Another derivation is given (and adopted by so learned an etymologist as H. Wedgwood) from the Portuguese pagão, 'a pagan.' It is possible that this word may have helped to facilitate the Portuguese adoption of pagoda; it is not possible that it should have given rise to the word. A third theory makes pagoda a transposition of [dagoba]. The latter is a genuine word, used in Ceylon, but known in Continental India, since the extinction of Buddhism, only in the most rare and exceptional way.
A fourth suggestion connects it with the Skt. bhagavat, 'holy, divine,' or Bhagavatī, applied to Durgā and other goddesses; and a fifth makes it a corruption of the Pers. but-kadah, 'idol-temple'; a derivation given below by Ovington. There can be little doubt that the origin really lies between these two.
The two contributors to this book are somewhat divided on this subject:—
(1) Against the derivation from bhagavat, 'holy,' or the Mahr. form bhagavant, is the objection that the word pagode from the earliest date has the final e, which was necessarily pronounced. Nor is bhagavant a name for a temple in any language of India. On the other hand but-kadah is a phrase which the Portuguese would constantly hear from the Mahommedans with whom they chiefly had to deal on their first arrival in India. This is the view confidently asserted by Reinaud (Mémoires sur l'Inde, 90), and is the etymology given by Littré.
As regards the coins, it has been supposed, naturally enough, that they were called pagoda, because of the figure of a temple which some of them bear; and which indeed was borne by the pagodas of the Madras Mint, as may be seen in Thomas's Prinsep, pl. xlv. But in fact coins with this impress were first struck at Ikkeri at a date after the word pagode was already in use among the Portuguese. However, nearly all bore on one side a rude representation of a Hindu deity (see e.g. Kṛishṇarāja's pagoda, c. 1520), and sometimes two such images. Some of these figures are specified by Prinsep (Useful Tables, p. 41), and Varthema speaks of them: "These pardai ... have two devils stamped upon one side of them, and certain letters on the other" (115-116). Here the name may have been appropriately taken from bhagavat (A. B.).
On the other hand, it may be urged that the resemblance between but-kadah and pagode is hardly close enough, and that the derivation from but-kadah does not easily account for all the uses of the word. Indeed, it seems admitted in the preceding paragraph that bhagavat may have had to do with the origin of the word in one of its meanings.
Now it is not possible that the word in all its applications may have had its origin from bhagavat, or some current modification of that word? We see from Marco Polo that such a term was currently known to foreign visitors of S. India in his day—a term almost identical in sound with pagoda, and bearing in his statement a religious application, though not to a temple.[[197]] We thus have four separate applications of the word pacauta, or pagoda, picked up by foreigners on the shores of India from the 13th century downwards, viz. to a Hindu ejaculatory formula, to a place of Hindu worship, to a Hindu idol, to a Hindu coin with idols represented on it. Is it not possible that all are to be traced to bhagavat, 'sacred,' or to Bhagavat and Bhagavatī, used as names of divinities—of Buddha in Buddhist times or places, of Kṛishṇa and Durgā in Brahminical times and places? (uses which are fact). How common was the use of Bhagavatī as the name of an object of worship in Malabar, may be seen from an example. Turning to Wilson's work on the Mackenzie MSS., we find in the list of local MS. tracts belonging to Malabar, the repeated occurrence of Bhagavati in this way. Thus in this section of the book we have at p. xcvi. (vol. ii.) note of an account "of a temple of Bhagavati"; at p. ciii. "Temple of Mannadi Bhagavati goddess ..."; at p. civ. "Temple of Mangombu Bhagavati ..."; "Temple of Paddeparkave Bhagavati ..."; "Temple of the goddess Pannáyennar Kave Bhagavati ..."; "Temple of the goddess Patáli Bhagavati ..."; "Temple of Bhagavati ..."; p. cvii., "Account of the goddess Bhagavati at, &c. ..."; p. cviii., "Acc. of the goddess Yalanga Bhagavati," "Acc. of the goddess Vallur Bhagavati." The term Bhagavati seems thus to have been very commonly attached to objects of worship in Malabar temples (see also Fra Paolino, p. 79 and p. 57, quoted under c. below). And it is very interesting to observe that, in a paper on "Coorg Superstitions," Mr. Kittel notices parenthetically that Bhadrā Kālī (i.e. Durgā) is "also called Pogŏdi, Pavodi, a tadbhava of Bagavati" (Ind. Antiq. ii. 170)—an incidental remark that seems to bring us very near the possible origin of pagode. It is most probable that some form like pogodi or pagode was current in the mouths of foreign visitors before the arrival of the Portuguese; but if the word was of Portuguese origin there may easily have been some confusion in their ears between Bagavati and but-kadah which shaped the new word. It is no sufficient objection to say that bhagavati is not a term applied by the natives to a temple; the question is rather what misunderstanding and mispronunciation by foreigners of a native term may probably have given rise to the term?—(H. Y.)
Since the above was written, Sir Walter Elliot has kindly furnished a note, of which the following is an extract:—
"I took some pains to get at the origin of the word when at Madras, and the conclusion I came to was that it arose from the term used generally for the object of their worship, viz., Bhagavat, 'god'; bhagavati, 'goddess.'
"Thus, the Hindu temple with its lofty gopuram or propylon at once attracts attention, and a stranger enquiring what it was, would be told, 'the house or place of Bhagavat.' The village divinity throughout the south is always a form of Durga, or, as she is commonly called, simply 'Devi' (or Bhagavati, 'the goddess').... In like manner a figure of Durga is found on most of the gold Huns (i.e. pagoda coins) current in the Dakhan, and a foreigner inquiring what such a coin was, or rather what was the form stamped upon it, would be told it was 'the goddess,' i.e., it was 'Bhagavati.'"