Swaveda

The Grammar of Hegemony: How Line 22 of the Allahabad Pillar Maps the Gupta Frontier

Line 22 of the Allahabad Pillar inscription uses precise Sanskrit compounds to define imperial outer limits, distinguishing tributary border realms like Kāmarūpa and Ḍavāka from directly ruled Gupta territory.

Meera Iyer for SwavedaNovember 2, 2026

Photo: Dr. Raju Kasambe, CC BY-SA 4.0, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File%3AKanheri_Budhist_Caves_Mumbai_by_Dr_Raju_Kasambe_DSCF9917_%283%29_01.jpg)
Want to correct something or add a source? Sign in below to contribute.

The 35-foot polished sandstone shaft standing inside the fort at Prayagraj bears a layered epigraphic history. Originally erected in the third century BCE to carry the edicts of the Mauryan emperor Aśoka, the Allahabad Pillar was reused in the fourth century CE by the Gupta court poet Harisena. Writing in classical Sanskrit prose and verse, Harisena composed a praśasti (panegyric) celebrating the reign of Samudragupta.

For epigraphists and legal historians, the core of the document lies in its grammatical distinctions. Harisena did not describe imperial conquest with a single, sweeping verb. Instead, he categorized conquered and subordinate territories through a hierarchy of political relationships, using specific Sanskrit compound terms to map the political geography of northern India.

The precise boundary of that administrative system is laid out in Line 22.

The Syntax of Subjugation

In Line 22 of the inscription, Harisena lists five eastern and northern border realms alongside nine republican tribal territories. The compound term introducing the monarchical border states reads:

समतट-डवाक-कामरूप-नेपाल-कर्तृपुर-प्रत्यन्त-नृपतिभिः samataṭa-ḍavāka-kāmarūpa-nepāla-kartṛpura-pratyanta-nṛpatibhiḥ

In plain English, the phrase translates to "by the border kings of Samataṭa, Ḍavāka, Kāmarūpa, Nepāla, and Kartṛpura".

The primary philological weight rests on the compound pratyanta-nṛpati (pratyanta, meaning border or peripheral marches, combined with nṛpati, meaning king or sovereign leader). In classical Sanskrit administrative syntax, a pratyanta-nṛpati was not a governor appointed by the imperial core, nor was he a monarch whose territory had been integrated into the bhukti (imperial province) network.

Harisena separates these peripheral sovereigns from the rulers of the Gangetic heartland, whom Samudragupta dealt with via prasabhoddharaṇa (violent extermination or total uprooting). The heartland kings lost their thrones, and their lands were directly annexed into imperial administrative units. By contrast, the pratyanta-nṛpati retained their local crowns.

The nature of their submission is defined in the compound that follows in Lines 22–23:

सर्व-कर-दान-आज्ञा-करण-प्रणामागमन-परितोषित-प्रचण्ड-शासनस्य sarva-kara-dāna-ājñā-karaṇa-praṇāmāgamana-paritoṣita-pracaṇḍa-śāsanasya

This compound spells out three specific obligations required of border rulers: sarva-kara-dāna (the payment of all required taxes or tributes), ājñā-karaṇa (the execution of royal imperial commands), and praṇāmāgamana (personally visiting the imperial court to render homage).

Through this syntax, evidence shows that the Gupta court recognized a distinct legal category for peripheral states: political entities that preserved internal autonomy while acknowledging external suzerainty through tribute and ceremonial court submission.

Mapping the Periphery

Among the five border polities named in Line 22, two point directly to the eastern Brahmaputra river basin: Kāmarūpa and Ḍavāka.

The mention of Kāmarūpa in the Allahabad inscription represents the earliest confirmed epigraphic occurrence of the name in historical records. Modern historical geography identifies Kāmarūpa with the central Brahmaputra Valley of Assam, centered around ancient Pragjyotishpura (near modern Guwahati).

Adjacent to Kāmarūpa stood Ḍavāka. Epigraphists including J.F. Fleet originally speculated that Ḍavāka corresponded to modern Dhaka. Subsequent historical analysis by N.K. Bhattasali re-anchored Ḍavāka to the Kopili-Jamuna river valley in Assam's modern Nagaon district, where the town of Doboka preserves the ancient place-name.

Scholars debate the exact duration of Ḍavāka as an independent polity. Archaeological evidence indicates that Ḍavāka operated as a flourishing riverine kingdom during the early centuries of the Common Era. By the fifth century CE, Ḍavāka was absorbed into the expanding Varman dynasty of Kāmarūpa, consolidating the central Brahmaputra valley under a single regional kingdom.

The presence of both Kāmarūpa and Ḍavāka as distinct entities in Line 22 confirms that in the middle of the fourth century CE, the Gupta empire faced a fragmented political landscape along its far eastern frontier.

Archaeological Realities on the Kapili River

The epigraphic distinction between direct imperial territory and peripheral tributary states is supported by archaeological surveys in Assam's Nagaon district.

Archaeological excavations led by the Directorate of Archaeology, Assam—including work at the Harlong Kuruwati site in Kampur—have documented stone temple plinths (adhiṣṭhāna, base foundation) and structural remains along the Kapili river valley. While structural stone temple remains recorded at these sites date primarily to post-Gupta centuries (roughly the tenth to twelfth centuries CE), regional artifact assemblages, including early brick alignments and chance finds of metalwork, reflect long-term historical settlement patterns in the ancient Ḍavāka region.

Crucially, the archaeological material from the Kapili Valley does not show the administrative footprint of direct Gupta imperial rule. Sites in the central imperial Gangetic plain feature standardized Gupta-era sealings, imperial coinage, and uniform administrative brick architecture. In contrast, four-century CE sites in the Brahmaputra and Kapili valleys display distinct regional material cultures that adopted selective cultural styles from the Gangetic plain while maintaining local artistic and political structures.

This archaeological reality aligns with Harisena’s deliberate grammatical framing. By classifying the kings of Kāmarūpa and Ḍavāka as pratyanta-nṛpati, the imperial chancellery admitted in stone what field survey confirms: the Guptas did not station garrisons or send provincial governors to the Brahmaputra Valley. Imperial power on the outer edge was maintained not through direct territorial occupation, but through formal tribute payments (kara-dāna) and obligatory diplomatic visits (praṇāmāgamana) to the court at Pataliputra.

Contribute to this article

Spotted an error, want to add a source, or have a correction? Sign in to send a contribution. Submissions are evaluated for factual accuracy before they change the article. Contrarian views are preserved publicly as reader notes.