Through intertextual analysis of ancient stone and metal images as well as written texts, this article explores the origins and identities of Tārā and Cundā, two Buddhist goddesses that are widely worshipped in the Indic and Tibetan Mahāyāna and Vajrayāṇa traditions. By charting the goddesses’ iconographic characteristics, their quality, and their relationships with Avalokiteśvara, it argues that the prototype of Tārā is Yaśodharā-to-be, and that Cundā is based on Sujata, the girl who gave the Buddha-to-be milk rice. The first part of the analysis shows clear coherence throughout ancient visual narratives, namely images featuring the girl in the Dīpaṃkara Buddha story, images featuring Avalokiteśvara and Tārā, and independent images of Tārā. The second part demonstrates clear coherence in the evolution of the visual narratives, which feature Sujata holding a vessel to see the Buddha-to-be and independent Cundā images. The article also shows that the visual narratives have intrinsic fidelity.
