The Dialect Correction
Ms. Rekha has joined a Class 4 Hindi classroom in a middle school in Bhiwani district. Thirty-six children, almost all of whom speak Haryanvi at home. In her first month she decided to 'clean up' their Hindi: she corrects 'म्हारे गाँव में' to 'हमारे गाँव में' and 'वे करै थे' to 'वे करते थे' in the middle of children's oral answers, deducts marks for dialect forms in written work, and reads aloud the three best notebooks every Friday. Her unit test shows comprehension is fine - class average 7.2 out of 10 on a passage with questions - but the average length of a free-writing answer has fallen from 41 words in July to 14 words in September. Pooja, who in July wrote six lines about her buffalo, now writes 'मेरी भैंस अच्छी है।' and stops. Oral participation fell from about 22 children answering per lesson to 6 or 7. One father says the teacher is teaching children to be ashamed of how their families speak.
1. Using the evidence in the setup, explain why free writing collapsed while comprehension scores held.
Read the worked answer
The two results measure different things, and only one of them involves risk.
Comprehension is receptive. A child can answer questions on a passage by recognising meanings already in the text; producing nothing original is required. So it stayed at 7.2.
Free writing is productive and exposed. Every sentence a child writes is a chance to be marked wrong. Rekha's corrections were public, mid-utterance (which interrupts planning), and evaluative - they cost marks, and Friday's read-aloud defines the target as error-free. The rational response is to write short, safe sentences that cannot be faulted. That is exactly the pattern: mean length dropping from 41 words to 14, and Pooja, who once wrote vividly, retreating to one flat line.
This is avoidance, not loss of competence. The change happened in six weeks, too fast for ability to fall, and only in the task that carries the penalty. The children still know the words; they have learned that using them is expensive.
A further signal: marking dialect forms as errors conflates two separate things - whether a child had something to say, and whether it was in standard Hindi - into one red mark. The comprehension test could not show this, which is why the two data points diverge.
2. How should Rekha mark dialect forms in written Hindi? Take a position and justify it.
Read the worked answer
Position: Haryanvi is a variety, not an error; standard Hindi is a register the child must add, not a replacement for home speech. A deficit stance misreads linguistic difference as cognitive or moral failure, and the evidence above shows what it costs.
Practical marking rules.
- Split the marks: one for content and organisation (does it say something, is it ordered, is there detail), one for standard written form (spelling, matras, agreement). A child can score high on the first while the second is developing.
- In a first draft, respond to meaning only. Standard forms are dealt with at the editing stage, where the child knows that is the focus.
- Keep a running class chart - 'जैसे हम बोलते हैं / जैसे लिखते हैं' - so the standard form is a visible second option, not a correction of the family.
- Use noticing questions: 'How would a news reader say this sentence?' Let the class generate the standard version.
- Never deduct for dialect in a draft. Assess standard forms on short, targeted editing tasks rather than on every piece of writing.
This preserves accuracy and voice together. A child who is never penalised for thinking aloud will write more, and more writing is the only route to better standard Hindi.
3. Draft a 20-minute oral activity for next Monday that builds standard Hindi without shaming any child, and say how Rekha will know it worked.
Read the worked answer
Activity: 'Two voices, one story' (20 minutes).
0–4 min. Pairs get a four-picture story. Each pair retells it to each other in the speech they use at home. No marks, no correction; Rekha circulates and listens only.
4–10 min. Pairs choose two sentences from their retelling and write them on a slip.
10–16 min. Rekha collects slips without names, writes three or four on the board and asks the class: 'If this were read on the radio, what would change?' Children offer versions; she writes both side by side and names the change ('यहाँ करै थे की जगह करते थे आता है'). Nobody is identified with an error; the class edits a text, not a child.
16–20 min. Chorus reading of the standard versions, then volunteers read aloud. Rekha notes who volunteers and which forms recur.
How she knows it worked. Three low-stakes indicators, tracked weekly: (1) number of children answering unsolicited in a lesson - target 15+ within three weeks, up from 6–7; (2) mean words in the next free-write - target back toward 35–40; (3) on the next editing task, how many children spot and convert the targeted forms. If hands rise but standard forms do not, the accuracy work needs more time; if neither moves, the classroom climate still needs attention.