The community in question
The South Asian community in Madagascar is known as the Karana. Their origins lie in Kutch and Surat in Gujarat, with arrival largely in the late 19th and early 20th centuries via the Mozambique Channel trading network that connected the Gujarati ports to the East African coast. Linguistically the community is Gujarati-speaking; religiously it is predominantly Sunni, Ismaili Khoja and Dawoodi Bohra Muslim. None of these markers are Tamil.
Why this matters here
An earlier, less careful version of a "global Tamil diaspora" map might have included Madagascar simply because the Karana are South Asian and the island is in the Indian Ocean. The Ecumene refuses that collapse. "South Asian" is not "Tamil" — treating it as such replicates the colonial archival habit of filing every brown-skinned merchant under a single "Indian" category, and erases the specific Gujarati Muslim history of the Karana, the specific Tamil Hindu history of the Jaffna kindred, and the specific Tamil Shaiva mercantile history of the Chettiar network all in one sweep.
External record
- UNHCR — "Madagascar's Karana people still awaiting nationality." Statelessness reporting documenting the Karana as a distinct Gujarati-origin community.
- Tilburg Law Review 19(1–2) 2014 — McInerney, "Karana statelessness in Madagascar." Peer-reviewed legal-anthropological treatment that consistently identifies the community as Gujarati-origin, predominantly Muslim, and distinct from any Tamil migration cohort.
- No counter-source. No peer-reviewed academic publication, no UNESCO body, no national statistics office, and no Tamil diaspora institution classifies the Karana of Madagascar as a Tamil community. The absence of a counter-claim is itself part of the record.
What this exclusion is, and is not
- Is: a methodological refusal to widen the "Tamil" category by ethnic proxy. The Ecumene admits a node only when Tamil-specific custodianship is attested in cited sources.
- Is not: a denial of the Karana community's own civilisational history, or of Tamil presence anywhere in the wider East African region. If Tamil-specific custodianship in Madagascar is later documented in Tier-A sources, that node can be added.
- Is not: a comment on the rights or statelessness situation of the Karana. That is a separate question handled by UNHCR and Madagascar's own civil society; TLTE has no standing on it.
Five real threads, none load-bearing
Even after the Karana exclusion is stated, readers keep reaching for Madagascar. The pull is not irrational. Five real threads keep it alive — none of them load-bearing enough to make Madagascar an Ecumene node. Naming them here is itself part of the methodology; an unnamed intuition keeps looping, a named one settles.
- Indian Ocean trade memory. Chola and later Tamil mercantile networks moved across the western Indian Ocean rim — East Africa, the Comoros, the Mascarenes. Madagascar sits inside that rim geographically, so the map intuitively reaches for it. The trade memory is real; the settled Tamil community on Madagascar is not.
- Austronesian substrate confusion. Malagasy is an Austronesian language (Barito family, Borneo origin). Pseudoscience occasionally folds this into "lost Tamil / Kumari Kandam" framings. The Atlas refuses that collapse — the organ is called the Ecumene precisely because Kumari Kandam is not used as a literal geography here.
- Réunion and Mauritius adjacency. The two confirmed Tamil indenture nodes nearest Madagascar are Réunion and Mauritius. Estate-labour and Hindu festival circuits did touch Madagascar's east coast in small numbers via those islands. Real, but diffuse — never a settled Tamil community of the scale the Atlas admits.
- Karana ≠ Tamil, but they are South Asian Muslim. "South Asian Muslim trading community in the Indian Ocean" can sound like it overlaps with Tamil Muslims of the Eastern Province. It does not — different language (Gujarati, not Tamil), different origin port (Kutch and Surat, not Kayalpattinam), different denomination cluster.
- A small contemporary Hindu Tamil professional presence in Antananarivo. Real, but uncited at scale, and not by any standard the rest of the Atlas uses. If Tier-A documentation later attests Tamil-specific custodianship in Madagascar, the node can be added; until then it cannot.
The exclusion is the finding. The recurrence is the method. Both are kept on this page so neither is hidden.
If you came here looking for it
If you reached this page expecting to see Madagascar on the Ecumene atlas — fair enough; the question gets asked often. The answer is: not here, not because no South Asians live there, but because the South Asians who live there are not Tamil. The right place to read about the Karana is UNHCR's statelessness work and the Tilburg Law Review record above. The right place to read about Tamil presence in the Indian Ocean rim is the rest of this organ.