Correcting the dialect out of the classroom
A Class III Bengali (Language I) teacher in a school in Purulia district has 41 children in one room. Twenty-seven speak Santali at home, nine speak a Kurmali variety, and five speak Bengali. The current lesson, "Amar Gaon", is in standard written Bengali. During oral work the children say moy instead of ami, koi instead of boli, and drop aspiration, so ghor becomes gor and khela becomes kela. The teacher's first response was to repeat the standard form in a louder voice until the child said it back; after three weeks, oral participation has fallen sharply. Muni, previously the liveliest speaker, now shakes her head when asked anything. The teacher's guide, printed by the textbook board, says that pronunciation errors should be corrected "immediately and firmly". A district inspector is due next month and will look at the oral Bengali register, in which the teacher must record each child's speaking. The headteacher says that correcting the dialect is the teacher's duty and that the children must learn "proper Bengali" before they can read and write it.
1. Distinguish a dialect feature from an error, and explain what each kind of correction does to the learner.
Read the worked answer
A dialect feature is a rule-governed variety of the same language, shared by a whole speech community. Moy/ami and koi/boli are not random mistakes; they are a different grammar, used consistently and understood by everyone in the room. An error is a departure from the target that blocks meaning or breaks a pattern the child is being taught at that moment.
The consequences of confusing them are predictable:
- Correcting dialect in oral production teaches the child that her home language is wrong. Muni did not fail to produce Bengali; she produced a variety and was told it was a fault. The result is exactly what happened: she stopped speaking. Oral output is the raw material for later writing, so silencing it costs more than any pronunciation gain.
- The standard variety is a legitimate long-term goal, but it is acquired by comparison, not erasure. Children told their speech is wrong do not reach standard Bengali faster; they speak less, volunteer less, and read less fluently because reading aloud becomes an exposure to correction.
Correct at once the error that blocks meaning, or a form the lesson is explicitly teaching and has just modelled. Note aspiration and pronoun choice for a planned comparison slot. "Immediately and firmly" in the teacher's guide applies to meaning-blocking errors, not to a child's regional accent.
2. Plan the oral segment of a 40-minute lesson on "Amar Gaon" so that the dialect is included rather than corrected.
Read the worked answer
The lesson still ends with standard written Bengali; the route there changes.
- Home-language brainstorm (3 minutes). "Tomar gaon sombondhe duto kotha bolo, jekono bhabe." Accept every contribution exactly as offered and write the ideas on the board in the child's own words. Put Muni's sentence up first, without comment.
- Same idea, two forms (3 minutes). Two columns: column A holds the child's form, column B the book's form. "Moy bhalo achi" beside "Ami bhalo achi". Say explicitly: "Duto-i thik; boi-te ei bhabe lekha hoy." This is comparison, not correction, and it gives the standard form a reason to exist.
- Book reading (3 minutes). Children read the paragraph from the book; the teacher models, the class echoes. Here the text, not the child, supplies the standard.
- Own sentence (1 minute). Each child says one sentence about their own village using the book's pattern, so the new form is used, not just heard.
Keep a private notebook of recurring forms heard across the week. Once a week, five minutes of bhasha jigyasha looks at one form as a comparison between two ways of saying the same thing. Never repeat a child's form back "correctly" in front of the class: that is the move that silenced Muni.
3. How will you fill the oral Bengali register and assess speaking without penalising dialect?
Read the worked answer
Mark the construct you actually want. If the register asks for "speaking in Bengali", record what the child can do: conveys meaning, speaks in connected sentences, uses new words from the lesson, takes turns. Dialect is not a rubric category, because it is not an achievement deficit.
A workable design:
- Three-point rubric per child, filled during pair work rather than in front of the class: (i) meaning is clear; (ii) length and connectedness; (iii) use of two or three target words from the lesson.
- A separate planning note, by group and not by name, of the standard forms children do not yet control. That note drives the weekly comparison slot.
- Reading aloud is assessed against the text actually read: accuracy and expression, not accent.
- Tell the children the criteria in advance. Children who know that their home speech is not being marked will speak, and speaking is the behaviour the register is trying to capture.
If the inspector asks, the register shows individual progress with dated evidence and a stated criterion. That is more defensible than a column of ticks for "correct pronunciation", which measures conformity and not language. The child who speaks in three sentences of her own variety is further along than the child who says one memorised standard sentence and then falls silent.