triumph of profile treatment is here! In European art surely one would have to go to the achievements of the quatro-cento for any parallel to equal it, or to the great Leonardo himself and his famous Mona Lisa” (26).

Khandalavala’s “inspired idealization” represented a compromise between the eternal ideal type and the realistic influence of a particular individual woman. The characterization of the artistic intervention is tied up with the idealized interpretation of the love relation between the prince and his concubine, as described in rhapsodic words:

Dickinson’s aesthetic preoccupations come through underneath the somewhat apologetical reworking by Khandalavala. He sets straight Dickinson’s misunderstanding of “the way of pleasure” There is a defensiveness about the erotic aspect of spiritual love. Later in the book, Khandalavala asserted:

In this formulation, art and appreciation for real-life beauty are intrinsically intertwined. What is remarkable is how little room was made for Banī-ṭhanī’s agency, notwithstanding Khandalavala being aware of her own authorship. While she was acknowledged as an inspiration, it was the patron and the painter who “fashioned the divine Radha and everything beautiful in womanhood.” No sooner had the role of Banī-ṭhanī been identified that she was rendered passive, inspiring, yes, but ultimately merely because she embodied the image of beauty in womanhood.

To support his thesis, Khandalavala was keen on finding textual evidence. Again with the acknowledged help of Khān, he located a passage were Nāgarīdās would describe the traits of Rādhā in terms of Banī-ṭhanī’s distinctive features (Dickinson and Khandalavala Reference Dickinson and Khandalavala1959: 4). As he put it in the brochure accompanying the folio album of Kishangarh painting published in 1971, “In [Nāgarīdās’] poetic description of the milkmaiden who became the adored of the Blue God he has in truth described the loveliness of his own mistress” (1971: 1). Navina Haidar, in her insightful discussion of the theory about Banī-ṭhanī as model, has rightfully pointed out a flaw in this logic: Many poems by Nāgarīdās express formulaic traits of the Goddess, the earliest ones composed long before Banī-ṭhanī came to Kishangarh (2004: 128, fn. 16). This weakens Khandalavala’s argument. The references to curved brows over drooping lotus-like eyelids, and elegantly curling hair locks are indeed standard fare, not just in Nāgarīdās’ poetry, but Krishna bhakti more generally. Yet the poem Khandalavala cites adds a more specific reference, namely to the long nose, compared to a cypress. That comparison is not formulaic, yet it is one of the further characteristics of the Kishangarhi facial type. Khandalavala translates:

What was the original poem underlying Khandalavala’s translation? The cypress-like nose is referenced only twice in Nāgarīdās’ collected works. One is from Braj-sār:Footnote ⁵⁷

Like Khandalavala’s English version, this poem progresses from the eyes and brows compared to lilies and bees, to the distinctive nose likened to the cypress, ending with smiling flowerbed-like lips. The “dove” might be a free translation for khañjana or “wagtails” in the last line of the Hindi poem. Strikingly, in his translation, Khandalavala left out the first line of the poem with its reference to the veil, which actually would strengthen his argument as it fits the image of “Lady with Veil” so well.

This is a passage from Nāgarīdās’ technical Rīti style work from his 1742, Braj-sār. This textbook of poetics has definitions in Dohā and examples in Kavitta meter. It illustrates all types of nāyikās, but this one comes under the heading bīrī daināntara priya badana ika ṭaka rījha citaibo barnanã (description of unblinkingly staring enchantedly at the beloved’s face, right after providing betel). It seems to be an example of the hero, rather than the heroine. Still, the poem matches the “Lady with Veil” painting with its references to the veil at the outset, the curved brows and eyes, the lock of hair, and especially to her particular style of earring (also omitted in Khandalavala’s translation). In addition, the lotus imagery of the poem is made manifest in the painting through the lotuses in the heroine’s hand. Only the hero staring at her is absent, but possibly this was only the right wing of a diptych that included a portrait of Krishna on the left wing, similar to the later matching set in the National Museum (Mathur Reference Mathur2000: 114–17, figs. 30–40, acc. no. 63.813–4). It seems highly likely then that this is the poem Khandalavala had in mind. And one wonders whether it was inscribed on the back. Nāgarīdās’ Rīti work lends itself well to illustration; in fact, another of its verses (the preceding Kavitta 31) is actually inscribed on the back of a painting by Nihālcand on the theme of lovers exchanging betel, similarly showing the exaggerated profile associated with Banī-ṭhanī (Pauwels Reference Pauwels2015: 165–7). To clinch the argument, the poem actually refers to Rādhā; it is remarkable that Khandalavala missed that too in his translation.

The other reference to the cypress-like nose is from a poem in Nāgarīdās’ collection of Festival Poems, or Utsav-mālā. Like the Braj-sār poem, this poem also uses the extended metaphor of the garden for the woman’s body:

Crucial to understanding the initial intent of this poem is the ritual context of the Sāñjhī festival, the autumnal flower festival celebrated by young girls, “tweens,” in between child and young woman. There is a painting attributed to Nihālcand on the theme of Sañjhī that illustrates the features mentioned in the poem (Pauwels Reference Pauwels2015: 181–3). Like the previous poem, it fits Khandalavala’s English version, in its description of the face from forehead to lips, so this is another candidate for what he translated, though he does not mention the autumnal fesival.Footnote ⁶⁰ In any case, Khandalavala was definitely onto something as he singled out the unusual reference to the cypress-like nose to make his point of the facial characteristics of Kishangarhi Rādhā being celebrated in Nāgarīdās’ work.

Whatever one might think of its worth, Khandalavala’s argument that Bani-ṭhani’s distinctive profile was the model for the paintings convinced many in the art world. The conjecture took on a life of its own and became hardened into “common wisdom.” The review of the Lalit Kalā Akademī book by noted art historian Stella Kramrisch for Artibus Asiae, lyrically summarized, “Nagaridas saw Bani Thani with the transforming eye of mystic; in ecstasy and, feeling one with her, his features echo hers …. He sees Bani Thani, as the essence of herself, aetherialized as Rādhā” (1961: 69). The connection with the Kishangarhi facial type was reiterated in subsequent influential generalizing works, each one contributing a bit more certainty to the original postulation. The keepers of Islamic and Oriental Antiquities at British Museum respectively, Douglas Barret and Basil Gray,Footnote ⁶¹ went a step further than “inspiration,” as they wrote about the “small court” of Kishangarh that “produced by a minor miracle the most important school of eighteenth-century Rajasthan painting.” They asserted that “There seems little doubt that Savant Singh’s identification of the two passions of his life [Krishna and Bani-ṭhani] was responsible for the small but magnificent group of pictures painted at Kishangadh between the early years of his love affair and his abdication,” speculating that Nihālcand’s “new and very beautiful type for the divine lovers” was “perhaps based on the features of Bani-Thani herself; though idealized it has the feel of an individual experience” (1963: 155–9). Similarly, curator and art historian Philip Rawson in his popularizing volume on Indian art succinctly proclaimed, “There can be little doubt that the reality of the royal love-affair was a special stimulus to the artist to ‘realize’ in his work the divine prototype.” He is slightly more circumspect in the caption for the “Lady with Veil” image in declaring it, “perhaps an idealized image of Bani Thani” (Rawson Reference Rawson1972: 148; my italics). Thus, Khandalavala’s theories were enthusiastically promoted in subsequent secondary literature.

All this was much to the dismay of the Kishangarh court, whose spokesperson Khān himself authored a rebuttal in 1975 in the journal Roopa-Lekha. While credited by Khandalavala for the identification of the lady in the painting of “Nāgarīdās doing pūjā” as Banī-ṭhanī, he professed having promptly discouraged the theory of Banī-ṭhanī as model for Rādhā. These reservations were taken seriously, among others, in the influential book on Kishangarh Painting by Mohinder Singh and Doris Schreier Randhawa. They were circumspect about the Banī-ṭhanī theory:

Banī-ṭhanī is somewhat grudgingly acknowledged as an inspiration, but an unholy one, and her influence on the Rādhā portrait is strongly denied. Her authorship is reduced; she simply has “professed an interest in Hindi poetry.” The Randhawas also signaled some distance from the Mona Lisa comparison, stating “the portrait of Radha by Nihal Chand … represents the Rajput ideal of feminine beauty at its best. Those who delight in parallels with western art call her the Indian Mona Lisa” (Randhawa and Randhawa Reference Randhawa and Randhawa1980: 9). Thus, the Randhawas qualified their statements, likely influenced by the stance of the Kishangarh Court.Footnote ⁶²

The matter remained contested. About a decade later, the Jaipur-based painter Sumahendra tried to reconcile the court’s denial of the romance by acknowledging the relationship between the prince and the performer, while simultaneously rendering it more chaste: “we can leave aside the so-called fabricated stories but can not deny their nearness. They loved each other may be like intellectual friends or having sentiments other than lovers. Their love could have been respectable rather than illegal. Pious rather than erotic” (1995: 38). Many less careful secondary sources followed suit sanitizing the romance by turning Banī-ṭhanī into a queen. Only a few noted the court’s and Khān’s reservations to the model theory.

In conclusion, the concept of Banī-ṭhanī as the La Gioconda-like model for Radha–Kishangarh appears to be a hybrid construct, resulting from a complex engagement of colonial and Indian scholars, just like the idea of the “bhakti movement” itself (Hawley Reference Hawley2015). It came about through multiple dialogical interactions between Western and non-Western agents, through mediations, collaborations, confirmations, and contestations. Dickinson’s curiosity may have been a catalyst, but the influential director of Lalit Kalā Akademī, Khandalavala, was the first to formulate the theory that Banī-ṭhanī was the model for the Kishangarhi Rādhā. While Dickinson had compared it with among others Italian art, Khandalavala was the one who first brought up Mona Lisa. Faiyāz ‘Alī Khān may have been instrumental in setting these scholars on her track, but later did his best to refute their theories. Mostly, the relationship between the prince and his concubine continued being characterized as a love affair or passionate attachment, even as Khān issued a sharp denial of a romantic engagement as well as of Banī-ṭhanī’s role as model. While discomfort with the equation of erotic and spiritual love remained, the consensus stressed how Sāvant Singh and Nihālcand together designed the new Kishangarhi type, inspired by, but with minimal agency of Banī-ṭhanī. Throughout all of this, very little attention was paid to the actual poetry of the prince or his concubine. The interpreters of Khetanchi’s canvas discussed in Section 1.1 were following a well-trodden path.

1.5 From India’s Spiritual Face to Cyber-Orientalist Love Story

However one evaluates Dickinson’s contribution to the Indian Mona Lisa trope, he surely put Kishangarh on the art history map. He also bequeathed India a more material legacy; this took the form of a donation from his own collection. The new National Museum’s permanent collection was based on what had been displayed during the grand “Masterpieces of Indian Art” exhibition held in 1948 in Government House (now Rashtrapati Bhavan) in Delhi to celebrate independence. This in turn had grown out of the assemblage of artefacts on their way back to diverse Indian institutions and collections from an exhibition at Burlington House in London that had been organized by the Royal Academy of Arts (November 1947–February 1948; Guha-Thakurta Reference Guha-Thakurta2004: 176–7). The London show had included artefacts on loan from British museam and private collectors that did not travel to India (Codrington et al. Reference Codrington, Irwin and Gray1949: 35–44). To fill the gaps, new loans were acquired for the Delhi occasion, among others a total of 241 items from Dickinson’s personal collection, second only to the Treasurywala collection (and later the Sarola collection; see Banerjee Reference Banerjee1992: 29). A substantial part of Dickinson’s loans was made up of Gandharan sculptures from the North-West,Footnote ⁶³ though his main contribution to the museum collection consisted of paintings, mostly Krishna devotional scenes. While he contributed samples from practically all Rajput schools, the catalogue did not specify, simply classifying as “Rajasthani.” Only the first of the Rajasthani paintings in the catalogue, an illustration of a manuscript of Keśavdās’ Rasik-priyā, is explicitly said to have been acquired from Kishangarh (Agrawala Reference Agrawala1948: 45, no. 386). Further included were illustrations of Bihārī’s famous Satsaī, as well as of the Rajasthani epic Ḍholā-Mārū, and a Rāga-mālā series. One painting is marked as Vāsaka-sajjā, but the dimensions do not match the Kishangarhi “Lady with Veil” (Agrawala Reference Agrawala1948: 48, no. 414). These generous contributions by Dickinson bolstered the new nation’s pride, complicating further his perceived role as the promotor of orientalist tropes about Kishangarhi painting.

The creation of the National Museum itself illustrates the complexity of even just material interchange between colonial and nationalist agents. As mentioned, this proud showcase of Indian art had grown out of the 1948 exhibition held in Delhi, which in turn had been conceived at the opportunity to assemble items on their way back to diverse Indian institutions and collections from the London exhibition of 1947–8. According to the Indologist V.S. Agrawala, who wrote the catalogue, the Indian version was conceived as “an instrument of prime importance for vitalizing the future cultural programme of the country” (1948: v). Elsewhere he went so far as to compare with a Vedic sacrifice, “A good Exhibition may be compared according to ancient Indian ideology to a public Yajna in which work of real value for the intellectual and aesthetic regeneration of society can be achieved” (Agrawala Reference Agrawala1949: 27). Thus, discursively, continuity of an indigenous tradition was foregrounded. The Indian curators had worked hard so the “aestheticized object of Indian art had emerged as a main field for the self-representation of the nation” (Guha-Thakurta Reference Guha-Thakurta2004: 175). Again, in the words of the catalogue, “Indian art richly documenting the past culture of India has a unique position … as revealing the mind of the Indian people” (Agrawala Reference Agrawala1948: v). One of the objectives in the new nation-building mode had been to come as close as possible to complete representation. To that goal contributions in addition to those on display in London had been solicited, which was where Dickinson’s loans had come in.

The success of this endeavor led first to the extension of the public display through 1949. The next step was the inauguration of the National Museum on August 15, 1949, which had been achieved through negotiations that succeeded in converting many loans to the exhibit into donations to the permanent collection, before everything was moved to the location at the crossing of Rajpath and Janpath in 1960. This move from loan to donation meant “asserting the priority of the new national claims of the capital and its central command over all existing holdings” (Guha-Thakurta Reference Guha-Thakurta2004: 201). In the process, the central administration “volunteered” the local museum directors who had been involved in exhibition organization to serve on the Art Purchase Commission, which in turn was to put pressure on other local museum administrators and collectors to secure objects for the permanent collection (Phillips Reference Phillips2006: 59–60).The National Museum’s history then, does not just embody the tension of nationalism with imperial discourses, but also one between the centralizing nation building project and provincial pulls. Assertion of nationalism came at the expense of local interests and regional pride. This tension is particularly felt in the case of the Kishangarhi paintings.

It is not entirely clear how the National Museum’s Kishangarh collection was built.Footnote ⁶⁴ The catalogue of the 1947–8 London exhibition makes reference to Kishangarhi painting in the introduction, but it only lists explicitly one on display, a temple background hanging (pichvāī) on loan from the State Museum of Baroda.Footnote ⁶⁵ As we have seen, the Delhi exhibition catalogue is very vague, designating most paintings generically as “Rajasthani” and identifies only one as obtained from Kishangarh. In Reference Mathur2000, then curator Dr. Vijay Mathur published a magnificent selection of the collection in his book Marvels of Kishangarh Painting. The accession numbers provide a clue as to the dates of acquisition (Kavita Singh, personal communication, August 2020). Only one painting has an accession number prefaced with 48, indicating it was acquired in 1948: the famous painting of Svāmī Haridās with Tānsen and the emperor Akbar.Footnote ⁶⁶ In his 1950 Mārg article, Dickinson discussed several paintings that were exhibited at the National Museum in Delhi, but it is not clear when (Khandalavala 1959: 4). Curator of paintings Adris Banerji indicated that twenty-two of those paintings “were brought to the National Museum, Delhi, by Dr. N. P. Chakravarty” (1954: 13). Chakravarty, as Director of the Archeological Survey of India, had been a major player in the “Masterpieces of Indian Art” exhibition of 1948: he was listed in the catalogue as secretary and member of the Executive committee (Agrawala Reference Agrawala1948: vi–vii).Footnote ⁶⁷ Upon retirement in 1950, he still served in the capacity of advisor to the Department of Archeology of the Government of India. Possibly, he was instrumental in the incorporation of Dickinson’s collection into the National Museum. In addition, the General Accession Register mentions that V.S. Agrawala, author of the Delhi exhibition catalogue, helped secure the Dickinson collection’s “‘waslis,’ manuscripts, ‘stuccos’ and sculptures” (Phillips Reference Phillips2006: 62). One suspects that Karl Khandalavala too played an important role in acquiring Kishangarh paintings through his presence at the deliberations of the Art Purchase Committee (Phillips Reference Phillips2006).

In 1963, a generous donation of no less than forty-six items, “masterly specimens of the Kishangarhi school,” was made by Sumer Singh, Maharaja of Kishangarh, whom we have already encountered as patron of Dr. Faiyāz ‘Alī Khān (Banerjee Reference Banerjee1992: 29; see also Mathur Reference Mathur2000: vii). This included some of the most famous and spectacular paintings that will be discussed in the following chapters, such as Boat of Love, Bani Thani, Sawant Singh and Bani Thani in a Mango Grove, and Krishna Holding Radha’s Odhani, besides Tambul Seva, and many others.Footnote ⁶⁸ Thanks to this donation, the National Museum has its own matching archetypical Kishangarhi Radha and Krishna set, which represents a turn-of-the-century 1800 specimen.Footnote ⁶⁹ The original “Lady with Veil” though is not at the National Museum.Footnote ⁷⁰ Yet that did not impede it from becoming “the spiritual face of the nation,” instantly recognizable throughout the country.

Two decisive steps in that direction again involved the agency of Khandalavala. First was the issuing of what was by now known as Portrait of Radha as a separate mounted color plate (Khandalavala Reference Khandalavala1971: 1, n. 3), at the occasion of the publication by Lalit Kalā Akademī of a portfolio with Kishangarh paintings (Lalit Kalā Series no. 12, 1971). That may well have played a role similar to what the gravure of Luigi Kalamatta did for the recognition of the Italian Mona Lisa in Paris in 1837 (Sassoon Reference Sassoon2001: 92). Khandalavala commented dryly on the raft of imitations that followed his publications:

All this contributed to the ubiquity of the “Lady with Veil” and its instant recognition.

The next phase in the portrait’s apotheosis was literally the nationalist stamp of approval. Out of a pool of more than 5,000 miniature paintings, the “Lady with Veil” was one of four selected to be the basis for a new stamp issued in 1973 by the India Post and Telegraphs Department. This was part of India’s first multicolor series printed from the Security Press in Nasik on the newly installed multicolor photogravure machine. The commemorative series of “Indian Miniature Paintings” was issued on May 5th of that year and was promoted vigorously, even by sales abroad at the Munich International Exhibition.Footnote ⁷¹ “The first-day sales alone brought in a revenue of Rs. 3 lakh,” according to the Annual Report of the Indian Posts and Telegraphs Department for 1973–4. The editorial of the July 1973 issue of the Philatelic Journal mentioned the extensive press coverage of the new series in philatelic journals in the United Kingdom, as well as in the Indian Press, where the Kishangarh stamp was singled out in particular (Mehta Reference Mehta1973a: 1). Clearly, this was a hit.

As is the case elsewhere with issuing national prints, the selection of the postal stamp miniature series was “more ad hoc and serendipitous than presumptions about their centrality to nation-building might suggest” (Penrose Reference Penrose2011, speaking about banknotes, but applicable here too). Practical considerations prevailed: All four miniatures were selected based on boldness of color scheme and adaptability after reduction to fit the stamp format (Mehta Reference Mehta1973b: 183–4). Still, the discursive context of articulating India’s art history as an elaboration of its spiritual aesthetic essence (Guha-Thakurta Reference Guha-Thakurta2004: 184–5) may well have played a role in identifying the Rādhā image as representative of Indian art. Earlier, in 1952, another Kishangarh painting had been adapted to produce a postage stamp of the Rajasthani poetess Mīrā. This was part of the series “Saints-Poets and Poets,” which also included Kabīr, Tulsīdās, and Sūrdās, showing pride in India’s “mystic” Hindu tradition. However, one should not too hastily paint everything with the broad brush of Hindu chauvinism. The series also included the Indo-Muslim poet Ghālib (as well as Tagore). In connection with exhibitions of Indian art, Cambridge-affiliated art historian Devika Singh has pointed out that besides Hindu chauvinists, the voices of progressive educationalists also figured importantly (2013: 1063–4). In the case of the choice of Radha–Kishangarh for the stamp, one suspects the influential Karl Khandalavala once again played a role, since he chaired the selection committee for the stamps.

Out of the 1973 miniature stamp series, the Kishangarh one was the only one with an inscription that geographically located it within one of the Rajput schools. This confirms the prestige the Kishangarh school had acquired by this time and its pride of place in the nation’s national art heritage. The writeup by Dhiru Bhai Mehta for the Indian Philatelic Journal specified that “the painting depicts perfect expression of the Kishengarh ideal of feminine charm and loveliness, added by the ornaments with which the Radha is adorned.” Voicing the more progressive stance, Mehta recognized the influence of Mughal painting, but stressed that the painter Nihālcand “imbued late Mughal Painting with serenity of rhythm and idyllic charm” (1973b: 184). We recognize some of the orientalist art-historical discourse appropriated for nationalist purposes discussed above. Besides the imperialist–nationalist antagonism, one may detect again tension between the centralizing discourse of the nation and assertion of diversity and local cultural pride. By this time, the historical context had evolved, and regional royal houses had become incorporated in the Indian Union. This tension is reminiscent of that at work earlier on in the establishment of the collection of the National Museum which had involved pressurizing lenders for their loans to the independence celebration exhibition to be made permanent (see Phillips Reference Phillips2006: 67–70).

The insistence on the regional label for the Rādhā stamp may well have come from the Royal House of Kishangarh, whose permission was sought for the publication of the stamp.Footnote ⁷² Its influence is evident in the inscription on the stamp simply as Radha–Kishangarh, which avoids reference to the purported model, the concubine Banī-ṭhanī, which as we have seen the court was uncomfortable with. The association with the latter though, lives on in the popular identification of the eighteenth-century painting as well as of the stamp, as we know from the discourses around Khetanchi’s oil canvas. In that light, it does not seem coincidental that the strong rejection of the Banī-ṭhanī as model trope would be articulated and published just a couple of years later by Faiyāz ‘Alī Khān in 1975. The issuing of the stamp causing a renewed interest in Banī-ṭhanī’s role as model for Kishangarhi Rādhā’s features, may well have prompted that sharp rebuttal.

Whatever its name, the image of Radha–Kishangarh is widely recognized as part of India’s national heritage; it figures, for instance, in the Masters Institute for Civil Services IAS exam multiple choice questions.Footnote ⁷³ Still, it became never as popular in currency iconography as the image of Bhārat Mātā or “Mother India,” the Goddess projected on a map, that became ubiquitous in stationary, calendar art, posters and the like (Ramaswamy Reference Ramaswamy2010: 15–7).Footnote ⁷⁴ This may be a more interesting comparison to make than with Mona Lisa, since both images carry a spiritual message that was appropriated for the new nation.Footnote ⁷⁵ Whereas the cartographically inflected Bhārat Mātā could be seen as representing “hot nationalism,” the Kishangarh image perhaps partakes in a local more “banal nationalism,” the material and cultural construction of routine regional identity (terms from Billig Reference Billig1995). While Radha–Kishangarh never acquired a logo status, it functions as an everyday element of visual culture, reifying the state and perpetuating the imagined local and regional community through art. The image is popularly used in schools in Kishangarh as a model for art work for the students and subsequently the best imitations are gifted to visiting dignitaries.Footnote ⁷⁶ In that sense, it can be seen to reproduce the social relations of local “stateness” (see Penrose Reference Penrose2011, in connection with banknotes).

The Radha–Kishangarh stamp was published amidst strong tensions between the royal courts and the Indian nation. The aforementioned Mahārāja Sumer Singh of Kishangarh had been active politically after the merger of Kishangarh into the Indian Union in 1948 and became representative of Kishangarh in the Rajasthan State Assembly (Member of the Legislative Assembly from 1967).Footnote ⁷⁷ After he passed away tragically in a car crash, the young Brajraj Singh was the last prince officially recognized by the president of the Indian Republic upon his accession in February 1971. Later that year, the derecognition of existing titles of royalty and the abolishment of the Princely Privy Purses was enacted as per the 26th amendment to the Indian constitution (Taft Reference Taft, Singh and Joshi2003: 128–9). The historical context was one of local (now ex-)rulers contesting vigorously in the Lok Sabhā elections of 1971.

The dire financial consequences of the abolishment of the Privy Purses for the royal houses also brought a flood of artefacts from royal collections on the art market, including miniatures for which the issuing of the stamp series of 1973 would bring publicity. They also accelerated the speed by which royal palaces were converted into “Heritage Hotels,” following the earlier example of Jaipur’s Rambagh Palace, which originally accommodated royal guests, but was turned into a hotel as early as 1958 (Taft Reference Taft, Singh and Joshi2003: 128). Similarly, Phoolmahal, the garden palace at the foot of the Kishangarh fort, built up and expanded over a period from 1870–1907, was turned into a hotel, but not until later.Footnote ⁷⁸ The tourist industry more generally led to the “packaging” of cultural heritage with concomitant emergence of visual images as icons, emphasizing the “romance and glories of the princely states” (Ramusack Reference Ramusack, Asher and Metcalf1994: 236, 242). Our Indian Mona Lisa would fall into that category.

Under the influence of what some have called “cyber-orientalism” (Henderson Reference Henderson, Henderson and Weisgrau2007: 61–81), Rajasthani images have become couched in the language of the “timeless, spiritual, colourful and exotic” (Henderson and Weisgrau Reference Henderson, Weisgrau, Henderson and Weisgrau2007: 225). This brings to mind the 1935 canvas of Rajput ladies dancing at the Gangaur festival in Udaipur under the camera lens of Western observers. In the words of Molly Aitken, such reveals the oriental trope of “quaintness as a cropped, postcard-thin projection” (Aitken Reference Aitken2017: 52–4). Banī-ṭhanī has undergone this fate, reduced to a two-dimensional cardboard cut-out, infinitely reproduceable, her quaint and exotic beauty fit for global tourist consumption, her romance with the prince now a Cinderella story ripe for an Indian Disneyland. And yet, the fairy tale remains contested, even by the very royals whose erstwhile palaces are now open for tour groups attracted by the myth. The purpose of the rest of this book is to find out who was the living, breathing woman behind the cyber-orientalist icon, and make audible what has been elided in the simplified promotional narrative.

***

The “Indian Mona Lisa” trope, as embodied in the 2006 oil canvas Bani Thani by Gopal Swami Khetanchi reproduced at the beginning of the chapter (Figure 1.1), turned out to be a subversion of the orientalist gaze. In referencing the eighteenth-century “Lady with Veil” from the Kishangarh school, it actually showed up the limits of eurocentrism, revealing western ignorance of other artistic traditions. The reception of Khetanchi’s painting’s in the popular press elicited discourses that challenged western ideals of womanhood, beauty, and art.

To be sure, the original portrait “Lady with Veil” in its eighteenth-century context had some mimetic elements. It was part of contemporaneous serial production of Mughal and Rajput portraits of women, which sometimes had links with European models, especially the bust portraits. Mostly the ladies portrayed had idealized features marked by regional styles, though they were at times based on individual traits of historical women, including some who were performers. This phenomenon in painting can be understood within a poetic imagery in both Hindi and Hindavi literature, where observed, ideal, and divine realities are telescoped onto one another.

Discourses around the “Lady with Veil” were popularized in the 1940s by Lahore-based British teacher Eric Dickinson. He described his “discovery” of Kishangarh painting as conforming to the art-historical discourse at the time, which was dominated by A.K. Coomaraswamy. Dickinson aptly dubbed “Lady with Veil” as Portrait of Radha, a title that has stuck. He deemed it the Kishangarhi tradition’s archetype, face of spiritual Rajput Art, yet he also postulated a historical intervention of painter Nihālcand and sponsor Sāvant Singh, to rival the neighboring schools.

While Dickinson recognized the poetic inspiration behind the canvases, he ignored the research from the Hindi world on the topic, though his informant and the Kishangarh Court’s spokesperson Dr. Faiyāz ‘Alī Khān was part of it. The unofficial title of the painting, “Banī-ṭhanī,” is courtesy of the director of the newly established national Lalit Kalā Academy, Karl Khandalavala, who identified its real-life model with Banī-ṭhanī, concubine of Prince Sāvant Singh, the patron of the paintings. He was also the first to compare with Lisa Gherardini, the model for the Mona Lisa. The portrait came to stand for the ideal spiritual essence of the new Indian nation’s Rajput Art.

The devotional inspiration of the Kishangarhi school was foregrounded by all, but the theory that Banī-ṭhanī’s features would underlie those of the archetype and the Goddess Rādhā caused some unease, possibly because she was a concubine and a slave. The Kishangarh Court opposed the theory and also defended the Kishangarhi school’s specific Vallabhan sectarian origin opposing the rival Nimbārkan challenge. While this resulted in competing editions of Sāvant Singh’s work (under the pen name Nāgarīdās), such subtleties did not rise beyond Hindi academic circles to make it onto the art-historical radar.

Meanwhile, the Kishangarhi school gained prominence thanks to Khandalavala’s publications. As its collection in the National Museum grew, the development climaxed when the school gained the “stamp” of nationalist approval in 1973. The postage stamp featuring “Lady with Veil” was issued among tensions between centralizing and regional political forces, in the wake of the derecognition of titles of regional royalty and the abolition of the Privy Purses in 1971. The name of the stamp, Radha–Kishangarh, affirmed the court’s insistence on its regional identity and its resistance to the theory that Banī-ṭhanī was the model for the portrait of the Goddess. Still, even though vigorously contested, the story of Banī-ṭhanī’s “romance” with Sāvant Singh gained steady ground and became popularized for the tourism sector under the “Indian Mona Lisa” banner. In the course of these processes, the erotic relationship with the prince was still felt to be problematic and sublimated in popular versions of the story. Eventually, Portrait of Radha, conflated with “Bani Thani,” developed into a cyber-orientalist icon. Along the way, what remained neglected was Banī-ṭhanī’s authorial voice as Rasikbihārī.

The following chapters look beyond the Indian Mona Lisa trope, setting out to recover what can be known of the real Banī-ṭhanī, alias the poetess Rasikbihārī. Starting with her origins as a slave, they restore the rich texture of a woman’s life, of an author-couple’s shared creative delight, but most of all, her voice and its legacy as it has survived over the centuries, even if detached from her image. By the end of the book, the full relevance of her being featured on the cover of Nouvelles de l’Inde’s special issue on Indian literature will become clear.