Constellations and Planets in Buddhist Astrology

If you observe the sun rising in the east and setting in the west over the course of a year, it would appear to follow along a certain band of the sky. This is called the ecliptic, which is defined as the “great circle representing the apparent annual path of the sun.” The illustration here should prove helpful. The moon likewise follows along this band of sky. Every night it will appear in a different section of it before returning to its original place, a cycle which takes around 28 days. These different sections can be called lunar stations in that the moon temporarily "resides" in each one before migrating to the next. The circle or band of sky can thus be divided into 27 or 28 parts. One could also divide it into 12 parts (the zodiac), which was the practice of the Babylonians by the year 500 BCE. This was taken up by the Greeks and by the second century BCE what we would recognize as western astrology emerged with the horoscope, which was based on the zodiac. The ancient Indian model of astrology, which was likewise influenced by Babylonian sources, initially based itself however on the lunar stations.

In ancient India the lunar stations were and still are called nakṣatra. In early Vedic times nakṣatra originally just meant star and later came to refer to constellations constituting lunar stations along the ecliptic. A complete list of 28 nakṣatra-s is first provided in the Atharva Veda. This was probably an indigenous development as there is no evidence the Indians received this model from Babylon. However, the Chinese by the fifth century BCE had developed their own model of 28 lunar stations (the xiu 宿), which led many in the last century to suspect both must have a common origin, though there is no substantial evidence to prove this. In 1919, the Assyriologist Carl Bezold thought he discerned Babylonian influence in Chinese texts, a thesis which was accepted by figures like Joseph Needham and Edward Schafer. David Pankenier, however, has recently (2014) has refuted this (see here).

Comparisons in any case between ancient Chinese and Indian models is problematic because as David Pingree has shown (see here), the Vedic literature does not provide accurate coordinates and later Indian star catalogs are inconsistent with respect to the yogatārā-s (the principal star of a nakṣatra). The origins and reasoning behind the Sanskrit names are also largely unclear aside from those named after their shape (Mṛgaśīrṣa = "deer's head"), which has led to suggestions they are not of Indian origin. Translators in China in any case were able to use the existing Chinese terms when translating Indian texts since both used 28 stations, though it was recognized only a few of the stars mutually correspond. The list is as follows:

01. 昴宿 Kṛttikā
02. 畢宿 Rohiṇī
03. 觜宿 Mṛgaśīrṣa
04. 參宿 Ārdrā
05. 井宿 Punarvasū
06. 鬼宿 Puṣya
07. 柳宿 Aślesā
08. 星宿 Maghā
09. 張宿 Pūrvaphālgunī
10. 翼宿 Uttaraphālgunī
11. 軫宿 Hasta
12. 角宿 Citrā
13. 亢宿 Svāti
14. 氐宿 Viśākhā
15. 房宿 Anurādhā
16. 心宿 Jyeṣṭha
17. 尾宿 Mūla
18. 箕宿 Pūrvāṣāḍhā
19. 斗宿 Uttarāṣāḍhā
20. 牛宿Abhijit
21. 女宿 Śravaṇa
22. 虚宿 Dhaniṣṭhā
23. 危宿 Śatabhiṣaj
24. 室宿 Pūrvabhādrapadā
25. 壁宿 Uttarabhādrapadā
26. 奎宿 Revatī
27. 婁宿 Aśvinī
28. 胃宿 Bharaṇī

In earlier Sanskrit works the list commences from Kṛttikā and later it is from Aśvinī, a change attributed to precession of the equinoxes. Indian astronomical models can exclude Abhijit and work exclusively with 27 nakṣatra-s. The Chinese system does not allow for this.

Generally in the 27 model each nakṣatra can occupy equal space, whereas in the 28 model the distances vary, and are measured by muhūrta (the time it takes the moon to pass through its space). A muhūrta is like an hour in that there are 30 per day, each comprised of 48 minutes (1440 minutes ÷ 48 = 30). Alternatively, Abhijit can be recognized yet subsumed under Uttarāṣāḍhā or Śravaṇa, as in the Mātaṅgī-sūtra 摩登伽經 (T 1300).

The number 27 divides into whole numbers such as with the ecliptic being divided into 108 pāda-s or quarters (.25), where each nakṣatra can be evenly assigned 4 pāda-s (108 ÷ 27 = 4). The twelve zodiacs, which were originally introduced to India from Babylonian and Hellenistic sources, are each assigned 9 pāda-s of the ecliptic (108 ÷ 12 = 9). The 9 pāda-s of a zodiac are divided among 3 nakṣatra-s which they are associated with. For example, Leo has the following assignments according to the Xiuyao jing 宿曜經 (T 1299), translated into Chinese (or rather compiled) in the mid-eighth century:

Maghā: 4 pāda-s.
Pūrvaphālgunī: 4 pāda-s.
Uttaraphālgunī: 1 pāda (the 3 remaining pāda-s are then subsequently assigned to Virgo).

On the Indian calendar the name of each day is derived from the nakṣatra in which the moon is in that night. The model Amoghavajra used does not use numerals to specify days and it is quite clear about this difference with the Chinese calendar. Some will predictably be visibly ahead or behind it, which is likened to a calf and its mother. This highlights that the model is arithmetic rather than observational. 

A month is named based on which nakṣatra the moon is in on the night of the fifteenth day of the śukla-pakṣa (waxing moon), which is the nominal full moon (the pūrṇāmānta method), though the actual full moon can visibly vary by a day, in contrast to the Chinese calendar where a month commences from the new moon (the amānta method). See here for the calendar table.

The aforementioned Mātaṅgī-sūtra in Chinese translation is exceptional in using the latter new moon method in contrast to the original Sanskrit version, but the gnomonic measurements of the Chinese translation also reflect revisions suited to a position north of India in Central Asian and moreover the text displays occidental influences. In Amoghavajra's Xiuyao jing the first month is Caitra (), corresponding to the second Chinese lunar month, though Vaiśākha can also be reckoned as the beginning of the year.

As a component to lunar astrology, each nakṣatra is associated with an Indian deity. This is a very ancient Indian custom. They are further associated with gotra names and foods, though these vary considerably in the texts. A person is also associated with the nakṣatra they are born under to which predictions about their personal character and fate are made. Auspicious activities are also prescribed for each day. This is easily determined if you know what day on the East Asian lunar calendar you were born on.

The following lists the Chinese xiu, Sanskrit nakṣatra, (Sanskrit month name), and associated deity plus variant in the Xiuyao jing if applicable. The associated deities are originally listed in the Nakṣatrakalpa of the Atharvavedapariśiṣṭā. It is unclear the reason behind the variant deities.

01. 昴宿 Kṛttikā (Kārttika) – Agni.
02. 畢宿 Rohiṇī – Prajāpati.
03. 觜宿 Mṛgaśīrṣa (Mārgaśīra) – Soma.
04. 參宿 Ārdrā – Rudra.
05. 井宿 Punarvasū – Aditi.
06. 鬼宿 Puṣya (Pauṣa) – Bṛhaspati.
07. 柳宿 Aślesā – Sarpa (Śeṣa)
08. 星宿 Maghā (Māgha) – Pitaras (Bhaga).
09. 張宿 Pūrvaphālgunī – Bhaga (Vasu).
10. 翼宿 Uttaraphālgunī (Phālguna) – Aryaman.
11. 軫宿 Hasta – Āditya (Savitṛ)
12. 角宿 Citrā (Caitra) – Tvaṣṭṛ.
13. 亢宿 Svāti – Vāyu.
14. 氐宿 Viśākhā (Vaiśākha) – Indrāgnī.
15. 房宿 Anurādhā – Mitra.
16. 心宿 Jyeṣṭha (Jyaiṣṭha) – Indra.
17. 尾宿 Mūla – Nirṛti.
18. 箕宿 Pūrvāṣāḍhā (Āṣāḍha) – Toya (Āpas)
19. 斗宿 Uttarāṣāḍhā – Viśvadeva.
20. 牛宿Abhijit – Brahmā.
21. 女宿 Śravaṇa (Śrāvaṇa) – Viṣṇu.
22. 虚宿 Dhaniṣṭhā – Vasu.
23. 危宿 Śatabhiṣaj – Varuṇa.
24. 室宿 Pūrvabhādrapadā (Bhādraphada) – Ajapāda.
25. 壁宿 Uttarabhādrapadā – Ahirbudhnya.
26. 奎宿 Revatī – Pūṣan.
27. 婁宿 Aśvinī (Āśvina) – Aśvin (Gandharva)
28. 胃宿 Bharaṇī – Yama.

The nakṣatra-s themselves are also regarded as deities in various esoteric Buddhist works. The respective deities are also represented in art. One fine example is the “Deities of the Five Planets and Twenty-Eight Constellations” 五星二十八宿神形圖 (Osaka City Museum of Fine Arts), which is attributed either to Zhang Sengyao張僧繇 (c. 490-540) or Liang Lingzan 梁令瓚 (a contemporary of Yixing 一行 683–727). See here for the full image file.

The iconography as line drawings is also cataloged in the Japanese Butsuzō zui 佛像圖彙 (see here for the text) from 1789.

The seven planets (Skt. sapta-grahāḥ) also naturally feature prominently in Buddhist astrology. These seven include the sun, moon and five visible planets. Uranus, Neptune and Pluto were unknown in the ancient world. In English “planet” used to include the sun and moon until the 1630s. Likewise in the sun and moon in Sanskrit can be considered graha-s. Here graha is defined as “seizing, laying hold of”. It seems there are multiple explanations for why graha would mean planet. Spirits and demons are understood as seizing and exerting influences on people, as so too it was believed the planets were capable.

In India an additional two “hidden planets” are added: Rāhu and Ketu, which then comprise nine planets.

Rāhu in Indian mythology is the demon said to be responsible for devouring planets and causing eclipses. The concept underwent development over time. He is first explicitly named in the Atharva Veda, and later in the Mahābhārata and Purāṇas associated with Svarbhānu who appears in the earlier Rg Veda. He was first described as a disembodied demonic head consuming the sun and moon, but later as a result of Greek influences in the early Gupta period he was revised as being the ascending node of the moon in astronomical literature, envisioned as a dark or hidden planet from which calculations could be made for predicting eclipses, and indeed it worked. 

Later this was coordinated with the entity Ketu, which became the descending node of the moon. The original Vedic meaning of ketu, however, meant rays of light and this came to refer to comets, the earliest reference of which is found in the Atharva Veda. Ketu was initially conceived of as solely the personification of comets, but sometime around or after 800 CE it was understood at the descending node of the moon. By the early second millennium Rāhu's imagery was further revised and he became a head of a serpent with Ketu as its tail owing perhaps to the influences from Middle Eastern astrology.

While Chinese texts have standard names for the seven planets, Chinese Buddhist works and almanacs will sometimes use transliterated loanwords from other languages. Amoghavajra provides in the Xiuyao jing the names of the seven planets in Chinese, Sogdian, Persian and Sanskrit. Their various names are as follows. The Persian terms are numerals, not planet names, used to indicate the day of the week.

Chinese: , 太陽.
Sogdian: , (myr).
Persian: (ēw).
Sanskrit: 阿彌底耶, 阿儞底耶 (āditya).

Chinese: , 太陰.
Sogdian: , (m'x).
Persian: 婁禍 (dō), 婁禍森勿 (dō šambih).
Sanskrit: 蘇摩 (soma).

Chinese: 火星, 熒惑.
Sogdian: 雲漠, 雲漢 (wnx'n).
Persian: (sĕ).
Sanskrit: 盎哦囉迦 (aṅgāraka).

Chinese: 水星, 辰星.
Sogdian: (ṭyr).
Persian: , (čahār).
Sanskrit: 部陀 (budha).

Chinese: 木星, 歳星.
Sogdian: 鶻勿, 溫沒斯 (wrmzṭ).
Persian: (panǰ).
Sanskrit: 勿哩訶娑跛底 (bṛhaspati).

Chinese: 金星, 太白, 啟明, 長庚.
Sogdian: 那歇, 那頡 (n'xyẟ).
Persian: (šaš).
Sanskrit: 戌羯羅, (śukra).

Chinese: 土星, 鎭星, , 填星.
Sogdian: 枳院, (kyw'n).
Persian: , (haft).
Sanskrit: 賖乃以室折羅, 室悉羅 (śanaiścara).

Planets associated with weekdays was a custom Buddhists adopted in India perhaps around the fourth of fifth century, which became it seems widespread among all Indians. The standard ordering of the seven-day week as we know it today originated in Egypt around the Common Era. The seven-day week was probably first employed around the Eastern territories of the Roman empire, the first usage found in the work of Dio Cassius (born 155 CE). It later became widespread from the third century before being adopted by Constantine in 321. Around the third to fourth centuries it was introduced into India. In the Vṛddhayavanajātaka, an astrological work by Mīnarāja dated to 300-325, the planets are for the first time in India listed in the temporal order of their regency over the days of the week. The earliest extant inscription referring to the seven-day week is dated to June 21st, 484 or 165 of the Gupta era during the reign of Budhagupta (r. c. 477-500).

This ordering of weekdays is a union of the Egyptian belief in deities overseeing each of the twenty-four hours and the Greek cosmological concept of concentric spheres. The spheres run in the descending order of Saturn, Jupiter, Mars, the Sun, Venus, Mercury and the Moon. The first hour of the first day is assigned to Saturn, the second hour to Jupiter, the third to Mars, and so on. The twenty-fifth hour (the first hour of the second day) is assigned to the Sun. The forty-ninth hour is assigned to the Moon. There are also earlier variant orderings which are reflected in Chinese translations.

As days of the week, each planet has astrological significance. Auspicious and inauspicious activities are assigned to each day. The fate of an individual is also influenced by which day they are born on. Furthermore, predictions about the future are made if certain weekdays fall on specific days of the lunar calendar.

In Hellenistic astrology, which deeply influenced Indian astrology and to some extent Chinese Buddhist models of astrology, the planets on the horoscopic chart are extremely significant. Jupiter, the Moon and Venus are positive or benefic, whereas Saturn and Mars are negative or malefic. The Sun and Mercury are of mixed qualities. These influences can be altered by various factors, such as which zodiacal house they are located in. The geometric relationships between the planets on a chart (called 'aspect') also influence events and individual fates. These concepts differ from native Chinese ones (Saturn for example is regarded as auspicious). These occidental ideas influenced late Tang literati society and actually displaced, at least for a time in the ninth century, the Chinese model in popular literature. However, at least one eminent court specialist in calendars in the following Song dynasty in the tenth century was also an expert in Hellenistic astrology.

Native Chinese astrology includes an art transliterated literally as “field allocation” 分野, which is a set of astral-terrestrial correspondences which assigns segments of the sky to geographical areas of China (designated by their ancient country names) and interprets the passage of planets through varying zones as being portentous, in particular conjunctions. This model can be traced back to the age of Confucius, long before any known Sino-Indian relations. Unlike in Hellenistic and Indian astrology, the concern is the state and not individuals. In the Warring States (403–221 BCE) period it was entirely sino-centric, but by the Han came to incorporate non-Chinese geographical zones as well as concepts such the five elements 五行 and yin-yang 陰陽 theory. By the late Tang (ninth century), the occidental and Chinese systems had been partially integrated.

In Mantrayāna Buddhist practice, there are mantras for each planet or all of them collectively, used either to enhance their positive influences or deflect their negative influences. These practices can be coordinated according to when it is astrologically appropriate to do so. In the Taishō canon the relevant texts are found between T 1302 ~ T 1312. The main mantras therein tentatively deciphered are as follows:

The Sun's mantra:
namaḥ ratna-trayāya namaḥ sūrya sarva-nakṣatra-rājāya oṃ amoghasya śāte svāhā

The Moon's mantra:
oṃ candra nakṣatra-rājāya śāte svāhā

Mars' mantra:
oṃ aṅgāraka arogya svāhā

Mercury's mantra:
oṃ budha nakṣatra svamina induja svāhā

Jupiter's mantra:
oṃ bṛhaspati nama pitṛvanāya mali vardhani svāhā

Venus' mantra:
oṃ śukra adhvan vrā rājāya śrī kavi svāhā

Saturn's mantra:
oṃ śanaiścara nakṣatra prahā manā rupaya puṣṭikari svāhā

Rāhu's mantra:
oṃ rāhuna asura-rājāya suma śatunaya śāntikari svāhā

Ketu's mantra:
oṃ vajra ketunā nakṣatra-rājāya hūṃ svāhā

There is a separate set of mantras appended to the Tejaprabha dhāraṇī (T 964), though these appear to have not been part of the original text:

namaḥ samantabuddhanāṃ oṃ mahāśubhiya svāhā

namaḥ samantabuddhanāṃ oṃ indrāya svāhā

namaḥ samantabuddhanāṃ oṃ varuṇāya svāhā

namaḥ samantabuddhanāṃ oṃ agniye svāhā

namaḥ samantabuddhanāṃ oṃ pṛthiviye svāhā

namaḥ samantabuddhanāṃ oṃ rāhula asurarājaya svāhā

namaḥ samantabuddhanāṃ oṃ praketunkṣatrarajāya hūṃ svāhā

namaḥ samantabuddhanāṃ oṃ aditya svāhā

namaḥ samantabuddhanāṃ oṃ saumaya svāhā

The planets each possesses unique features in iconography where they are depicted in human forms. The earliest extant representations of the planetary deities in India are from the fifth century (Gupta period). Chinese artists generally depicted the deities in Chinese garb with associated animal caps. One noteworthy extant specimen is "Tejaprabhā Buddha and the Five Planets" 熾盛光佛并五星圖 from Qianning 乾寧 4 (897 CE) by Zhang Huaixing 張淮興 from Dunhuang (British Museum Asia OA 1919,1-1,0.31).

A Brahmin depicting Saturn, carrying a staff and directing the ox. Atop him is an ox head.

The female figure on the bottom is Venus depicted as a beautiful lady playing the pipa 琵琶. Her skin tone is snow white, reflecting a Tang aesthetic. She has a bird hat atop her head.

At the bottom right the four-armed heterodox figure is Mars. He has a donkey atop his head. He carries an arrow, bow, double-edged sword and trident.

To the top there is another female figure dressed in Chinese attire holding a brush and paper with a monkey atop her head. She is Mercury.

The minister in the top left is Jupiter. He carries a plate of flowers and fruits while wearing a pig hat.

The aforementioned “Deities of the Five Planets and Twenty-Eight Constellations” also depicts the planets in similar garb, though they ride animals rather than wearing animal caps:

To summarize, I want to emphasize that Buddhist traditions, both in India and East Asia, were full of astrological lore, beliefs and practices. This influenced both religious practices as well as the art record. Planetary invocations and star worship, it seems, were quite common during the so-called "golden ages" of Buddhism in both India and China. I have not investigated Tibetan Buddhism in this respect closely, but I can refer readers to Edward Henning's quality work on the subject of astrology in Tibet (see here). I have seen for myself how important astrology is for many Lamas and monks around India and Nepal, and they've assured me how popular it is. It seems historically this was just as much the case as well.

In an upcoming paper to be formally published I will discuss how originally Buddhism and perhaps the Buddha himself dismissed astrology and prohibited monks from practicing it. The paradox, of course, is that historically many Buddhists did in fact practice astrology despite such prohibitions and refutations of astrology. The history of all this is rather complex and hopefully within a few years my dissertation on the subject will be finished and I can share it with everyone.


Barton, Tamsyn. Ancient Astrology. New York: Routledge, 1994.

Beck, Roger. A Brief History of Ancient Astrology. Oxford: Blackwell Publishing, 2007.

Chan Man Sing 陳萬成. "Du Mu yu xingming" 杜牧與星命. Tang yanjiu 唐研究 8 (2002): 66-68.

David Pingree. “Identification of the Yogatārās of the Indian Nakṣatras.Journal for the History of Astronomy 20 (1989): 99-119.

Pankenier, David. Astrology and Cosmology in Early China Conforming Earth to Heaven. Cambridge: Cambridge University Press, 2013.

Yano, Michio. “Calendar, Astrology, and Astronomy.” The Blackwell Companion to Hinduism. Oxford: Blackwell Publishing, 2003.

Yano, Michio 矢野道雄. Mikkyō senseijutsu 密教占星術. Tokyo: Toyoshoin, 2013.

Decoding Sanskrit Mantras with Japanese Sources

As I noted in an earlier post (see here) there are often great challenges posed when attempting to decode a mantra or dhāraṇī from garbled archaic Chinese transliteration. Fortunately, there are modern resources such as dhāraṇī dictionaries which have done a lot of the necessary preliminary work, but there are still plenty of cases where you are entirely on your own.

Modern Japanese and Chinese academic works can cite these texts and just insert the original Chinese, though if you are translating a text into English this is not possible. The Sanskrit must be deciphered, or an attempt at least must be made. Occasionally a key aid to this process will be old Japanese texts – now digitized – which both preserve accurate transliterations and define words.

Recently I've been working with esoteric texts related to astral practices. One such work is the Big Dipper Seven Stars Homa Rite 北斗七星護摩法 (T 1310) attributed to Yixing 一行 (684-727). This is not actually his work, but I will publish a paper detailing why in the future. It contains a number of mantras or dhāraṇīs in garbled Middle Chinese. One of them is a “praise” and is found in a few other esoteric texts. One does provide transliteration in Siddhaṃ (Sanskrit) script, but it is largely incomprehensible. The text is the Ritual of the Deity Mahākāla 大黑天神法 (T 1287). It provides the following:


ayaṃtu deva sagasra kaṃnarendra sakaradayā pravara dharmma kṛtadhikara vidharmma ca prasama saukhyai nimeta bhuta meta prakaṣaya tadiha śramathāya dhamaṃ1

Here we have the dhāraṇī transliterated with Chinese characters as they were pronounced in Middle Chinese (the pronunciations are better preserved in Japanese than in modern Mandarin). The accompanying Siddhaṃ, transliterated into Roman below, is clearly Sanskrit, but quite baffling. An uncritical religious practitioner might assume – as was often done in the past – that Sanskrit is the language of the devas, hence it is not necessarily possible to understand the meaning. You just pronounce it and the power inherent in the language is supposed to work. From the perspective of sacred Sanskrit phonology that might be true, but only if you precisely pronounce each syllable properly.

Needless to say, Sanskrit transliterated with Chinese characters will not accurately represent Sanskrit pronunciation. To complicate matters further the scribes who reproduced the Siddhaṃ in the copied texts – both in China and Japan – might not have understood Sanskrit. The alphabet might have been sacred, but grammars and Indian teachers were scarce, especially after persecution of Buddhism under Emperor Wuzong in 845 and the collapse of the Tang dynasty (907).

I should note here that Sanskrit studies did continue in Japan throughout the medieval period to some extent. One noteworthy figure in this respect was Myōkaku 明覺 (1056–c.1122). He was a Tendai scholar monk of the late Heian period (794–1185), noted for his work on phonetics and Siddhaṃ. In his youth he studied at Mt. Hiei 比叡山 before later residing at Onsenji 溫泉寺. In addition to his linguistic works, his Daizuigu darani kanchū 大隨求陀羅尼勘註 (T 2242) and Dai bucchō nyorai hōkō Shittatahattara darani kanchū 大佛頂如來放光悉怛他鉢怛囉陀羅尼勘註 (T 2235) are annotated dhāraṇī texts complete with Siddhaṃ. Nevertheless, it doesn't seem like such studies amounted to comprehensive understandings of grammar or vocabulary, such as would be the case in Sanskrit studies today.

Returning to our dhāraṇī in question, Jōnen 靜然 in his Gyōrin Shō 行林抄 (T 2409), compiled in 1154, provides it along with a critical evaluation of differing manuscripts available to him (see here for the text). The corrupted Siddhaṃ version differs from the one cited above.

He initially states that it is titled in various ways including “Praise to the Eight Groups of Devas and Nāgas” (梵字天龍八部讃) or “Petitioning the Devas and Nāgas” 請天龍. Judging from the title, it was probably a short text brought to Japan by Kūkai 空海 in 806 as it appears in his catalog (see here). Jōnen's work notes there was the Sanskrit version of the text as well as bilingual Sanskrit-Chinese version. This same dhāraṇī appears incorporated into a few texts in the Taishō under different names.

His notes become critical in deciphering the dhāraṇī. For example, he notes zuo ga su ra 左誐素囉 (sagasra in the corrupted Siddhaṃ above) in the Chinese should be bhujaga asura. He identifies the first as a Nāga and the latter as a non-deva (asura). In alternate manuscripts the first word in Chinese normally commences with a labial consonant (something like [b]). However, here it was omitted, and moreover the words ended up combined as the Chinese had a tendency to drop the initial [a] from Sanskrit words. Fortunately Jōnen could compare and report from several manuscripts.

Another error in the text above is vidharmma ca (“...and unlawful/wrong“). He gives voddhaṃvacaḥ and suggests it might mean “Buddha speaks” (佛語). It should read buddhaṃ vacaḥ. Again, without these notes it would be arguably impossible to confidently decipher the dhāraṇī in the absence of alternate manuscripts, especially for someone with elementary Sanskrit knowledge such as myself.

The end result of his notes results in a dhāraṇī tentatively as follows:

ayantu deva bhujaga asura kiṃnarendra śakradayā pravara dharma kṛtādhikārā buddhaṃ vacaḥ praśama saukhya nimitta bhūta mita prakāśya tadiha śravaṇyai dharmaṃ

Compare this with the alternate version cited above:

ayaṃtu deva sagasra kaṃnarendra sakaradayā pravara dharmma kṛtadhikara vidharmma ca prasama saukhyai nimeta bhuta meta prakaṣaya tadiha śramathāya dhamaṃ

This illustrates how poorly Sanskrit – even in fragments – was preserved in East Asia (the Tibetans were far superior in preserving mantras with their phonetic alphabet which was derived from a Sanskrit script). Even in earlier ages the lack of understanding of Sanskrit was lamented by some erudite scholars like Xuanzang (602-664) 玄奘, who visited India between 633-645. He was quick to point out “accented” forms of the language that he disapproved of both domestically and abroad. He also provided updated transliterations in Chinese, which within a few centuries likely deviated from the intended pronunciation, given how spoken Chinese rapidly evolved. The character e (meaning "accented") appears 93 times in his travel account the Record of Travels to Western Lands 大唐西域記.

Xuanzang, echoing the widespread Buddhist belief in his time, believed in the divinity of Sanskirt. He describes the languages of India as follows, which probably echoes the sentiments of his Indian colleagues:

《大唐西域記》卷2:「詳其文字,梵天所製,原始垂則,四十七言也。寓物合成,隨事轉用。流演枝派,其源浸廣,因地隨人,微有改變,語其大較,未異本源。而中印度特為詳正,辭調和雅,與天同音,氣韻清亮,為人軌則。隣境異國,習謬成訓,競趨澆俗,莫守淳風。」(CBETA, T51, no. 2087, p. 876, c9-14)

Their letters were established by Brahma and have been passed down from their beginnings until now, being forty-seven in number. They combine to form words according to the object [declension?] and shift in use according to the action [inflection?]. It has spread around and branched off, its source being deep and broad. Due to regions and peoples there have been some changes, though the words are generally not different from the original source. Central India is especially proper, their diction being elegant and the same sound as devas, with a character sharp and clear, which is a model for people. The neighboring countries have become accustomed to erroneous pronunciation. In their chaotic ways and base nature they do not maintain genuineness.

After the Tang dynasty there were fewer Indians visiting China. In fact, the translation projects of the Northern Song dynasty (960-1127) ended up as failures even with Indian staff and official funding from the court.2

To some extent there was still an interest in Sanskrit for a few centuries at least among some intellectuals. Shen Kuo 沈括 (1031-1095), an intellectual with various interests including Sanskrit, was aware even in his day of Sanskrit influences in Chinese linguistics. Consider the following remark from his work Dream Pool Essay 夢溪筆談.


As to the study of phonetics, the methods have been refined from Shen Yue 沈約 [441–513] dealing with the four tones to when Indian Sanskrit studies were introduced into China.

He was aware of the phonetic table of Sanskrit, though perhaps this was simply an intellectual curiosity on his part.

As noted above, the situation was different in Japan where Mantrayāna survived intact in Shingon and Tendai lineages. Moreover, alongside this there was an interest in Siddhaṃ studies (Jpn. shittangaku 悉曇) until the present day. Figures like Jōnen and Myōkaku preserved this inherited field from Tang China that was otherwise lost there.

Kōyasan, Japan.
This further highlights the value of Japanese Buddhism in understanding earlier forms of Chinese Buddhist literature and culture. The collapse of the Tang dynasty saw much Buddhist civilization forgotten. A great deal of Chinese literature was preserved exclusively in Japan, which the Chinese later became aware of and reintroduced. It was actually more than literature that was preserved: the Yogācāra school of East Asian (Faxiang 法相 in Chinese) and Mantrayāna as living traditions were preserved in Japan, along with other material aspects of Tang Buddhist culture including architecture, attire, rites and so on.

In the modern day if you want to know classical Chinese Buddhism, you need to look to Japan. Arguably the true heirs to Tang Buddhism are in Japan. This is further made clear when you consider how Buddhism in China was basically decapitated by the communists in the twentieth century and only pieced back together in Taiwan.



1 (CBETA, T21, no. 1287, p. 357, b21-c4)

2 For a good study of this subject see Tansen Sen, Buddhism Diplomacy and Trade The Realignment of Sino-Indian Relations 600-1400 (Honolulu, HI: University of Hawai'i Press, 2003).