A shakti peetha is a place where a piece of Sati is held to have fallen. The word joins shakti, power, to peetha, seat, and these are counted as the seats where the goddess is present in the land itself rather than in an image someone installed. The first thing a reader wants to know is how many there are, and almost no page answers it honestly: eighteen, forty-two, fifty-one, fifty-two, seventy-two and a hundred and eight are all in circulation. They are not competing errors. They are different enumerations, compiled in different centuries for different purposes, and the shorter ones are mostly earlier stages of the longer ones rather than rivals to them. This page gives each count the text it comes from, and says plainly which of those attributions hold and which do not.

Counted by these texts

• Aṣṭādaśa Śakti Pīṭha Stotra (traditionally attributed to Śaṅkarācārya) — 18 — The enumeration this corpus follows, because it is the only one of the six that names every member in a fixed order a reader can check against the verse — it opens lankayam shankari devi and closes kashmiretu sarasvati. Two cautions. The attribution to Shankaracharya is traditional: several hundred works are ascribed to him and scholarship accepts a small fraction of them, and hymns are where that ascription concentrates, so this corpus writes attributed to rather than composed by. And the Kularnava Tantra is separately credited with an eighteen; no source consulted here shows the two eighteens to be the same list, so they are not treated as one.
• Kubjikā Tantra — 42 — The stratum the fifty-one grew out of, which is why it belongs on the panel even though few readers will have met the number. Sircar's account of the Pithanirnaya is explicit that its list of fifty-one was built by enlarging a list of forty-two.
• Pīṭhanirṇaya, also transmitted as the Mahāpīṭhanirūpaṇa — 51 — The count almost everyone means by the Shakti Peethas, and the only one of the large enumerations with a critical edition behind it — Sircar based his 1948 study on this text. Two things about it are usually left out. It is late: Sircar puts the fifty-one at the seventeenth or eighteenth century, so the most famous list is also among the youngest. And it is derived: the forty-two above, enlarged by the ten Mahavidyas. Popular sources also credit the fifty-one to the Devi Purana, and treat this treatise as a section of the Tantra Chudamani rather than a work of its own; the number travels more reliably than the title it is hung on.
• Tantra Chūḍāmaṇi — 52 — Fifty-two is the fifty-one with one more seat, and the difference is normally explained as a later tradition splitting a site or adding one, chiefly in Bengal. Treat this pairing as the received one rather than the demonstrated one: the same title is elsewhere credited with fifty-one, and no source consulted here settles which reading its manuscripts support.
• Devī Gītā — a circulating attribution this corpus could NOT corroborate — 72 — Recorded here so a reader who meets the number has something to weigh it against. Seventy-two travels almost entirely inside a single repeated formulation that assigns fifty-one to the Devi Purana, a hundred and eight to the Devi Bhagavata, fifty-two to the Tantra Chudamani and seventy-two to the Devi Gita. It appears in no scholarly treatment consulted here, and the attribution contradicts itself: the Devi Gita is not a separate work but a section of the Devi Bhagavata Purana, sitting in the chapters immediately after the pitha list that the same sentence credits with a hundred and eight. The likeliest reading is that one entry was split in copying. Until a verse is located, this is a claim about what circulates, not a citation.
• Devī Bhāgavata Purāṇa, seventh skandha, ch. 30 — 108 — The Puranic enumeration, and the one most clearly shaped by its own number: a hundred and eight is how this literature counts sacred things, so the chapter reads as a hymn of the goddess's places more than as a pilgrim's itinerary. The chapter and its list are verifiable; that the list holds exactly a hundred and eight entries is asserted by the sources consulted here rather than counted by this corpus, and is recorded on that footing.

Places written so far

• Shankari — The verse opens with her, and the shrine it opens with was pulled down in 1622 and its stone built into a fort that still stands there.
• Kamakshi — Kanchipuram is a seat of the goddess in two different senses, and the two have nothing to do with each other.
• Shrinkhala — Third of the eighteen, and a place where a pilgrim arrives to find no shrine at all — only protected monuments.
• Chamundeshwari — The hill fuses two goddesses the text keeps separate — the one who killed Chanda and Munda, and the one who killed Mahisha.
• Jogulamba — Her shrine was destroyed in the fourteenth century and her image spent six hundred years sheltering in the temple next door.
• Mallikarjuna — Two of the tradition's great circuits stop on the same hill — Shiva as Mallikarjuna, Parvati as Bhramaramba, a few steps apart.
• Mahalakshmi — A goddess with Lakshmi's name, a lion at her feet and a Shiva-linga above her crown — two traditions in one image.
• Ekaveerika — Two shrines at one hill are both handed to pilgrims as this peetha, and only the smaller one carries the verse's name.
• Mahakali of Ujjain — Ujjain gives the verse's ninth peetha two addresses, and the busier of the two is not the one that carries the name.
• Puruhutika — The tenth peetha stands in a town the tradition calls the foot of a demon whose head is Gaya and whose navel is Jajpur.
• Girija Devi — The eleventh peetha is named for a place whose location scholars still argue about — and it is not the shrine that is in doubt.
• Manikyamba — The twelfth peetha stands in a town whose name means Daksha's garden — the sacrifice the whole story starts from.
• Kamarupa — The one seat in the eighteen with no image in it — the object of worship is a cleft in the rock, kept wet by a spring.
• Madhaveshwari — The fourteenth peetha stands at the confluence of three rivers, and its sanctum holds no image at all — only a covered cradle.
• Vaishnavi — The fifteenth peetha has no image and no lamp: what is worshipped is fire coming out of the rock by itself.
• Mangala Gauri — The sixteenth peetha stands above a town the whole subcontinent visits for its dead — and the goddess is not why they come.
• Vishalakshi — The seventeenth peetha is a small temple in a lane above the ghats, in a city that gives a small shrine a great deal to compete with.
• Sharada — The hymn closes in Kashmir, at a shrine that is a ruin — and that gave a script and a region their name.

Questions to explore

• Every list I find gives a different number of Shakti Peethas. Which one should I actually follow, and does it matter which?
• The fifty-one are matched to the fifty-one letters of the Sanskrit alphabet. Did the letters shape the list, or the list the letters?
• Three of the eighteen stand where a jyotirlinga stands. Do the two circuits meet anywhere else, and is that a coincidence of geography or of history?