Variables

(2) This hill is non-fiery.

(3) Therefore this hill is not smoky.

F. ARISTOTLE.

Demonstration through the medium of essence and cause.

The two relations, viz. " self-sameness" (or " inherence ")and ''effect", as expounded by Dharmaklrti, correspond to the two relations, viz. " per se " (inherence) and " causal ', as expounded by Aristotle in his Posterior Analytics (bk. i, ch. iv, pp. 253-5, and bk. i, ch. xxiv, p. 301).

(1) Per se (essence or inherence). —The predicate (major terms) is said to be related to the subject (middle term per se essentially), if the properties of the former are inherent in the definition of the latter.

(2) The causal.—The predicate (major term), which is essentially present with the subject (middle A History of Indian Logicject (middle term) stand to each other in the relation of cause and effect, e.g.

all that is smoky is fiery.

(3) Non-perception (anupalab-dhi); e.g. whatever thing (being perceptible) is not perceived is non-existent.

j. Nature of the universal preposition.

The relations which estab' lish invariable concomitance othe middle term with the major term are the bases of universal propositions. "Invariable concomitance" is designated in Sanskrit as vyapti (pervasion or co-presence), ndntarzyaka (non-separation); and avinabhava (the relation owing to which one cannot exist without the other).

A History of Indian Logicterm), is said to be the cause of the latter.

j. Nature of the universal proposition.

The proposition in which the predicate is related to the subject per se or causally is a universal one.

Aristotle in his Posterior Analytics (bk. i, ch. iv, pp. 253-5, 6. F. Owen's edition) calls that (the) universal (major term) which is predicated " of every " and " per se ", that is, which being predicable of the middle term per se is predicated of it in every instance. In the Posterior Analytics (bk. i, ch. xxiv, p. 301) Aristotle further observes that the universal (major term) is the cause of the middle term, which is essentially co-present within it. In Prior Analytics, bk. i. ch. xxiv, pp. 138-9, Aristotle says that in all syllogisms we must have a universal proposition (premise) which is shown by the universal term.

A History of Indian LogicG. UDDYOTAKABA (600-50 A.D.) fe. Syllogism from a sign (linga-paramarsa).

Uddyotakara, who completely incorporated syllogism into inference in his Nyaya-varttika,* defines inference as knowledge which is preceded by the perception of the middle term {sign) and recollection of its invariable concomitance with the major term in the following form: " I perceive that this hill has smoke, which I remember to be invariably concomitant with fire, and hence I infer that this hill has fire."

The above may be properly put in the following form:—

(1) Whatever is smoky is fierv.

(2) This hill is smoky.

(3) Therefore this hill is fiery.

The first part (major premise) of the above inference is called vydpti (a universal proposition) ; the second part (minor premise) is called upanaya (an application of the universal);

and the third part is called anu-miti (inferential knowledge or conclusion). The first two parts (premises)

constitute what is

A History of Indian LogicG. A&ISTOTLE. k. Enthymeme from a sign.

Aristotle, in his Prior Analytics (bk. ii, ch. xxvii, pp. 238-40), speaks of two kinds of enthymemes, viz.

the enthy-raeme from a sign and that from a likelihood. Now, the enthymeme from a sign in the first figure (especially if the signiis infallible, tekmerion) leads to a conclusion which is necessarily true. It is this kind of enthymeme from a sign that corresponds to the syllogism from a sign * (linga-paramarsa) as expounded by Uddyotakara.

Just as in the Nyaya-varttika of Uddyotakara the syllogism (paramarsa) comprises only the first two parts of an inference, so also in the Arabic Logic,' (e.g.inbk. iii of the Risa-lah Shamsiyyah, dated about 1250 A.D.) the syllogism (qayas} comprises only the two premises and not the conclusion. It may be added that the Arabic Logic of the schools of Baghdad (ninth century A.D ), Kufa (750 A.D.), and Bassora (700 A.D.) is re

A History of Indian Logiccalled by Uddyotakara pard-тагба '(syllogism) or linga-paramarsa (syllogism from a sign), and all the three parts combined together are designated as para-тагйба rйpanumana (a syl-logistic inference). Since the time of Uddyotakara no further development has been made in the form of the syllogistic inference, except that the three parts of it have been called res-pectively the instrument (ka-rana), operation (vyapara), and consequence (phala). The first two parts (premises) combined together have been uniformly designated as paramarsa (syllogism or enthymeme).

A History of Indian Logicputed to have been derived from the Organon of Aristotle, as taught in a developed form in the Syro-Persian School of Gundeshapur' in Sueiana about 350 A.D.

According to certain Roman commentators * too, the premises alone constitute the syllogism.

A History of Indian Logic6. Migrations of the Logical Theories of Aristotle from Alexandria into India (175 B.C.-600 A.D.)

Considering the antiquity of the syllogism as propounded by Aristotle and the close connexion that exists between it and the syllogism promulgated in the Hindu Logic, we may fairly conclude that the latter was greatly influenced by, if not based on, the former. Aristotle's works were brought down to Alexandria (in Egypt) by Callimachus, the celebrated librarian of Ptolemy Philadelphus during 285-247 B.C., and it seems that copies of some of these works reached India through Syria, Susiana, Bactria, and Taxila in subsequent times. From the stages in the development of the syllogism in Hindu Logic, as indicated above, it will appear that Aristotle's works migrated into India during three distinct periods.

The first period extends roughly from 175 B.C. to 30 B.C., when the Greeks occupied the north-western parts of

A History of Indian LogicIndia ' and had their capital at Sakala, officially called Euthv-demia (modern Sialkot) in the Punjab. The work of Aristotle of which we find a trace in this period is the Art of Rhetoric, which was evidently a favourite subject of study among the Indian Greeks, and from which the syllogism of five members as illustrated in the Caraka-samhita, referred to above, seems to have been derived. It is worthy of note that the first trace in India of Aristotle's syllogism is met with in a work the author of which was the chief physician to King Kaniska, who reigned in the Punjab, if not exactly in the city of Sakala, at any rate near to it. The second period extends from about 39 B.C. to 450 A.D., when the Romans, masters of Alexandria, Syria, and Persia, carried on a brisk trade between Alexandria and India. The work of Aristotle which comes to our notice in this period is the Prior and Posterior Analytics (and possibly also the De Interpretatione). from which Aksapada, Nagarjuna, Vasubandhu, and Dignaga seem to have, as shown above, borrowed the definition of some of the most important logical terms, the explanation of the various structures of the syllogism and the idea of a universal position. the basis of a true syllogism. The two chief among these logicians, Aksapada and Dignaga, were inhabitants of Kathiawar (Prabhasa) and Conjeeveram (Kanchi), which were the principal seaports on the eastern and western coasts of India, frequented by merchants and travellers from Alexandria. It is probable that the Prior Analytics was widely read in those days, either in the original or in vernacular translation. The introduction of different parts of the Greek Prior Analytics into Indian Logic must needs have been gradual, as these had to be assimilated into and harmonized with the parts previously ingrafted into Indian thought and language. The third period extends from about 450 A.D. to 600 A.D. when the Syro-Persian school of Gundeshapur, ' established in Susiana (Persia) in 350 A.D. on the dispersion there of some of the beat works of the school of Alexandria, was in the height of its glory and spread its light all around. But it is not quite certain whether any influence was exerted at this period.

I have endeavoured in the preceding pages to show that the works of Aristotle were very well known in India during the first six centuries of the Christian era. As regards the remarks of Cicero and Strabo, quoted by Sir A. Grant, that most of Aristotle's " writings had been lost," I have to state that after the death of Aristotle his entire works were bought up by Callimachus * for the

A History of Indian Logiclibrary of Alexandria, where they were duly appreciated and whence they gradually spread to India and other countries. To the scholars at Athens and Rome these works were practically lost, until copies of some of them reached the island of Rhodes, where they were edited by Andronicus in 50 B.C. Even the edition of Andronicus was not available in the Middle Ages to the Greeks and Romans, who depended for their knowledge of Aristotle on the Latin translation of Boethius (480-525 A.D.). But the original works of Aristotle seem to have been carefully preserved in Alexandria, and on the downfall of the Greeks and Romans they found their way into Syria and Persia, whence they reached the Arabic school of Bagdad about the beginning of the ninth century A.D- The original Greek texts of Aristotle's works after these strange vicissitudes reached the country of their birth via Con-stantinople about 1204 A.D. The presumption, therefore, is that from the third century B.C. to 1200 A.D.

Aristotle's works were more extensively read and better appreciated in the East than in the West.

A History of Indian LogicAPPENDIX С.

THE U3S4VERSITY OF NALANDA.

(ABOUT 300—850 A.D.).

Nalauda was a village which is identified with modern Bara-gaon, 7 miles north of Rajgir, in Behar.

Though occasionally mentioned in the Pali literature, Nalanda was not of great importance before the rise of the Mahayana at the beginning of the Christian era. Nagarjuna, about 300 A.D., and Arya Deva, about 320 A.D., were the earliest scholars to take an interest in the educational institution at that village.

A Brahmana named Suvisnu, a contemporary of Nagarjuna, is said to have estab-. lished 108 temples there in order that the Abhidharma of the Mahayana might not decline. About 400 A.D. the Chinese pilgrim, Fabian, visited this place, which he calls " the village of Nalo." He saw there a tower which had been erected on the spot where Sariputra, the right-hand disciple of Buddha, had entered Nirvana. Early in the 7th century A.D. another Chinese pilgrim, the famous Hwen-thsang, visited Nalanda and halted there 16 months to study the Sanskrit language under Silabhadra. According to him the site of Nalanda was originally a mango garden which was bought by 500 merchants at a cost of ten crores of gold pieces and given to Buddha.After the Nirvana of Buddha, five kings, named Sakraditya, Buddha Gupta, Tathagata Gupta, Baladitya, and Vajra, built five Sangharama or monasteries at Nalanda. A king of Central India established another magnificent monastery, and began to build round these edifices a high wall with one gate. A long succession of kings continued the work of building, using all the skill of the sculptor, till at the time of Hwen-thsang in 637 A.D. the whole was " truly marvellous to behold." In the establishment were some thousands of monks, all men of great ability and learning. They were very strict in observing the rules of Vinaya, and were looked up to as models by all India. Learning and discussing, they found the day too short, day and night

A History of Indian Logicthey admonished each other, juniors and seniors mutually helping to perfection. Learned men from different cities came to Nalanda to acquire renown, and some persons even usurped the name of Nalanda students in order that they might be received everywhere with honour. '' Of those from abroad who wished to enter the schools discussion, the majority, beaten by the difficulties of the problems, withdrew; and those who were deeply versed in old and modern learning were admitted, only two or three out of ten succeeding.'' Hwen-thsang mentions some celebrated men of Nalanda, such as Dharmapala and Candrapala, Gunamati and Sthiramati, Prabhamitra and Jina-mitra, and Jnanacandra and Silabhadra.

Another Chinese pilgrim named 1-tsing, who resided in Nalanda for ten years (probably 675—685 A.D.), says that there were eight halls and 300 apartments in the monastery of Nalanda with more than 3,000 resident monks. The lands in its possession contained more than 200 villages which had been bestowed upon the monastery by kings of different generations.

Nalanda assumed the character of a university from about 450 A D.* Baladitya, king of Magadha, who built a monastery at Nalanda, was a contemporary of the Hun king Mihirakula, who reigned first in Sakala and afterwards in Kasmira. Now Mihirakula began his reign in 515 A.D., and his contemporary, Baladitya, must also have lived about that time. There were three predecessors of Baladitya who built monasteries at Nalanda. Of them, the earliest, named Sakraditya, must have reigned about 450 A.D. if we suppose 25 years as the average duration of the reign of each of them. The year 450 A.D. is then the earliest limit which we can roughly assign to the royal recognition of Nalanda. The latest limit which we know with certainty is

A History of Indian Logic750 A.D., when Kamalasila (q.v.) was the'professor of Tantras at Nalanda. But as we read in the accounts of Vikramasila that there was for some time an intercourse between that university and Nalanda, we may suppose that the latter continued to exist approximately until 850 A.D.

According to Tibetan accounts' the quarter in which the Nalanda University, with its grand library, was located, was called Dharmagafija (Piety Mart). It consisted of three grand buildings called Rafcnasagara, Ratnodadhi, and Batnaranjaka, respectively. In Ratnodadhi, which was nine-sfcoreyed, there were the sacred scripts called Prajnaparamita-sutra, and Tantrik works such as Samaja-guhya. etc.

After the Turuska raiders had made incursions in Nalanda, the temples and Caityas there were repaired by a sage named Mudita Bhadra. Soon after this, Kukutasiddha, minister of the king of Magadha, erected a temple at Nalanda, and, while a religious sermon was being delivered there, two very indigent Tirthika mendicants appeared. Some naughty young novice-monks in disdain threw washing-water on them. This made them very angry. After propitiating the sun for 12 years, they performed a yajna, fire-sacrifice, and threw living embers and ashes from the sacrificial pit into the Buddhist temples, etc. This produced a great conflagration which consumed Ratnodadhi. It is, however, said that many of the Buddhist scriptures were saved by water which leaked through the sacred volumes of Prajnaparamita-sutra and Tantra.

A History of Indian LogicAPPENDIX D.

A LIST OF KINGS OF THE PALA DYNASTY OF BENGAL AND BEHAR.

(From Tibetan sources.}

In the Tibetan books,* such as Pag-sam-jon-zang, Lama Tara-natha's Chos-byun. etc., we find a short account of the kings of the Pala dynasty of Bengal. Go Pala, the founder of the dynasty, lived principally in Pundra-vardhana. His successor, Deva Pala, annexed Varendra to his kingdom. Deva Pala's grandson, Dharma Pala, conquered Magadha and annexed it to Bengal. Dharma Pala's power is said to have extended in the east to the ocean, in the west to Delhi, in the north to Jalandhara, and in the south to the Vindhya ranges. It is stated that during his reign Santa Raksifca died. Now Santa Baksita visited Tibet during the reign of Thi-srong-deu-tsan in 749 A.D., and worked there for 13 years, that is, till 762 A.D. His death must therefore have taken place after 762 A.D. Dipankara Srijnana alias Atisa, High-priest Vikramasila, who was a contemporary of king Naya Pala of Magadha, visited Tibet in company with Nag-tsho-lotsava in 1040 A.D. during the reign of Lha-tsun-byan-chub, son of Lha-lama-ye-ses-hod, who held his court at Tholing in Nari. These facts throw a good deal of light on the dates of the Pala kings. It is further stated that the death of Mahi Pala is exactly synchronous with that of the Tibetan king Khri-ral. Now Khri-ral (or Ral-pa-can) died in 899 A.D.* This fixes the date of the death of Mahi Pala. As the period of reign of each of the kings that preceded and succeeded Mahi Pala is definitely stated by Lama Taranatha, and also by the author of the Pag-sam-jon-zang, there is no difficulty in ascertaining the dates of the Pala kings. Proceeding in this way, we can fix the dates as follows :—

A History of Indian Logic1.

Go Pala 2. Deva Pala 3. Rasa Pala

660—705 AD.

705—753 A.D.

753—765 A.D.

4. Dharma Pala 5. Masu Raksita 6. Vana Pala 7. Mahi Pala 8. Maha Pala 9. Samu Pala '

10. 6restha Pala or Praistha Pala 11. Canaka

12. Bhaya Pala 13. Naya Pala 14. Amra Pala 15. HastiPala 16. Ksanti Pala 17. Rama Pala 18. Yaksa Pala

A History of Indian LogicThe researches on the Palakings, by the late Dr. Rajendra Lal Mitra,arrived at a conclusion which is somewhat different

from mine.Dr. Mitra's list of Pala kings* is given below:—

1. Go Pala.. .. .. 855—875 A.D.

2. Dharma Pala .. .. 875— 895 A.D.

3. Deva Pala .. .. 895— 915 A.D.

4. Vigraha Pala I .. .. - 915— 935 A.D.

5. Narayana Pala .. .. 935— 955 A.D.

A History of Indian Logic765—

829 A.D. 829

— 837 A.D.

837— 847 A.D. 847—

899 A.D. 899

— 940 A.D.

940— 952 A.D. 952—

955 A.D. 955

— 983 A.D.

983—1015 A.D. 1015—

1050 A.D.

1050—1063 A.D. 1063—

1078 A.D.

1078—1092 A.D. 1092—

1138 A.D.

1138—1139 A.D.

6. Raja Pala .. .. 955— 975 A.D.

7. .... Pala .. .. 795— "995 A.D 8. Vigraha Pala II .. .. 995—1015 A.D.

9. Mahl Pala .. .. 1015—1040 A.D.

10. Nava Pala .. .. 1040—1060 A.D.

11. Vigraha Pala III .. .. 1060—1080 A.D.