Sir John Marshall Dug Taxila From 1913 to 1934 and Found Three Cities but No Lecture Hall, While the Jatakas Have a Prince Arriving With a Teacher's Fee of 1,000 Pieces

Taxila, about 35 km (22 miles) northwest of Rawalpindi in Pakistan’s Punjab, appears in the Buddhist Jataka stories as the place young men went to finish their education. The Punjab government’s Taxila Museum goes further. Its page on the Jaulian monastery says the monastery “was home to one of the world’s oldest universities”, and the page names no source. Sir John Marshall, who excavated the site from 1913 to 1934, wrote in his 1951 report: “Whether there was any systematic organisation for the professors or their pupils we do not know.”

Three cities, each built beside the last

Marshall’s digs exposed three city sites on one plain, and each belongs to a different set of rulers. The Bhir Mound is the oldest. According to UNESCO it was probably founded in the 6th century BCE by the Achaemenians, the Persian dynasty that held Gandhara, the region around Taxila. In the 1951 report the lowest settlement is probably fifth century BCE, “though it may have been earlier”. The third layer is the town Alexander entered in 326 BCE, in Marshall’s reading, and the second is the town of later Mauryan rule in the third century BCE.

Early in the 2nd century BCE, Marshall wrote, the Bactrian Greeks moved the capital across a stream to Sirkap, a walled city laid out on a grid. Sirkap’s founding falls in the middle of that century in UNESCO’s dating, decades later than Marshall’s. The city stayed in use under Greek, Saka (Scythian), Parthian and Kushan rulers. Who built the third city, Sirsukh, depends on the source. The 1918 guide said the Kushans probably built it under Kanishka, the 1951 report places the move in the time of Vima Kadphises, and Britannica names Kanishka. For its part, UNESCO says only that Sirsukh dates to the 1st century CE and that its excavated walls are of dressed stone with rounded bastions.

Buddhist stupas (domed relic mounds) and monasteries fill the hills around the three cities. The masonry at Mohra Moradu follows a style that came in during the second century CE, Marshall found, and some work at Jaulian belongs to the fifth. The end came with the White Huns. He put their invasion at about 460 CE, using the Chinese pilgrim Sung Yun’s report, and described wrecked monasteries and the skeletons of the slain lying in the debris. When the Chinese monk Xuanzang visited around 630 CE, Marshall wrote, he found most of the monasteries still ruined. Xuanzang’s wider travels are covered in the article on Harshavardhana.

What the spade settled and what it did not

Alexander Cunningham matched the ruins to ancient Taxila in 1863 to 1864 and 1872 to 1873, Britannica says, and Marshall then dug for more than twenty years. His three volumes, published by Cambridge University Press in 1951, cover the work done “between the years 1913 and 1934”.

The Dharmarajika stupa shows how far the digging could go. When Marshall began there in 1913 the base lay under 20 or 30 feet of fallen debris. The Taxila Museum’s page says the stupa “was erected by Ashoka himself”. Marshall was more guarded and inferred an Ashokan origin from the stupa’s name and from the years Ashoka spent at Taxila as viceroy for his father Bindusara, then added that nothing of the original structure, or of any other monument at Taxila, can be definitely recognised as Mauryan.

One find does tie the period to the ground. A broken marble pillar with an Aramaic inscription, built into a wall at Sirkap, honours an official named Romedote under a governor called Priyadarsi. Following F. C. Andreas’s reading, Marshall wrote that there was virtually no doubt Priyadarsi was Ashoka, ruling Taxila before he became emperor. The Dictionary of Pali Proper Names, citing the Buddhist Divyavadana, records a revolt at Taxila that Ashoka was sent to put down. Britannica says Taxila passed into the empire of Chandragupta Maurya within a decade of Alexander’s death.

What the Greeks saw in 326 BCE

The earliest outside descriptions come from Alexander’s campaign, and the surviving accounts rest on the writings of his companions and contemporaries, as Marshall noted. In Arrian’s version, translated by E. J. Chinnock in 1884, Alexander crossed the Indus and reached Taxila, “a large and prosperous city, in fact the largest of those situated between the rivers Indus and Hydaspes”. The governor, Taxiles, received him in a friendly manner. Strabo called it “a city which is large and has most excellent laws”.

Teachers appear in these accounts as individual sages with followers. Arrian says that at Taxila Alexander saw the naked sect of Indian philosophers and wanted one to join him. The oldest, Dandamis, “of whom the others were disciples”, refused to come and would not let the others go.

A later visitor is more doubtful. Philostratus, the biographer of Apollonius of Tyana, describes a Taxila about the size of Nineveh, fortified in the Greek manner, with narrow streets and a plain palace. Marshall wrote that the story mixes in much fiction, but that some details find “remarkably strong corroboration” in his own discoveries.

What the Jatakas say about students

The Jatakas are Buddhist stories of the Buddha’s former lives, and in many of them a young man goes to Taxila to study. In Jataka 252, in W. H. D. Rouse’s English version, a king of Benares sends his sixteen-year-old son there with one-soled sandals, a sunshade of leaves and a thousand pieces of money. The teacher asks whether he has brought a fee “or do you wish to attend on me in return for teaching you”. The prince has: “I have brought a fee with me.” The text explains that resident pupils wait on the teacher by day and learn from him at night, while fee-paying pupils “are treated like the eldest sons in his house”.

The Vinaya, the Buddhist monastic code, tells of the physician Jivaka, a foundling raised by a prince. Jivaka went to Taxila to learn from one world-renowned physician and studied there seven years. His teacher then told him to search around the city with a spade for any plant that was not medicinal. None turned up, and the teacher said his learning was done.

G. P. Malalasekera’s Dictionary of Pali Proper Names gathers the references. Pupils studied the three Vedas and eighteen sciences, including archery, medicine and elephant-handling. The usual fee was a thousand pieces. Pupils could be flogged with a bamboo stick, and only Brahmin or Kshatriya (warrior-caste) pupils appear to have been admitted. Malalasekera also points out that Taxila is never mentioned in the suttas, the discourses ascribed to the Buddha himself, though the Jatakas present it as a great centre of learning from before the Buddha’s day.

Scholars treat these texts with care. E. B. Cowell, the editor of the standard English translation, wrote in 1895 that the verses are the older core of the Jataka. Sri Lankan tradition has the prose stories translated into Pali about 430 CE by Buddhaghosa, and Cowell accepted that they took that form not long after, from material handed down from much earlier. Cowell added that he had no confidence in the historical credibility of the framing episodes that tie each story to an incident in the Buddha’s life. In Education in Ancient India (second edition, 1944), A. S. Altekar wrote: “Unfortunately we know very little about its educational activities from non-Jataka sources.” The Jatakas’ usual figure of five hundred pupils per teacher is “more conventional than real” in his judgment, and he put a normal class at no more than 20.

Bharata, Panini and Chanakya, as tradition has them

Britannica records that, according to the Ramayana, Bharata founded the city and named it after his son Taksha, its first ruler, and that tradition places the first recitation of the Mahabharata at Taxila, at King Janamejaya’s snake sacrifice. Two historical names are tied to the city by later tradition.

The grammarian Panini, author of the Ashtadhyayi, was born at Shalatula near Attock on the Indus, according to the University of St Andrews’ MacTutor biography, which also says the dates given for him are pure guesses and range from the 4th to the 7th century BCE. A man from Salatura “must have been” a student at Taxila, Altekar reasoned, which is an inference from geography. F. W. Thomas, in a 1944 letter printed in Marshall’s report, allowed only that Taxila’s interest in grammar might have a remote origin.

Chanakya’s link is also late. The Mahavamsa, in Wilhelm Geiger’s translation, says “the brahman Canakka” anointed Candagutta after killing the last Nanda king, and does not place him at Taxila. The Dictionary of Pali Proper Names calls Canakka “A brahmin of Takkasilā” and cites the Mahavamsa and other Pali works for his story. The same is said of Kautilya, the name on the Arthashastra, in a single sentence of Altekar’s with no text cited.

A university, or a city of independent teachers

Marshall’s own wording moved over time. In 1918 his guide said Taxila “enjoyed a great reputation as a University town”. By 1951 he still called Jataka-era Taxila “a university centre”, but wrote that the teachers were probably independent of the state and of one another, though some may have belonged to an assembly, and that there was no reason to suppose they had property, endowments, common lecture halls or libraries, apart possibly from an assembly hall. Two letters from scholars are printed in an appendix. F. W. Thomas called the so-called universities “organised groups of independent teachers” and wrote that real universities with endowed monasteries were created by Buddhism, at places such as Nalanda and Vikramashila. Lionel Barnett inferred that scholars of each branch probably formed assemblies where they were numerous enough, but said it is unknown whether the faculties ever met together.

Flatly, in 1944, Altekar wrote: “Taxila did not possess any college or university in the modern sense of the term.” Each teacher, helped by advanced students, formed an institution by himself, he wrote, and there were no degree examinations. By his date the excavations had found no large buildings that could be hostels or lecture halls. Britannica’s entry by the Lucknow historian Baij Nath Puri agrees that Taxila was “not a university town with lecture halls and residential quarters” as Nalanda was, and says the teacher housed his own pupils, who paid in cash or in service.

Some sources use the word university without hedging. Malalasekera’s dictionary itself calls Takkasila a university, and the Taxila Museum’s Jaulian claim is stronger still, though Marshall described that monastery as one “of moderate dimensions”. UNESCO’s summary says that from the 5th century BCE to the 2nd century CE Taxila was an important Buddhist centre of learning.

Taxila today

Taxila was inscribed on the World Heritage List in 1980 under criteria (iii) and (vi). The listing covers 18 archaeological sites, ten in Punjab and eight in Khyber Pakhtunkhwa. Risks named by UNESCO include industrial estates in the valley, limestone blasting and quarrying, looting at the monastery sites, and funding it calls inadequate.

In 2026 the repairs themselves came under question. Concern about the work at Mohra Moradu and Sirkap began in March, Dawn reported on 2 July 2026, when a third party sent photographs to Pakistan’s delegate at UNESCO in Paris. The agency then asked the government to reverse “reconstructions” at the two sites and warned of the danger list. Pakistan’s Department of Archaeology and Museums and the heritage ministry made a joint technical visit with UNESCO on 12 June, and UNESCO asked for the heritage impact assessment and before-and-after photographs. Punjab archaeology director general Malik Zaheer Abbas told Dawn that calling the work reconstruction was not accurate. Dawn also cited ministry sources who said UNESCO had mentioned delisting in a meeting, which came from those sources and not from a published UNESCO statement.

Pakistan Today reported on 29 July 2026 that a Senate committee had been told UNESCO had no further objections to the work at Mohra Moradu and Sirkap, after Pakistan submitted the technical documentation.