This guide adapts rules and examples from Microsoft’s Localization Style Guide for Sanskrit (India) (originally written for software/UI localization). The underlying linguistic rules apply universally — to legal contracts, medical documents, marketing copy, and any Sanskrit translation work. Restructured and reformatted as a general Sanskrit translator reference by ChatsControl.

Sanskrit Translation Style Guide — Voice, Word Choice & Common Pitfalls (Legal, Medical, Marketing, IT)¶

TL;DR¶

• Sanskrit translation across all spheres should maintain a respectful, polite, formal tone — Sanskrit’s structure naturally enforces formal register, making it suitable for medical, legal, marketing, and IT contexts.
• Address user formally with त्वम् (singular), युवाम् (dual — two people), यूयम् (plural — three or more). Sanskrit’s three-number system (singular/dual/plural) requires precise person-count handling.
• Strictly avoid Nuqta (़) in Sanskrit — it’s a Devanagari diacritic for borrowed sounds (Urdu/Arabic/English), not native Sanskrit. Many Hindi-Sanskrit translators accidentally include it.
• Distinguish colon (:) from Sanskrit visarga (ः) — visually similar but linguistically distinct. राम: (with Latin colon) is wrong; रामः (with proper visarga) is correct.
• Sanskrit has no capitalization — Devanagari is unicameral. Acronyms in English (DNS, HTML, LED, CPU) kept in Latin script with English pluralization rules; first occurrence shows full form in parentheses.

Register and tone for modern Sanskrit translation¶

Always keep the tone of voice respectful, polite, formal. Sanskrit’s grammatical structure enforces this naturally — there is no informal/conversational register in modern Sanskrit usage.

Why this matters: Sanskrit’s inherent formality makes it suitable for legal documents (contracts, court rulings translated for archival), religious/cultural content (where Sanskrit is the source language), scientific/technical writing (Ayurveda, Yoga, classical sciences), medical texts (Ayurvedic patient materials), and modern UI (where formal register is acceptable, even expected). There is no marketing-friendly “casual Sanskrit” — the language is what it is.

Abbreviations¶

You might need to abbreviate words due to lack of space, especially in UI.

Abbreviated terms can be used in plural. Example: TVs → टीवीस्.

Acronyms¶

Acronyms made of initial letters of compound terms. Well-known: DNS (Domain Name Server), HTML (Hypertext Markup Language).

• Most acronyms kept in English due to wide acceptance/usage in Sanskrit. Plural follows English rules (LEDs, CPUs).
• If an English acronym is to be used throughout text: first occurrence — acronym (full form). Subsequent — acronym only.

Example:

• First: DNS (Domain Name System)
• Later: DNS

Capitalization¶

Capitalization does not apply to Sanskrit. Devanagari is a unicameral script — single form is used for each letter. No uppercase/lowercase distinction.

Punctuation¶

Critical distinction: The colon (:) is different from the Sanskrit visarga (ः). They are visually similar but linguistically distinct. Many translators mistakenly use the Latin colon in place of visarga.

• Wrong: राम: (with Latin colon)
• Correct: रामः (with proper visarga ः)

Ensure correct visarga (ः) is used in Sanskrit text.

Bulleted lists¶

In Sanskrit, punctuation (full stops, question marks) used in bulleted lists only if the list item is a complete sentence. If not a complete sentence, no specific punctuation rules apply.

Example:

Dashes and hyphens¶

Hyphen (-) — shortest dash. Use for:

• Compound words
• Breaking words at end of line
• Page numbers, dates, number intervals
• Establishing relationships between concepts

En dash (–) — Alt+0150 or Ctrl+- (numpad). Used in arithmetic and negative numbers.

Example: Temperature is -10°C → तापमानं -10°C अस्ति।

Em dash (—) — Alt+0151 or Ctrl+Alt+- (numpad). Avoid extensive use of English em dash. Use commas or parentheses instead. Can start new sentence. Feel free to change sentence (remove em dash) for natural Sanskrit.

Example:

Quotation marks¶

Straight quotation marks ( ” ” ) typically used in Sanskrit.

• Use quotation marks when someone is speaking.
• Do not use quotation marks when referring to UI items.

Usage of Nuqta¶

Nuqta (़) is a diacritic used in Devanagari to describe modern sounds borrowed from languages NOT native to Devanagari — mainly Urdu (Arabic, Farsi) and English. It appears as a dot below the letter and is used in some Hindi letters: क़, ख़, ग़, ज़, फ़.

Nuqta is NOT used in Sanskrit. Ensure Nuqta is strictly avoided in Sanskrit text.

This is a common error when translators familiar with Hindi work on Sanskrit content — Hindi conventions automatically apply Nuqta; Sanskrit forbids it.

Numbers, symbols, non-breaking spaces¶

Numbers. In documentation, numbers 1-10 spelled out; other numbers as numerals.

Non-breaking space. Use between numeral and symbol/unit of measure. Examples: 30 cm, 1 h, 75 %, 20 °C.

Ampersand. Always translate “&” as “च” when in running text. Don’t keep “&” in target unless part of a tag, placeholder, shortcut, or code.

Date¶

Date format: DD/MM/YYYY (e.g., 01/01/2025).

When only month and year (e.g., “March 2025”): MM YYYY format.

Inclusive language¶

General principles: comply with local language laws; use plain language; mind regional/cultural references; represent diverse perspectives; don’t generalize/stereotype; don’t use profane/derogatory terms or military/political jargon.

Term replacements¶

Avoiding gender bias¶

Use gender-neutral alternatives. Avoid compounds with gender-specific terms (नरनायकः, स्त्रीशिक्षिका). Use gender-neutral terms (नेता, शिक्षकः).

When presenting generalization, use plural noun forms (जनाः, प्रत्येकं, छात्राः).

Don’t use gendered pronouns (सा, तां, सः, तं) in generic references. Instead:

• Rewrite to use second or third person (भवान् or कश्चन).
• Rewrite to have plural noun and pronoun.
• Refer to person’s role (पाठकाः, कर्मचारी, क्रयणकर्ता, ग्राहकाः).

Real people: Use the pronouns the person prefers (सः, सा, they, or another). Gendered pronouns are fine for real people who use them.

Accessibility¶

Focus on people, not disabilities. Sanskrit, being highly structured, allows flexibility while maintaining respectful tone.

Keep paragraphs short. Aim for one verb per sentence. Read text aloud and imagine it spoken by a screen reader.

Applications, products, features¶

Trademarked application/product names are not translatable. Verify before translating.

Version numbers¶

Version numbers always contain a period.

Trademarks¶

Trademarked names and “Microsoft Corporation” shouldn’t be localized unless local laws require translation and approved translated form is available. Reference: https://www.microsoft.com/en-us/legal/intellectualproperty/trademarks.

Reference materials: authoritative Sanskrit references¶

Normative references:

1. Ashtadhyayi (अष्टाध्यायी) — Pāṇini’s foundational Sanskrit grammar work. Definitive on classical Sanskrit grammar.
2. Laghushiddhant Kaumudi (लघुसिद्धान्तकौमुदी) — concise traditional Sanskrit grammar reference.
3. Any English to Sanskrit dictionary — multiple acceptable.

Note: If more than one spelling is acceptable, opt for the spelling adopted on Microsoft approved terminology and in the normative dictionaries. Use consistent spelling across Microsoft products.

Informative references:

1. Sanskrit Documents — https://sanskritdocuments.org
2. Monier-Williams Sanskrit Dictionary — https://www.sanskrit-lexicon.uni-koeln.de/monier/
3. Digital Corpus of Sanskrit (DCS) — http://www.sanskrit-linguistics.org/dcs/ (Corpus for linguistic research)

FAQ¶

What’s the register for Sanskrit translation across professional contexts?¶

Sanskrit’s grammatical structure enforces a respectful, polite, formal tone naturally. This makes Sanskrit suitable for technical, legal, medical, and ceremonial contexts. There’s no informal/conversational register in modern Sanskrit usage — the language inherently carries formality.

How should I address users in Sanskrit translation?¶

Use त्वम् (tvam, singular) for one user, युवाम् (yuvām, dual) for two users, यूयम् (yūyam, plural) for three or more. Sanskrit’s three-number distinction (singular/dual/plural) requires precision. Avoid भवान् in generic addressing — it’s masculine; for gender-neutral, prefer त्वम् forms.

What is Nuqta and why must it be avoided in Sanskrit?¶

Nuqta (़) is a Devanagari diacritic (dot below letter) marking borrowed sounds from Urdu, Arabic, Farsi, English (क़, ख़, ग़, ज़, फ़). Hindi uses Nuqta extensively; Sanskrit does NOT use it. Many Hindi-Sanskrit translators accidentally include Nuqta — strictly avoid in Sanskrit output.

How does Sanskrit handle capitalization and acronyms?¶

Sanskrit has no capitalization — Devanagari is unicameral (single case). Acronyms commonly kept in Latin English (DNS, HTML, LED, CPU) due to wide acceptance. Plural follows English: LEDs, CPUs. First occurrence in text: acronym (full form). Subsequent occurrences: acronym only.

Which Sanskrit normative references should I follow?¶

Ashtadhyayi (Pāṇini’s grammar — foundational Sanskrit grammar work), Laghushiddhant Kaumudi (concise traditional Sanskrit grammar reference), and any reliable English-Sanskrit dictionary. Informative: Sanskrit Documents (sanskritdocuments.org), Monier-Williams Sanskrit Dictionary (sanskrit-lexicon.uni-koeln.de/monier), Digital Corpus of Sanskrit (sanskrit-linguistics.org/dcs).

What’s the difference between colon (:) and visarga (ः) in Sanskrit?¶

Visarga (ः) is a Sanskrit phoneme marking aspiration at the end of certain words (रामः, गुरुः). It is NOT the Latin colon (:). They look similar but are linguistically distinct Unicode characters. Always use the Devanagari visarga (U+0903) in Sanskrit text, not Latin colon (U+003A).

How does Sanskrit handle date formats?¶

Date format: DD/MM/YYYY (e.g., 01/01/2025). When only month and year: MM YYYY format. Months may be kept in English transliteration (मार्च for March, जनवरी for January).

Sources¶

• Microsoft Localization Style Guide for Sanskrit (PDF)
• Microsoft Globalization Style Guides hub
• Monier-Williams Sanskrit Dictionary
• Sanskrit Documents
• Digital Corpus of Sanskrit