Chapter 1: The Evidence in the Ground

---

In 1989, a team from the Odishan Institute of Maritime and Southeast Asian Studies began digging at Manikapatna, a village on the coast of Chilika Lake, in Puri district. The site had been identified as a potential ancient port — the geography was right, with the lake providing a sheltered anchorage and the Satapada ghat offering access to the open sea. But what the excavation produced, over four seasons between 1989 and 1993, was not just evidence of a port. It was evidence of a world.

From the same strata, the diggers pulled Roman amphora fragments — the broken handles and rims of Mediterranean wine jars that had traveled five thousand miles to reach this shore. They found Puri-Kushana coins from the first century CE. They found Chinese celadon, the pale green stoneware that was China’s luxury export for a thousand years. They found a potsherd with a Kharosthi inscription — a script from the northwest of the subcontinent, from Gandhara and Central Asia, a thousand miles from the Odisha coast. They found rouletted ware, the distinctive Indian pottery whose concentric tooth-wheel patterns have turned up in archaeological digs from Bali to Egypt. They found an inscribed pendant reading “Sadabhu Tisha,” palaeographically dated to the Mauryan period — the third century BCE.

And they found boat nails. Iron boat nails, iron harpoons, iron spearheads, iron fish hooks. Iron slags from local metalworking. Bangles made of terracotta, faience, glass, and conch shell. Beads of agate, carnelian, bone, and soft stone. The inventory of a harbor town that manufactured, traded, repaired, and provisioned ocean-going vessels.

Here was the problem, and the revelation. A single site, on a lake shore in coastal Odisha, had yielded material culture from Rome, China, Central Asia, Sri Lanka, and Southeast Asia — spanning roughly two thousand years. The later excavation phases, reaching medieval strata, added Rajaraja Chola coins from the tenth century, Sahassamalla coins from the Polonnaruva period in Sri Lanka, Chinese coins from the fourteenth century, Yuan and Ming dynasty porcelain, and Arabic glazed pottery. When the Deccan College re-excavated in 2010, the picture only deepened. An Indo-Arabian stone anchor was found on the coast nearby; geochemical provenance analysis — petrographic, trace element, Sr-Nd isotopic — traced its basalt to a specific lava flow at Palitana in Saurashtra, Gujarat.

Manikapatna, as the excavation reports would note, is the only site on India’s entire east coast from which such a variety of ceramics from different civilizations has been recovered. Not Arikamedu, the famous Roman-trade site in Tamil Nadu. Not Tamluk, the ancient Tamralipti. Only Manikapatna.

This chapter is about what was found in the ground, carved in stone, and recorded in texts. It is the evidence chapter. Before asking what Kalinga’s maritime world meant — before tracing the cultural transfers, the civilizational impact, the modern implications — the first question is simpler and harder: what do we actually know?

---

The Ports

A coastline is not a port. A port requires a specific convergence: a natural harbor or sheltered water, access to a river system leading inland, proximity to a production hinterland, and position on a trade route that ships actually sail. The Kalinga coast — roughly seven hundred kilometers from the Subarnarekha estuary in the north to the Godavari delta in the south — had all of these, and had them in abundance. Rivers punctuated the coast at regular intervals: the Subarnarekha, the Baitarani, the Brahmani, the Mahanadi, the Daya, the Rushikulya, the Vamsadhara, the Nagavali. Each estuary was a potential port. Several were actual ones.

Tamralipti (modern Tamluk, now in West Bengal but historically part of the Kalinga trading network) was the anchor of the northern stretch. It sat at the head of the Hooghly distributary, where the river system of the Gangetic plain met the Bay of Bengal. Every Chinese pilgrim who traveled to India by sea knew Tamralipti. It was the gateway — the port where you arrived from Sri Lanka or Southeast Asia and picked up the river route inland to the Buddhist monasteries of Bihar and Bengal, or where you boarded a ship to go the other direction. Xuanzang, the seventh-century Chinese pilgrim, described it as a flourishing port. Faxian, two centuries earlier, had embarked from Tamralipti on his return voyage to China, sailing first to Sri Lanka and then east across the open ocean. The archaeological record at Tamluk includes extensive pottery assemblages consistent with long-distance maritime trade. Tamralipti was not exclusively Kalingan — it sat at the boundary between Kalinga and Vanga (Bengal), and different periods saw different political control — but it was part of the same coastal trading system, and Kalingan merchants used it as freely as Bengali ones.

Chelitalo is the port Xuanzang described in his seventh-century account — he called it Che-li-ta-lo. It was, he reported, a port from which ships sailed to Simhala (Sri Lanka) and to China. Scholars have identified it with the Manikapatna area on the Chilika coast, or with a nearby location. The identification is not certain, but the description is specific: a departure point for Bay of Bengal crossings, situated in the Kalinga region. When Xuanzang visited, circa 638-645 CE, the port was active and known to the wider maritime world. That a Chinese pilgrim — who had traveled overland across Central Asia to reach India — should specifically note this Kalingan port as a departure point for sea voyages tells you something about how integrated the port was into the broader Indian Ocean network. It was not a local fishing harbor. It was a node in a system that connected India, Sri Lanka, Southeast Asia, and China.

Palur — or Paloura, as Ptolemy rendered it — may be the most significant of all the Kalinga ports, at least from the standpoint of ancient geography. Ptolemy’s Geography, compiled around 150 CE in Alexandria, describes Paloura as an important emporium on the east coast of India. But Ptolemy adds a detail that elevates Palur from a trading port to something more: he mentions an Apheterion near the town. The word is Greek: it means “point of departure.” Specifically, it was the recognized launching point from which ships crossed the open Bay of Bengal to Chryse — the “Golden Land” — identified with Suvarnabhumi, the Sanskrit name for the gold-bearing regions of Southeast Asia.

Think about what this means. In the second century CE, a Greek geographer in Alexandria knew that ships bound for Southeast Asia departed from a specific point on the Kalinga coast. This was not secret knowledge. It was published geography, available to anyone who consulted Ptolemy’s work — the most influential geographical text of the ancient and medieval world. Palur was not just a Kalingan port. It was an international waypoint, the last Indian landfall before the transoceanic crossing.

The scholar S. Levi identified Palur with Dantapura of the Buddhist texts and Dandagula of Pliny’s Natural History. A village named Palur still exists at the Rushikulya estuary in Ganjam district. Archaeological exploration there has unearthed Chinese celadon ware, Roman rouletted pottery, and amphora fragments — the same material signatures found at Manikapatna, confirming that this was indeed a port with long-distance connections. But Palur has received nothing like the systematic excavation that Arikamedu or even Manikapatna has. It sits in the ground, largely unasked.

Pithunda enters the record through two very different sources. The Jain text Uttaradhyana Sutra describes it as an important center at the time of Mahavira, in the sixth or fifth century BCE — a place frequented by merchants from Champa (a region associated with ancient Vietnam). Kharavela’s Hathigumpha Inscription, written several centuries later, records that in his eleventh regnal year, the Kalingan king conquered Pithunda and “ploughed it with a plough yoked to asses” — a vivid humiliation — after defeating the Tramira (Tamil) confederacy. Sylvain Levi placed Pithunda south of Palur, near Chicacola and Kalingapatnam in the Srikakulam area. Its exact location remains debated, but the convergence of the Jain text and the Kalingan inscription — two independent sources, separated by centuries, both identifying Pithunda as a significant place — is more than coincidental.

Kalingapatnam, on the Vamsadhara River in what is now Srikakulam district, Andhra Pradesh, bears Kalinga’s name directly. It sits in the southern stretch of the old Kalingan territory. Archaeological evidence confirms port activities, though excavation has been limited. The name itself is the strongest evidence: a port named after Kalinga, at the mouth of a river, on the coast that Ptolemy and the Periplus described.

Dosarene appears in the Periplus of the Erythraean Sea, the anonymous first-century CE Greek navigational guide to the Indian Ocean. The Periplus lists it among the ports on the east coast of India. Ptolemy mentions it too. The exact identification is uncertain — east coast Indian ports in Greek sources are notoriously hard to pin down — but scholars place it on the Kalinga coast. The Periplus is more detailed about western Indian ports (the Malabar coast, Gujarat) than eastern ones, so the mention of Dosarene at all indicates that the Kalinga coast was known to the Mediterranean trading world, even if less thoroughly documented than the west coast.

The Paradip area, at the Mahanadi estuary — where modern Paradip Port now stands — shows archaeological evidence of Buddhist maritime activity: a ninth-century Avalokiteshvara image (a Buddhist deity associated with compassion and, significantly, with the safety of seafarers), and an eleventh-century image of the Goddess Oddiyani Marichi from Marichpur, worshipped by sailors and traders seeking safe voyages. A fifteenth-century Persian navigational tract mentions “Faradip.” The Mahanadi was one of the great rivers of Kalinga — its delta was the agricultural heartland, and its mouth was the natural location for a port connecting the interior to the sea.

The pattern across all these sites is consistent. Where a major river met the Bay of Bengal on the Kalinga coast, there was a port. Where there was a port, there is archaeological evidence of long-distance trade — pottery, coins, inscriptions from multiple civilizations. The Kalinga coast was not a wall between the interior and the ocean. It was a membrane, permeable in both directions.

---

The Pottery

Of all the material evidence for ancient Indian Ocean trade, rouletted ware may be the most useful — not because it is beautiful (it is competent but not spectacular), but because it is distinctive, widely distributed, and scientifically testable.

Rouletted ware is a fine pottery from India’s Early Historic period, roughly 500 BCE to 200 CE. It is characterized by concentric bands of rouletted impressions — marks made by pressing a toothed wheel or comb into the interior base of the vessel while the clay was still soft. The result is a pattern of regular, repeating indentations arranged in rings. It is not decorative in the usual sense. It looks more like a maker’s mark, or a quality indicator, or simply a potter’s habit. But it is consistent enough, and distinctive enough, that when you find it, you know where it came from.

And it has been found everywhere.

In Odisha: at Sisupalgarh, the ancient fortified city near Bhubaneswar whose earliest occupation dates to 804-669 BCE, where Monica L. Smith and R. Mohanty conducted systematic excavations between 2005 and 2009. At Manikapatna, alongside the Roman amphorae and Chinese celadon. At Tamralipti.

In Southeast Asia: at Sembiran and Pacung on the north coast of Bali, where excavations between 1987 and 1989 recovered 72 fine-fabric rouletted sherds from Sembiran alone. At Tra Kieu in Vietnam. At Beikthano in Myanmar. At Bukit Tengko Lembu in Malaysia. At Kobak Kendal in Indonesia. The distribution maps in the archaeological literature look like someone scattered pins across the entire Bay of Bengal and Indian Ocean littoral.

At Arikamedu in Tamil Nadu — the site Mortimer Wheeler excavated in 1945, the one that first demonstrated Roman trade with India — rouletted ware is the signature ceramic. This is the reference site, the place against which all other rouletted ware finds are compared.

And here is the critical finding: when researchers subjected rouletted ware from Sembiran (Bali), Kantarodai (Sri Lanka), and Arikamedu (Tamil Nadu) to X-ray diffraction analysis and neutron activation analysis, the results showed that all the rouletted ware came from the same production source region. The clay was Indian. The pots were made in India and carried across the ocean.

What they carried with them — or, more precisely, what they accompanied — is the real question. Rouletted ware is not itself a luxury good. Nobody was sailing two thousand miles to buy a dish with wheel marks on it. The pottery traveled as part of a larger cargo, or as the personal possessions of traders and sailors, or as containers for other goods. Its presence at a site is a signal: Indian Ocean trade happened here. Indian merchants, or goods that passed through Indian hands, reached this shore. The pot is a proxy for the network.

The Sembiran finds in Bali are particularly significant because of their dating. The analytical work by Calo and colleagues, published in Antiquity through Cambridge University Press, suggested that the links between Bali and the Indian subcontinent reached back to the late first millennium BCE — approximately 200 years earlier than previously estimated. If the dating holds, this pushes the beginning of regular India-Southeast Asia maritime contact back to at least the third or second century BCE — the period of Ashoka and Kharavela. The pottery and the inscriptions, the material evidence and the textual evidence, converge on the same era.

Ford and colleagues, in their revisiting of rouletted ware evidence published in Antiquity, and Schenk, in his study of rouletted ware in Indian Ocean exchange networks, have both emphasized that this ceramic type is not merely an Indian export. It is an indicator of a trade system — a network of maritime routes, port towns, and merchant communities that linked India’s east coast to Sri Lanka, Southeast Asia, and ultimately the Mediterranean. The rouletted ware found at Berenike in Egypt — a Roman Red Sea port — was interpreted as the personal possessions of Indian merchants or sailors who had traveled westward. The same ceramic type, found on both sides of the Indian Ocean, is the material signature of a connected world.

---

The Beads

If rouletted ware is the signature pottery of the early Indian Ocean trade, then carnelian and agate beads are the signature luxury goods. Small, durable, easily transported, and highly valued across cultures, these semi-precious stone beads were one of ancient India’s most successful exports.

The manufacturing center was primarily Gujarat — the workshops of Khambhat (Cambay) and the Deccan, where the raw material, carnelian and agate nodules from the Deccan Traps volcanic formation, could be sourced, cut, drilled, polished, and strung. The technology was sophisticated: etching techniques that produced white patterns on red carnelian, long drilling methods that created thin, elongated beads, and heat treatment that deepened the stone’s color. These were not craft goods in the casual sense. They represented accumulated technological knowledge passed through generations of specialist artisans.

The beads traveled across the Bay of Bengal. At Iron Age sites in Cambodia and Thailand, Alison Kyra Carter subjected agate and carnelian beads to LA-ICP-MS analysis — laser ablation inductively coupled plasma mass spectrometry, a technique that vaporizes a microscopic sample of the stone and reads its elemental fingerprint. The results, published in the Journal of Archaeological Science: Reports in 2016, showed that many of the beads were produced from raw material derived from the Deccan Traps of India. The stone was Indian. The craftsmanship was Indian. The beads had crossed an ocean.

At Manikapatna, beads were part of the standard assemblage — terracotta, agate, soft stone, bone — alongside the iron implements and the ceramics from multiple civilizations. The beads at Manikapatna may not have been locally manufactured from Gujarati raw material. They may have been trade goods passing through the port, or they may have been made locally from imported or locally sourced stones. The excavation reports do not resolve this question fully. But their presence confirms that Manikapatna was integrated into the same bead-trade network that linked Gujarat’s workshops to Southeast Asian burial sites.

Bellina’s research on beads, social change, and interaction between India and Southeast Asia demonstrated that long-distance exchange networks crossing the Bay of Bengal penetrated deep into Southeast Asia in the period 500-1 BCE. The beads were not merely decorative. In many Southeast Asian cultures, they served as markers of social status, as grave goods, as components of ritual exchange systems. An Indian bead arriving at a Southeast Asian port was entering a local system of meaning and value — it was being absorbed, not merely received.

A caveat is necessary. The bead trade was predominantly a Gujarat-to-Southeast-Asia phenomenon. The specific role of Kalingan bead production — as distinct from Gujarat’s dominant Khambhat industry — has not been isolated by the archaeological evidence. Kalingan ports may have served as transshipment points for Gujarati beads heading east, or they may have had their own bead manufacturing (Odisha has its own agate and carnelian sources). The evidence does not yet distinguish between these possibilities. What it does establish is that the Kalinga coast was part of the bead trade network, whether as producer, intermediary, or both.

This is a pattern that recurs throughout the evidence. The Kalinga coast sits within a larger Indian Ocean trading system. Isolating the specifically Kalingan contribution from the broader east coast Indian or pan-Indian contribution is often difficult. The pottery could have been made in Tamil Nadu. The beads could have been crafted in Gujarat. What we can say with confidence is that the Kalinga coast — Manikapatna, Palur, Tamralipti, the rest — was a node in the network, not a bystander to it. Ships called here. Goods accumulated here. The material evidence proves presence. Proving origin requires more work.

---

The Inscriptions

Pottery and beads are mute witnesses. They prove trade happened but not who traded, or why, or under what political conditions. For that, you need inscriptions — words cut into stone by people who wanted their words to last.

Hathigumpha: Kharavela Speaks

The Hathigumpha Inscription has already appeared in this series. Seventeen lines of Prakrit in Brahmi script, carved on the brow of the Elephant’s Cave at Udayagiri near Bhubaneswar, recording the regnal achievements of King Kharavela of the Mahameghavahana dynasty, sometime in the second or first century BCE.

For the maritime story, the relevant details are these. The inscription describes a powerful state — Kalinga under Kharavela — with military reach from Magadha in the north to the Tamil confederacy in the south. In the eleventh regnal year, Kharavela conquered Pithunda, a port identified by the Jain Uttaradhyana Sutra as a place frequented by merchants from Champa. In the fifth year, he extended a canal originally built by the Nanda kings three centuries earlier — infrastructure that connected the agricultural interior to the coast. The inscription references a “catumga” army (fourfold: infantry, cavalry, elephants, chariots) and implies a state with the organizational capacity to project power along the coastline and into the sea.

The inscription does not explicitly describe overseas trade. It does not mention boitas or sadhabas or Southeast Asia. But it describes the conditions under which maritime trade flourishes: a powerful state controlling the coast, with infrastructure connecting river and sea, military dominance over rival coastal powers, and a port (Pithunda) already known to merchants from Southeast Asian-linked regions.

A warning about the Hathigumpha Inscription is essential. The stone is heavily damaged. Large portions of the text are reconstructed from partial letter forms, and different epigraphists — K.P. Jayaswal, R.D. Banerji, and their successors — have produced significantly different readings. Some of the more dramatic claims attributed to Kharavela depend on specific readings that other scholars dispute. The inscription is real and important, but it should be treated as partially legible testimony, not as a clear document.

Ashoka’s Kalinga Edicts: The Context of Conquest

The Dhauli and Jaugada edicts — the Separate Kalinga Edicts of Ashoka, carved approximately 261 BCE — do not directly address maritime trade. Their relevance is contextual. They establish that Kalinga was important enough to the Mauryan Empire to justify a war of conquest, and that the territory was large enough to require two separate administrative centers (one for northern Kalinga at Dhauli, one for southern Kalinga at Jaugada). They confirm that Kalinga was, in the mid-third century BCE, an independent, organized, and militarily capable state — the kind of state that can sustain international trade.

The edicts also tell us something about the aftermath. Ashoka’s edicts address the mahamatras (administrators) of Kalinga and instruct them to govern with justice and fairness. The implication is that Kalinga was being integrated into the Mauryan administrative system — which, among other things, would have connected its ports to the wider Mauryan commercial infrastructure. Whether Mauryan control enhanced or diminished Kalingan maritime trade is an open question. It could have done either, or both at different times.

Southeast Asian Inscriptions: Kalinga’s Name Abroad

The most striking evidence that Kalinga’s reach extended to Southeast Asia comes not from Indian soil but from Southeast Asian inscriptions that invoke the Kalinga name directly.

The Kalingga kingdom of Java. Between the sixth and eighth centuries CE, a kingdom called Kalingga (also rendered Ho-ling or Heling in Chinese records) existed on the north coast of Central Java. The name is not a coincidence or a linguistic accident. Chinese Tang dynasty records note: “Ka-ling is also called Djava, it is situated in the southern ocean, at the east of Sumatra and at the west of Bali.” The kingdom was substantial enough to appear in Chinese diplomatic records and to produce its own inscriptions.

Two inscriptions mark its existence. The Tukmas inscription, discovered on the western slope of Mount Merapi, Dusun Dakawu, Lebak village, Magelang Regency, Central Java, is written in Pallava script in Sanskrit — a Southeast Asian inscription using Indian script and language. The Sojomerto inscription, discovered in Sojomerto village, Batang Regency, Central Java, is written in Kavi script in Old Malay and dated to the seventh century CE. The kingdom was ruled, at its peak, by Queen Shima (Ratu Sima), who brought it to its greatest extent before her death around 695 CE, after which it weakened and was eventually absorbed by Srivijaya.

A kingdom in Java named after a kingdom in India. The naming is the evidence. It is as if someone founded a city in Southeast Asia and called it “New York” — the name itself records the origin of the founders or the cultural model they emulated.

The Vo-Chanh Rock Inscription in Vietnam. Dated to the second or third century CE, this Sanskrit inscription refers to the first kingdom of Champa under the royal family of Sri Mara, identified as being of Kalingan origin. If the identification holds — and it is accepted by a significant number of scholars, including R.C. Majumdar, who devoted much of his career to ancient Indian connections with Southeast Asia — then the founding dynasty of one of Vietnam’s most significant ancient civilizations traced its ancestry to Kalinga.

The Champa connection runs deeper than a single inscription. Approximately 400 Champa inscriptions have been found, in Cham, Sanskrit, and Arabic, across the kingdom’s cultural centers: Vijaya (Binh Dinh), Kauthara (Nha Trang), Panduranga (Phan Rang), Indrapura (Dong Duong), Amaravati (Quang Nam). The Cham script descends from the South Indian Brahmic Grantha script. Fourth-century CE inscriptions of King Bhadravarman — the Cho Dinh and Hon-Cut inscriptions — show script similarity to Kalingan epigraphic traditions. Champa adopted Sanskrit as its scholarly language and Hinduism, especially Shaivism, as its state religion. The kingdom lasted for over a millennium, from roughly the second century CE to the fifteenth, and its cultural DNA was, in significant part, Indian — and arguably, in its earliest phase, specifically Kalingan.

A note of scholarly caution. The Vo-Chanh inscription’s identification of the Sri Mara dynasty as Kalingan is not universally accepted. The inscription is fragmentary, and the “Kalingan origin” reading depends on specific interpretive choices. Some scholars argue the connection is overstated. Others, including the Indian Council of World Affairs in its work on India-Vietnam civilizational linkages, accept it as credible. The evidence is suggestive, not conclusive.

The Shailendra dynasty theory — R.C. Majumdar’s 1933 proposal that the powerful Shailendra dynasty of the Indonesian archipelago originated from Kalinga — deserves mention precisely because it illustrates the boundary between evidence and aspiration. The Shailendras built Borobudur, one of the greatest Buddhist monuments on earth. If they were Kalingan in origin, the implications would be enormous. But modern scholars have largely rejected the theory. The evidence does not support it, and the desire for it to be true — the desire for Kalinga to be the origin of Borobudur — has sometimes outrun the archaeology. This is worth being honest about. Not every connection that is claimed is a connection that is proven.

---

The Texts

Inscriptions are carved in stone and survive by accident of weather and neglect. Texts — written on palm leaf, paper, or parchment — survive by the more precarious mechanism of being copied, translated, and valued by successive generations. Several categories of texts illuminate the Kalinga maritime world.

The Jataka Tales

The Jatakas are Buddhist birth stories, compiled from roughly the fourth century BCE onward, recounting the previous lives of the Buddha. They are literature, not history. But they are literature that reflects a world, and the world they reflect is one saturated with maritime trade.

The Baveru Jataka (No. 339) describes Indian merchants exporting peacocks to Baveru — identified with Babylon, Mesopotamia. The detail is vivid: the merchants discover that the Babylonians have never seen a peacock and are astonished by the bird. The story establishes that Indian trading voyages to the Persian Gulf were a known, familiar phenomenon by the time the tale was composed. The Suppuraka Jataka (No. 463) describes the adventures of a master mariner named Suppuraka — maritime navigation as a recognized skill, the sea as a known if dangerous space, the merchant-sailor as a character type that an audience would find plausible.

Multiple Jatakas mention Suvarnadvipa (Southeast Asia, the “Golden Island”), Ratnadvipa (Sri Lanka, the “Gem Island”), and Baveru (Babylon). The Kalinga-Bodhi Jataka (No. 479) is set in the Kalinga kingdom itself. The cumulative picture is of a culture in which long-distance maritime trade — to the west and to the east — was so familiar that it served as a natural setting for moral parables.

The Jatakas are not evidence of specific voyages. They do not name ships, or ports, or dates. They are evidence of a culture for which maritime trade was a structural fact of life — unremarkable enough to serve as the backdrop for stories about other things entirely. That is, in some ways, stronger evidence than a shipping log would be. A culture records its routines in its stories.

The Periplus of the Erythraean Sea

The Periplus, a first-century CE Greek navigational and commercial guide, is a working document written for traders, not scholars. It describes the ports, goods, and sailing conditions of the Red Sea, East Africa, the Persian Gulf, the Arabian Sea, and the Indian coast. Its author — anonymous, probably an Egyptian Greek merchant — had either sailed these routes himself or compiled his information from those who had.

The east coast of India receives less attention than the west coast in the Periplus. Western Indian ports — Barygaza (Bharuch), Muziris (on the Malabar coast), Nelcynda — are described in commercial detail: what goods are available, what prices are fair, what political conditions apply. For the east coast, the information is sparser. But Dosarene is mentioned as a trading location on what scholars identify as the Kalinga coast. The Periplus describes the Indian east coast as producing fine textiles, spices, and precious stones — all goods consistent with what Kalinga is known to have traded.

The comparative sparseness of the Periplus on Kalinga ports is itself informative. The document was written from a Mediterranean perspective, by someone whose trade routes ran primarily from Egypt to western India. The east coast trade — India to Southeast Asia — was a different network, run by Indian merchants sailing routes that Greek and Roman traders did not frequent. The Periplus glimpses the Kalinga coast from the outside, as a place that existed in a trading world the author was aware of but not deeply familiar with.

Ptolemy’s Geography

Ptolemy’s Geography (c. 150 CE) has already been discussed in the context of Palur and the Apheterion. Its significance for the Kalinga maritime story cannot be overstated. Ptolemy was not a traveler; he was a compiler, working in Alexandria from reports, sailing directions, and earlier geographical works. His identification of Paloura as an important emporium and his mention of the Apheterion — the departure point for ships sailing to Chryse (Suvarnabhumi, Southeast Asia) — represent the accumulated geographical knowledge of the second-century Mediterranean world about the eastern Indian coast.

Ptolemy mentions other points on the Kalinga coast — areas near Konark, Puri, and Balasore (Kosambi). The detail is not exhaustive, but it is specific enough to show that the Kalinga coast was mapped, at least approximately, in the most authoritative geographical work of the ancient West. Ships sailed from here to Southeast Asia. Ptolemy knew it. His readers in Alexandria knew it.

The navigation implication is important. During the early centuries of the Common Era, ships crossing the Bay of Bengal did not typically sail directly from Sri Lanka to Southeast Asia. Instead, they followed the coast northward from Sri Lanka to the Kalinga coast — to Palur, to Chelitalo — and from there made the open-ocean crossing to Suvarnabhumi. The Kalinga coast was not merely a place where goods were produced. It was a waypoint — a place where ships stopped, provisioned, and waited for favorable conditions before attempting the most dangerous leg of the voyage. This made Kalingan ports critical to the entire Bay of Bengal trading system, even for trade that originated elsewhere.

The Chinese Pilgrims

Three Chinese Buddhist pilgrims left records that illuminate the Kalinga maritime world, not as traders but as travelers who used the same routes and ports.

Faxian (c. 337-422 CE) traveled to India around 399 CE and returned by sea, sailing from Tamralipti to Sri Lanka and then east across the Bay of Bengal. His account is the earliest firsthand Chinese description of the maritime route that connected India to China via Southeast Asia. He describes the conditions of the voyage — storms, navigation by stars, the terrifying open-ocean crossing — with the detail of someone who experienced them.

Xuanzang (c. 602-664 CE) is the most important of the three for the Kalinga story. He visited Odisha around 638 CE and described the Buddhist centers of the region. He noted the port of Chelitalo as a departure point for ships sailing to Sri Lanka and China. He visited Pushpagiri, a Buddhist establishment that scholars have increasingly identified with the Diamond Triangle sites in Jajpur district. His account of Pushpagiri describes a still-functioning Buddhist center with monks, students, and connections to the wider Buddhist world.

Yijing (635-713 CE) traveled to India by sea. His records contain a detail that directly connects Kalinga to Java: in 664 CE, a Chinese Buddhist monk named Huining arrived in Heling (Ho-ling) — the Kalingga kingdom of Java — and stayed for about three years. With the assistance of Jnanabhadra, a Heling monk, Huining translated numerous Hinayana Buddhist scriptures. The Kalingga kingdom was not merely an Indianized political entity. It was a functioning Buddhist center, with monks who could collaborate with Chinese scholars on scripture translation. The exchange was not abstract. It was specific: a Chinese monk, a Javanese monk, a Kalingan-named kingdom, and the production of Buddhist texts.

The Chinese celadon and porcelain found at Manikapatna corroborate the textual record. The texts describe monks traveling the routes. The ceramics prove that goods traveled the same routes. Text and material evidence converge.

Tamil Texts

The Kalingattuparani, a twelfth-century Tamil war poem by Jayamkondar, celebrates the Chola invasion of Kalinga under Kulottunga Chola I. It is a parani — a genre of victory poem — and its value is not in maritime evidence directly but in confirming that Kalinga remained a significant, named kingdom in the Tamil political imagination into the medieval period. Kalinga was not a forgotten backwater. It was a rival worth celebrating a victory over.

The Sangam literature (c. third century BCE to third century CE), the earliest stratum of Tamil literary production, describes busy maritime ports on the east coast. The Pattinappalai’s 301 lines on the Chola port of Puhar (Kaveripoompattinam) depict a city of large ships, fishermen, markets, and festivals — a port culture that existed alongside, and interacted with, the Kalingan port culture further north. The Hathigumpha Inscription, from the Kalingan side, mentions the “Tramira” (Tamil) confederacy. The picture from both Tamil and Kalingan sources is of two maritime cultures on the same coast, sometimes cooperating, sometimes warring, always aware of each other.

---

The Buddhist Connection Sites

Buddhism traveled the same routes as cotton and carnelian. Monks sailed on the same ships as merchants. And in Odisha, the institutional infrastructure of Buddhism — the great monasteries and universities — was concentrated in a cluster of sites that scholars call the Diamond Triangle.

Ratnagiri, Lalitgiri, and Udayagiri — three sites in Jajpur district, not to be confused with Kharavela’s Udayagiri-Khandagiri caves near Bhubaneswar — form a complex of Buddhist monasteries that were active from the early centuries of the Common Era through the twelfth century. Their peak period was the seventh to tenth centuries CE, when Ratnagiri in particular rivaled Nalanda as a center of Buddhist learning.

Ratnagiri is the largest. Excavations have revealed monumental mahaviharas (great monasteries), votive stupas, sculpted Buddhas and Bodhisattvas of extraordinary quality. The site was a Vajrayana center — the esoteric, tantric form of Buddhism that would later be transmitted to Tibet and Indonesia. This is significant for the Southeast Asian connection because Vajrayana Buddhism had documented exchange networks reaching from Odisha to the Indonesian archipelago. The Buddhist art of Ratnagiri shows stylistic connections with both Pala Bengal and Southeast Asian Buddhist art, confirming that this was not an isolated monastery but a node in an international network of Buddhist institutions.

Lalitgiri is among the earliest Buddhist sites in Odisha. An important relic casket was discovered here. Udayagiri is a large monastic complex. Together, the three sites represent the institutional foundation of Odisha’s Buddhist tradition — the places from which monks trained, studied, and traveled outward.

The Chinese pilgrims connect these sites to the wider world. Xuanzang’s visit to Pushpagiri (increasingly identified with sites in or near the Diamond Triangle) in 639 CE confirms that these institutions were known to the international Buddhist community. Monks and scholars from India, Sri Lanka, Nepal, China, and Southeast Asia visited them. The flow was not one-directional. Buddhist knowledge moved outward from these sites to Southeast Asia and beyond, but ideas, students, and even Buddhist texts moved inward as well.

The ASI (Archaeological Survey of India) has been conducting ongoing excavations at Ratnagiri from 2024 through the present, unearthing new artifacts that continue to deepen the picture. The three sites are on the UNESCO Tentative World Heritage List as a serial nomination. If the nomination succeeds, it would bring international recognition to what has long been an underappreciated chapter in the history of global Buddhism.

The connection between the Diamond Triangle and the maritime trade routes is structural, not metaphorical. Buddhist monasteries needed patronage. Maritime merchants provided it. The merchants needed the social credibility and cosmic legitimacy that religious association provided. The monks needed the trade routes to spread the dharma. The relationship was symbiotic. When monasteries flourished, trade routes were well-maintained. When trade declined, monasteries lost patronage and shrank. The archaeology at Ratnagiri tracks this relationship: the site’s peak period coincides with the peak of Bay of Bengal trade, and its decline in the twelfth century coincides with the broader disruption of Indian Ocean trading systems.

---

What We Don’t Know

Honesty requires a section on the gaps.

No Kalingan ship has been found. The boita — the ocean-going vessel that the Bali Jatra festival commemorates, that the sadhabas sailed to Southeast Asia — exists in literary descriptions, in ritual memory, and in the iron boat nails recovered from Manikapatna. It does not exist as a recovered hull. No shipwreck specifically attributable to a Kalingan vessel has been identified. The Godavaya shipwreck off Sri Lanka’s southern coast (second century BCE to first century CE), with its cargo of Indian iron ingots and glass, demonstrates that Indian ships sailed and sometimes sank. But it is not a Kalingan ship. We know the boats existed because we know the trade happened and because we find the boat nails. But we do not have the boat itself. This is a significant gap. Without hull remains, we cannot know how boitas were constructed, how large they were, what their sailing characteristics were, or how they compared to other vessels of the period.

Much of the Kalinga coast is unexcavated. Manikapatna received serious archaeological attention in 1989-1993 and again in 2010. Sisupalgarh has been systematically excavated. But the vast majority of potential port sites along the seven-hundred-kilometer Kalinga coast have received little or no professional excavation. Palur — which Ptolemy identified as the departure point for Southeast Asia — has been explored but not systematically excavated. Kalingapatnam, Pithunda, Dosarene, and other identified or probable port sites remain largely in the ground. The evidence base could expand dramatically with sustained excavation, but sustained excavation requires funding, institutional will, and decades of patient work. What we have is what has been dug. What has been dug is a fraction of what exists.

Dating is often imprecise. Archaeological dating depends on a combination of stratigraphy (the layer a find comes from), typology (comparison with similar finds from dated contexts), and absolute dating methods (C14 radiocarbon dating, thermoluminescence, etc.). For many Kalingan finds, the available dating is typological rather than absolute — the pottery “looks like” second-century-BCE rouletted ware because it resembles rouletted ware from dated contexts elsewhere. This is standard archaeological practice, but it introduces uncertainty ranges of decades or even centuries. The Sisupalgarh excavations, with their C14 dates, represent the gold standard. Most other Kalingan sites do not have comparable absolute dating.

Distinguishing “Kalingan” from “east coast Indian” trade is difficult. This is perhaps the most important caveat. Rouletted ware was produced somewhere on the Indian east coast — possibly Tamil Nadu, possibly the Deccan, possibly Kalinga itself. The exact production centers are still debated. Geochemical sourcing studies have shown a common origin for rouletted ware found across the Indian Ocean, but that origin has not been pinpointed to a specific region of India, let alone to Kalinga specifically. The beads are largely Gujarati in origin. The coins found at Kalingan ports come from all over India and beyond.

What this means is that the Kalinga coast was demonstrably part of a major Indian Ocean trade network, but its specific contribution — as producer, as transshipper, as middleman, as origin point for specific goods — is harder to isolate than the popular narrative suggests. The claim “Kalinga traded with Southeast Asia” is well-established. The claim “Kalinga was the primary Indian connection to Southeast Asia” is much harder to prove and probably overstates the case. Tamil Nadu, Bengal, Gujarat, and other Indian regions were equally engaged. Kalinga was important. It was not the only node. The danger in telling this story is the temptation to claim more than the evidence supports — to make Kalinga the protagonist of a story that had many protagonists.

The “Indianization” debate reframes everything. The older scholarly model, associated with George Coedes and R.C. Majumdar, described the spread of Indian culture to Southeast Asia as a one-directional process: Indians brought civilization to a passive Southeast Asia. Modern scholarship, beginning with O.W. Wolters’ “localization” model and continuing through Sheldon Pollock’s “Sanskrit cosmopolis” framework, has fundamentally revised this picture. Southeast Asian societies were active participants. They selected which Indian cultural elements to adopt and which to reject. They adapted what they adopted. The process was bilateral, not imperial. When Kalingan merchants arrived in Java or Vietnam, they did not impose their culture on blank slates. They entered existing societies that chose, for their own reasons, to engage with Indian religious, linguistic, and artistic traditions.

This matters because it changes the credit structure. Kalinga did not “civilize” Southeast Asia. Kalingan traders, monks, and cultural elements were part of a larger exchange that Southeast Asian societies actively shaped. The evidence in the ground supports connection, exchange, and mutual influence. It does not support cultural conquest.

---

What the evidence in the ground does support, taken together, is substantial enough without inflation. A coastline studded with ports. Pottery from India in Bali, Indonesia, Vietnam, Myanmar, Malaysia, Sri Lanka, and Egypt. Beads from India in Cambodian and Thai burial sites. Inscriptions invoking Kalinga’s name in Java and Vietnam. Chinese pilgrims noting Kalingan ports as departure points for the Bay of Bengal crossing. A Greek geographer in Alexandria marking Palur as the recognized launching point for Southeast Asia. A single excavation site on the Chilika coast yielding material culture from Rome, China, Central Asia, Sri Lanka, and the wider Indian Ocean world.

The evidence is distributed, partial, and in many cases awaiting more excavation, better dating, and sharper provenance analysis. But its collective weight is unambiguous. The Kalinga coast was not a backwater. It was not a stretch of sand between the Mahanadi and the Godavari where fishermen launched their boats and nothing more happened. It was one of the nodes — not the only node, but one of the important ones — in a trading system that connected the ancient world across the greatest ocean on earth.

The pottery in the ground at Manikapatna came from everywhere. And goods from the Kalinga coast went everywhere in return. The next chapter traces what those goods were, how the trading system actually worked, and what the monsoon cycle meant for the economy of a civilization built on the rhythm of the wind.

---

Next: Chapter 2 — The Goods and the Routes