Foreignizing the Sacred: Domestication and Foreignization in A. K. Ramanujan's Translation of Samskara

Foreignizing the Sacred is a dissertation-style study of A. K. Ramanujan's English translation of U. R. Ananthamurthy's Kannada novel Samskara (1965), first published in English by Oxford University Press in 1976 and still the most cited Indian translation in postcolonial classrooms. The novel's world — an agrahara of orthodox brahmins, its ritual vocabulary of purity and pollution, its food codes and its crisis over a corpse — arrives in English through a translator who was also a poet-linguist, and the question this study asks is the one Lawrence Venuti made central to translation theory: does the translation domesticate the source culture into fluent English, or foreignize the English reader into contact with what remains untranslatable?

The method is a descriptive, target-text-oriented analysis. I build a parallel-text corpus of forty passages sampled across the whole novel and weighted toward ritual, food and social-interaction scenes; code every culture-bound item in each passage against a seven-strategy codebook (retention, retention with in-text gloss, paratextual support, domestication, explicitation, omission, syntactic calque); and track every retained Kannada and Sanskrit word in a lexicon and paratext tracker that records how and where the reader is helped. Frequency tables and charts establish the dominant strategies; close readings of six extended passages — including the famous paratext of the untranslated title itself — test what the frequencies mean.

The study frames its findings through Venuti's distinction and Tejaswini Niranjana's postcolonial critique, and states its own limits honestly: I read the translation, not the Kannada, and the design says so.

Indian English departments teach Samskara in English so habitually that the translation disappears — students quote Ramanujan's sentences as if Ananthamurthy had written them, and the theoretical questions that come with that invisibility are left unasked. Is the ritual world of the novel being foreignized into the reader's awareness, or domesticated into fluent, frictionless English? Venuti's categories offer a way to answer, but they are rarely tested against an Indian text with a countable corpus; Niranjana's colonial-context argument raises the stakes but is seldom brought down to the level of particular words and sentences. There is a need for a small, honest, reproducible study that codes Ramanujan's actual strategies passage by passage, tracks how the paratext compensates for retained foreignness, and states its own linguistic limits rather than hiding them — so that the translation debate moves from admiration to evidence.

Abstract

Samskara is the classic test case of Indian translation: a modernist Kannada novel whose plot is a ritual question and whose texture is Sanskritised Kannada, rendered into English by A. K. Ramanujan, a bilingual poet-translator. This study codes Ramanujan's translation strategies across a parallel-text corpus of forty passages, asking whether his English domesticates Ananthamurthy's ritual world or foreignizes it — and what that choice means in India.

Introduction

Venuti's distinction between domestication — making the text read like native English — and foreignization — making the reader register the foreign — has become the standard vocabulary for judging translation, while Niranjana's Siting Translation argues that in colonial and postcolonial contexts translation is never innocent: it can flatten asymmetrical languages or expose the violence of the asymmetry. Samskara sits exactly on this fault line. Its title is left untranslated in the English edition, with the subtitle A Rite for a Dead Man doing compensatory work; its brahmin world is carried by words such as madi and agrahara whose translation is itself an interpretive act. Ramanujan's own afterword discusses these difficulties, which makes the book an unusually self-aware corpus.

Literature gap

• Classroom discussion of Samskara in English almost always treats the novel as if it were written in English; the translation's strategies are admired or assumed, not counted.
• Indian translation historiography (Sujit Mukherjee's Translation as Discovery) and postcolonial theory (Niranjana) discuss Ramanujan's work in general terms, without a coded, reproducible reading of this text.
• Venuti's categories are usually applied to European classics; studies that test them against an Indian bhasha-to-English canon text with a systematic corpus remain rare at the MA level.

Proposed study

• A forty-passage parallel corpus sampled evenly across the novel and stratified by scene type: ritual and purity scenes, food scenes, social confrontation, and interior monologue.
• A seven-strategy codebook with decision rules and worked examples, applied to every culture-bound item in the corpus; a second reader codes a ten-passage subset to check consistency.
• A lexicon and paratext tracker: every retained Kannada/Sanskrit item, its first appearance, the glossing apparatus around it, and its recurrence — including the title-page and afterword as paratext.
• Frequency analysis (strategy distribution overall and by scene type) plus close readings of six extended passages where strategy and interpretation interact.

Expected findings and feasibility

The study expects to show a split strategy: retention with paratextual support dominating the ritual lexicon — foreignization at the word level — while domestication dominates syntax and idiom, delivering a foreign world in fluent sentences. The subtitle and afterword emerge as the load-bearing compensation for what the lexicon withholds.

Feasibility: the text is short, in print and affordable; the method needs only a passage log, the codebook and Excel; the workload fits twenty weeks; and the one serious limitation — I read English and Tamil, not Kannada — is built into the design as a target-text study, with two Kannada-speaking postgraduate students back-checking a ten-passage subset.

In scope

• Ramanujan's English translation of Samskara as the primary corpus, read as a target text.
• Forty coded passages plus six extended close readings, with page references to the edition used throughout.
• Culture-bound items: ritual vocabulary, kinship and caste terms, food and purity codes, forms of address, proverb and idiom.
• Paratexts belonging to the translation: title and subtitle, translator's afterword, in-text glosses and typographic signalling (italics).
• Frequency tables and charts of strategy distribution, and a ten-passage second-reader consistency check.

Out of scope

• Source-text philology: no independent analysis of the Kannada original beyond the back-checked subset.
• Reception study: no reader surveys or reviews analysis; the reader is a theoretical construct from Venuti.
• Comparison with other translations of Samskara into English or other languages.
• The novel's film adaptation and its subtitling choices, noted only as future scope.

Research design: qualitative-descriptive with a quantitative layer — a target-text-oriented descriptive translation study. The unit of analysis is the culture-bound item within a passage; the unit of observation is the passage itself.

Instruments: the passage log, the codebook, the decision diary for borderline cases, the lexicon-and-paratext tracker and the reliability note from the second reader.

Analysis plan: strategies are counted as types, not tokens — a word retained fifty times counts once — so the frequencies describe choices, not repetition. The distribution is reported overall and by scene type; the lexicon table is read for patterns in glossing effort (where the reader is helped, where abandoned); and the six close readings interpret the numbers through the theoretical frame. All quotations follow MLA 9, with the edition fixed on the first citation.

This is a literary study, so the architecture below is the research pipeline: how a reading of one book becomes coded, checkable evidence and then an argument.

1. Frame layer — Venuti's domestication/foreignization distinction and Niranjana's colonial-context argument define what counts as evidence: choices a translator made that a reader can feel.
2. Corpus layer — the forty-passage log with page references and scene tags, the fixed edition, and the rule that every claim must point to a logged passage.
3. Coding layer — the seven-strategy codebook, the decision diary for borderline cases, and the second-reader consistency check that keeps the categories from dissolving into opinion.
4. Argument layer — frequency tables, the lexicon and paratext tracker, and six close readings that give the numbers their interpretive weight.

flowchart TD
  A["Frame: Venuti, Niranjana, Mukherjee"] --> B["Build passage log: 40 stratified passages"]
  B --> C["Tag scene types: ritual, food, social, monologue"]
  C --> D["Code culture-bound items with the 7-strategy codebook"]
  D --> E{Second reader agrees?}
  E -->|No| F[Revise codebook and re-code borderline cases]
  F --> D
  E -->|Yes| G["Kannada back-check of 10 passages by PG students"]
  G --> H["Lexicon and paratext tracker: retained items and glosses"]
  H --> I[Frequency tables by strategy and scene type]
  I --> J["Close readings of 6 extended passages"]
  J --> K["Argument chapter: foreignizing lexicon, domesticating syntax"]
  K --> L[Dissertation draft in MLA 9 format]
  L --> M[Seminar presentations and viva preparation]

The loop between coding and the codebook is the method's honesty: the categories were revised twice in the pilot before any counting began, and every revision is dated in the decision diary so the final codebook is reproducible by another student.

• Replicate the study with Kannada source-text access, adding translation-accuracy and omission analysis at the sentence level.
• Compare Ramanujan's Samskara with his Speaking of Siva to test whether his strategy set is author-specific or translator-signature.
• Extend the frame to a Dalit or women's translation from Tamil or Kannada, where caste and gender vocabulary sharpens the domestication question.
• Study the 1970 film adaptation's handling of the same ritual vocabulary in subtitles.
• Survey MA classrooms on how students cite the translation — turning this study's premise into reception data.
• Contribute the coded glossary to a digital annotated edition of the translation for classroom use.