Assam HS Sanskrit common mistakes
Assam HS Sanskrit · Understand the traps and how to avoid them.
Writing the visarga as a full 'ha' or omitting it at the end of a word when transcribing Devanagari into Roman or while reciting sandhi forms.
The visarga (ः) is a light breath marker visually close to other marks; students who learned letters as isolated shapes treat it as decorative and default to the more familiar 'ha' sound or drop it entirely.
What to do instead
Drill a two-column list of word-final -aḥ, -āḥ, -iḥ forms with audio or teacher modelling; before every transcription, circle every visarga in the source passage and verify each one appears as 'ḥ' in the answer.
Applying sandhi mechanically without checking whether the junction is internal or external, producing forms like *rāma + aśva → rāmāśva instead of the correct rāmo 'śva.
Candidates memorise the vowel-guṇa rule (a + a → ā) in isolation and apply it to every adjacent a-sound, forgetting that before a voiced vowel or certain consonants the preceding -a becomes -o with elision of the initial a.
What to do instead
Build a decision chart: first identify whether the second word begins with a vowel or consonant, then apply the visarga/guṇa branch, then re-read the joined form aloud to test whether it matches known textbook examples such as rāmo 'śva, te 'pi, so 'yam.
Declining a noun by matching the ending of the stem rather than identifying its gender and class, so that a masculine -a stem like rāma is declined with the feminine -ā paradigm (rāmāyaḥ, rāmāyāḥ for dative/ablative/genitive).
The written endings -a and -ā look almost identical in Devanagari, and candidates often learn paradigms by visual pattern rather than by gender, so the feminine ā-paradigm bleeds into masculine a-stems.
What to do instead
For every noun in a translation passage, write a three-part tag (stem, gender, case-number) before translating; practise paired drills contrasting rāma/rāmā, kavi/kavī, and guru/guru in the same case to make the gender distinction automatic.
Translating a Sanskrit sentence word-by-word in English order, leaving the finite verb in the middle and missing the final verb that carries the sentence's main assertion.
Sanskrit is verb-final and highly inflected, but candidates who think in English word order stop at the first familiar noun and never reach the verb, producing fragmentary or reversed translations.
What to do instead
Use a fixed reading routine: scan to the last word first, identify the finite verb and its person/number, then work backwards assigning each noun to a case role; write the case label above each word before composing English.
Confusing the six kāraka roles when translating, especially treating the instrumental (tṛtīyā) as agent and the ablative (pañcamī) as source indiscriminately, so that 'rāmeṇa' is rendered 'for Rāma' or 'from Rāma'.
Case endings overlap in form (e.g., -ena, -āt, -tas) and candidates learn them as a list of suffixes without linking each to a question word (kena? kasmāt? kva?).
What to do instead
Memorise the kāraka question words as a set (kartā ko? karma kim? karaṇa kena? sampradāna kasmai? apādāna kasmāt? adhikaraṇa kva?) and answer each question explicitly for every sentence in practice passages before translating.
Conjugating a verb from the wrong gaṇa, for example forming *likhati from likh as if it were a class-1 thematic verb when the expected present is likhati but the imperfect or causative is misformed, or producing *bhavati from as/√bhū without the guṇa.
The ten gaṇas are taught as lists of endings, and candidates memorise the present-tense personal endings without learning which roots take which class, so they default to the -a thematic pattern for every root.
What to do instead
Keep a root card for each verb studied: root, gaṇa, present third-person singular, imperfect, and one derived form; test yourself by generating all three persons in both numbers from the card before attempting translation.
Misidentifying a gerund (ktvā/lyap) or infinitive (tumun) as a finite verb, then supplying a subject and tense that the sentence does not contain.
Forms like kṛtvā, gatvā, and draṣṭum look like verbs and carry verbal meaning, but they are indeclinable participles; candidates who rely on meaning alone miss the -tvā and -tum markers.
What to do instead
Highlight every word ending in -tvā, -ya, -tum, or -itum in a passage and label it 'participle/infinitive - no person or number'; rewrite the sentence with an explicit finite verb before translating.
Reading a compound (samāsa) as a sequence of separate words, so that rājaputra is translated 'king and son' or 'the king's son' without deciding whether it is a tatpuruṣa, dvandva, bahuvrīhi, or avyayībhāva.
Compounds are written without spaces and the case relationship between members is unmarked; candidates who have not learned the compound types guess from the first member's meaning.
What to do instead
Practise splitting compounds by inserting the missing case ending (rājñaḥ putraḥ = tatpuruṣa), and classify each compound in a passage into one of the four main types with a one-line justification before translating.
Answering appreciation or rasa questions about a prescribed verse with a plot summary instead of naming the alaṃkāra, rasa, or vṛtta and quoting the relevant words as evidence.
Candidates prepare verse meaning thoroughly but not the technical vocabulary of poetics, and the question wording ('identify the figure of speech') is read as 'explain the verse'.
What to do instead
For each prescribed verse, prepare a four-line note: metre (with gaṇa pattern), alaṃkāra (named, with the two entities compared), rasa, and one key phrase; practise writing answers that begin with the technical term and then quote.
Losing marks on translation into Sanskrit by writing a literal English-to-Sanskrit calque, e.g., placing the subject after the verb or using the wrong case for the object of a verb of giving or speaking.
Composition is usually practised less than comprehension, and candidates transfer English syntax and prepositional logic into Sanskrit, ignoring that dative marks the recipient and accusative the thing given.
What to do instead
Before writing, draft the Sanskrit sentence in the order subject–object–verb and check each noun against its kāraka question; keep a checklist of common verb-government patterns (dadāti + dative recipient, vadati + accusative addressee) and apply it to every composed sentence.