Arikamedu's Roman Pottery and a Coin Record That Doesn't Add Up

Walk the excavated mounds at Arikamedu today, near Puducherry, and you are standing on one of the most argued-over sites in Indian Ocean archaeology. The pottery here is not in dispute. What is in dispute — or at least, what deserves more scrutiny than it usually gets — is what the coins found nearby, and not found here, are trying to tell us.

Arikamedu first drew serious attention when Mortimer Wheeler excavated the site in the late 1940s, identifying what he read as a Roman "trading station" on the Coromandel coast per the Archaeological Survey of India's site records. Wheeler's interpretation leaned heavily on a distinctive ceramic type now called rouletted ware — flat-based dishes with concentric rows of stamped decoration on the interior base — alongside fragments of amphorae (Latin for the large, two-handled jars that Mediterranean traders used to ship wine and olive oil; singular: amphora). Subsequent excavations, including the long-term work led by Vimala Begley starting in the 1980s, refined and in places revised Wheeler's chronology, pushing back the site's earliest occupation and complicating the neat "Roman trading post" story into something messier and more locally rooted per Vimala Begley's excavation reports published through the University of Pennsylvania Museum.

That messiness is the point. Rouletted ware is not, strictly, proof of Roman presence at all. Archaeologists now generally treat it as a ware produced somewhere in the Bay of Bengal region — possibly eastern India — and traded widely across South and Southeast Asia, showing up from Sri Lanka to the Mekong delta per research summarized in the Journal of Archaeological Science. Its design vocabulary echoes Mediterranean tableware, but the clay fabric and firing point to local or regional manufacture. So when you find rouletted ware at Arikamedu, you are seeing evidence of a trade network that touched Mediterranean forms and tastes — not necessarily evidence that Roman ships tied up at the jetty.

The amphora sherds are a different matter. These are harder to explain away as imitation. Amphora fragments recovered at Arikamedu have been associated with wine and oil transport typical of Mediterranean shipping in the early centuries CE, and their presence, alongside other imported fine wares, is the strongest single argument for direct or near-direct maritime contact between the Roman-linked trading world and this stretch of the Tamil coast per the site excavation summaries cited by UNESCO's Silk Roads programme.

Where the coins complicate the picture

Here is the puzzle. If Arikamedu were the primary node of Roman contact on this coast — the port the Periplus of the Erythraean Sea (Greek for "Voyage Around the Red Sea," a first-century sailing manual describing Indian Ocean trade routes) gestures toward when it mentions a market town called Poduke on this coast per the Wilfred Schoff English translation of the Periplus — you might expect it to be rich in Roman coin finds. It is not, comparatively. Tamil Nadu's Roman coin hoards, the assemblages that get cited again and again in discussions of Indo-Roman trade, cluster more heavily at inland sites — Karur foremost among them, along with other locations in the Kaveri (Cauvery) river valley — rather than at the coastal port itself per catalogue data discussed in Roman India coin-hoard surveys published by the Royal Numismatic Society.

This is not a new observation among archaeologists working on the region, but it is one that rarely gets walked through for a general reader. A port that handled sustained Mediterranean shipping traffic, evidenced by pottery, ought on a simple model to accumulate coin loss and coin hoarding at roughly the same rate as the inland markets it fed. Instead, the coin record and the pottery record point in different directions: pottery says "coastal contact," coin concentration says "inland exchange and redistribution."

There are several ways to read this gap, and it is worth being honest that this is where evidence gives way to informed argument rather than settled fact. One possibility is that Roman coins moving inland were being used less as currency and more as bullion — raw metal melted down or hoarded by inland elites rather than staying in circulation at the port of entry — which would explain why ports show fewer coins even where trade goods are abundant. Another is that coastal sites like Arikamedu saw more coin recycling or restriking into local issues, erasing the archaeological signature of Roman currency even where the metal itself persisted. A third possibility, favored by scholars skeptical of overreading the "Roman" label, is that much of what reached Arikamedu came through intermediary trade — goods and forms moving hand to hand along the coast and up rivers — without requiring a Roman ship, or a Roman coin, ever to touch the site directly. These three readings are debated among numismatists and archaeologists working on Indo-Roman contact, and no consensus has settled the question.

None of this undoes the case that Arikamedu was plugged into a trade world that included Mediterranean commodities and Mediterranean-influenced tableware. The amphora sherds are real, the site's occupation across multiple centuries is real, and its mention in ancient geographic sources is a reasonable if not airtight identification. What the coin gap does is discipline the story. It is tempting, writing about Roman contact with India, to reach for a single triumphant port and call it done. Arikamedu resists that. The pottery says the Mediterranean world reached this coast. The coins, thin on the ground here and thick further inland, say the mechanics of that contact — who carried the money, who kept it, and where it ended up buried — are still an open question, and one that better excavation data, rather than better rhetoric, will eventually have to settle.