regions, and throughout the Sirana Valley and in the Mymensingh District a kind of
mixture of Bengali and Tibeto-Burman called Haijong is also spoken hy low-caste tribes.
In the centre of the delta, in the Districts of Ehulna, Jessore, and Earidpur, the language
is in a transition stage. We see the standard dialect of Central Bengal gradually merging
into the dialect of Dacca, and, if it is desired, we can class the speech of these Districts as
a sub-dialect of Eastern Bengali, called the East-Central, Along the eastern littoral
of the Bay of Bengal we find another distinct dialect, also of the Eastern type, called
South-Western Bengali, and inland there is one more curious dialect, called Ohakma
(with an alphabet of its own) spoken by tribes of the Chittagong Hill Tracts.
Having now completed a rapid survey of the various dialects, we may take stock
and see how many people speak Bengali in its proper home.
gallon the "se^gaii'area. **' This is shown in the following table. Details will be found
in the sections dealing with each dialect separately : —
Name of Dialect.
Name of Dialect.
-Central or Standard
Western-(inchding Kharia Thar, Mai
Paharia, and Sarakl sub-dialects) .
NoEfchera (including the KOch and
Eajbangsl (including Bahe sab-
Eastern (including Eajang and East-
•Central sub-dialects) .
South-Eastern (including Chakma) .
TOTAL for Bengal
/Iddl—South-Eastern Bengali, spoken
TOTAL for Assam
GRAND TOTAL for Bengali spoken in
the Bengali-speaking area .
Bengalis helong to an intelligent and well-educated nationality, and have spread
far and wide over India as clerks, or in the practice of the learned professions. . It is,
therefore, of interest to note how far the Bengali language
^ Bengali as a foreign langu, j^ ^^^ ftfe time as a foreign speech, over the rest
of India. We have counted up the number of people who*
speak Bengali at home, let us now see how many people speak it abroad. As the
returns of this Survey do not take cognisance of tlie languages spoken hy small groups
of people who are away from their homes, we shall not follow them, but shall take
instead the figures of the Census of 1891* l?or obvious, reasons it is impossible to
classify these entries according to dialect, and we must content ourselves with noting
that the latter is unspecified. I shall commence with those portions of theLower
1 These are Census figures, and are not based on special local returns.