register-discourse-and-audiovisual-constraints

name	register-discourse-and-audiovisual-constraints
description	Reviews a translation's register and discourse choices on the Hallidayan field/tenor/mode model, its cohesion and pragmatic linkage, politeness and face, transitivity and modality shifts, House's overt/covert axes, its audiovisual (AVT) signifying-code analysis, and its subtitling-specific shortening constraints (it does not carry dubbing isochrony/lip-sync or voice-over criteria); it reviews the analysis, it does not produce the finished translation, subtitle file, or publication decision.
kind	skill
status	ready
provenance	{"principles":["P019","P021","P052","P053","P064","P065","P066","P067","P068","P069","P085","P090","P091"],"claims":["C00009","C00010","C00040","C00171","C00172","C00173","C00174","C00175","C00176","C00177","C00178","C00179","C00180","C00181","C00182","C00183"],"evidence":[],"source_anchors":[],"authored_from_digest":"f5c9a89060d151dab00a1708c447f7a22ea2d9777a5d19a441e648675aa9f744"}

Register, Discourse And Audiovisual Constraints

Purpose

This skill reviews translation at the level of register and discourse using the Hallidayan model — field, tenor, mode — and checks cohesion, pragmatic linkage, transitivity and modality shifts, and politeness/face management. It gives special attention to audiovisual and subtitling constraints, where hard limits on space, legibility, and timing force a shortening that can re-code interpersonal meaning rather than merely omit it. Note that House's two same-named axes are distinct and must not be conflated: the overt/covert translation-type typology names a translation's mode (P021), while the overtly-/covertly-erroneous error taxonomy classifies an error's type/origin — a covertly erroneous dimensional register or genre mismatch versus an overtly erroneous denotative or target-system error — not its severity (P065) — same words, different judgments. It reviews a translation, a translation choice, or a translation-studies analysis; it does not produce the finished translation or subtitle file.

When to use

• A translation's register profile (field/tenor/mode), genre, and function need comparing against the source's for a quality judgement.
• A cohesive tie, an implicit pragmatic link, or a transitivity/modality shift may have changed meaning.
• A face-threatening act, its irony, or a marked discourse such as camp is being shifted across cultures.
• A subtitle or dubbed line is constrained by on-screen space, timing, camera cuts, or legibility.
• An audiovisual text's acoustic and visual codes need checking against its linguistic track for coherence.

Procedure

1. Analyse an audiovisual text across all of Chaume's ten signifying codes — one linguistic code, three further acoustic codes, and six visual codes — using multimodal transcription of frames, shots, and phases with a metafunctional interpretation, checking that the linguistic code reads with an oral register, that visual and colour codes are respected so an unfamiliar symbol is explained and coherence with the image is kept, and that the review accounts for subtitling's overt, vulnerable status and the subtitler's frequently uncredited work (P019).
2. Decide, before judging any specific choice, whether the translation under review is functioning as overt — seeking second-level functional equivalence at language, register, and genre because its source is tied to a specific culture and time — or covert — reading as an original through a cultural filter — treat the distinction as a cline rather than a binary, and check that a version rather than a translation was produced where the source genre has no target equivalent (P021).
3. Check whether subtitling's habitual preference for brevity has merely omitted politeness markers or has actively re-coded the interpersonal dynamics — for instance whether a concise, subject-dropping target style has flattened a source character's face-threatening directness into unwarranted familiarity — and verify that any dropped hedge, discordant interpersonal signal against the moving image, or untranslatable face-protecting pronoun is flagged as a systematic sacrifice rather than judged phrase by phrase (P052).
4. Use politeness theory to pinpoint exactly how a translation shifts a face-threatening act and its irony, read a marked discourse such as camp as political rather than purely aesthetic, check that its minimization or amplification is read against the target culture that produced it rather than in isolation, and flag any collapsing of a subculture's internally distinct, ironic lexicon into one neutral mainstream term (P053).
5. Use the Hallidayan model to relate lexicogrammatical choices to the text's function in its sociocultural framework, analysing register as field, tenor, and mode mapped onto the ideational, interpersonal, and textual metafunctions, and reading transitivity, modality, thematic structure, and cohesion to check whether a shift such as active-to-passive or a nominalization has changed who is represented as responsible; this skill owns the full field/tenor/mode register-and-cohesion analysis, while naming the specific Vinay & Darbelnet or Catford procedure or shift that a marked register feature motivates routes to the sibling translation-procedures-and-shifts (P064).
6. Assess translation quality by comparing the source and target register profiles — field, tenor, mode — together with their genre and function, and classify any mismatch found as a covertly erroneous dimensional mismatch in register or genre, or as an overtly erroneous denotative or target-system error (P065).
7. Treat cohesion — Halliday and Hasan's five types of reference, substitution, ellipsis, conjunction, and lexical cohesion — as language-specific, checking that the target rebuilds its own web of cohesive ties rather than copying the source's, that any translator-typical explicitation of cohesive ties has not caused an unintended functional shift, and that a gender- or subject-inflected target's forced disambiguation of a deliberately ambiguous source was a deliberate, loss-aware choice (P066).
8. Check that pragmatic meaning has been managed across cultures by making an implicit link explicit wherever target receivers lack the source receivers' background knowledge, that a presupposed reference has been made explicit when needed, and that implicature has been handled through Grice's maxims of quantity, quality, relevance, and manner plus politeness, with awareness that different cultures operate different cooperative maxims (P067).
9. Watch how discourse-level choices shift meaning: verify whether a transitivity shift has changed the ideational function, whether a modality shift has changed a truth value, whether a lexical choice has imposed an unintended ideology, and whether the interpersonal function of hedges, evaluative or attitudinal language, or pronoun choice has been downplayed or shifted in a way that reveals the translator's own interpretation (P068).
10. For subtitling, confirm the rendering stays within the hard space and time limits of about two lines of 38 Roman characters or 13 to 15 Chinese or Japanese characters and around six seconds, respects camera cuts, matches duration to dialogue rhythm, follows the near-universal guidelines to simplify grammar and lexis while keeping interactional features only partly, and applies one of Pedersen's strategies — retention, specification, direct translation, generalization, substitution, omission, or official equivalent — to any culture-bound reference, marked speech, song, or piece of humour (P069).
11. Weight the seriousness of any face-threatening act as a cultural variable set by the social distance and relative power of the speakers, check that politeness has been relayed trans-culturally with the texture-level modification its varying linguistic realization requires, and confirm that a subtitler has relayed both the character's on-screen discourse and the scriptwriter's separate signal to the mass-audience auditor, since film dialogue is doubly designed (P085).
12. Apply discourse and register analysis as a flexible toolkit rather than a rigid one-to-one model: check whether an apparent source-target mismatch is in fact a legitimate strategy such as explicitation or compensation rather than an error, and flag any uncritical imposition of English-oriented frameworks or of Gricean, fluency-biased assumptions onto other or non-European languages or onto text types they do not suit (P090).
13. Treat subtitling as an overt, vulnerable translation constrained by on-screen space, legibility, and timing that force shortening; check whether that shortening has normalized an informal sociolect by increasing cohesion and imposing conventional grammar, verify that a lost tenor marker such as a reported-speech subjunctive has been compensated elsewhere, and remember that a soundtrack may carry tone a text-only analysis alone will miss (P091).
14. Synthesize the findings from highest impact to lowest: a register or discourse mismatch that reverses who is represented as responsible, or that erases a marked identity's political charge, outranks a subtitling-constraint trade-off that merely simplifies grammar, and a systematic interpersonal re-coding outranks an isolated lexical slip.

Inputs

• The source and target texts, or the audiovisual material together with its subtitle file, and any stated register, discourse, or subtitling constraints.
• The reasoning offered for the translation decision under review — its orientation (overt/covert), brief, strategy, and any quality claim made.

Output

Per finding: name the flaw and the principle it violates, apply the correction, state the residual uncertainty and the trade-off it reflects, and end with a concrete next step, ordering findings highest-impact first. When the request is an advise or compare rather than a post-hoc critique, apply the same checks forward — in advise mode, a recommendation naming the governing principle(s) and orientation with its residual trade-off; in compare mode, a side-by-side of what each option preserves and costs, ending in a brief-weighted recommendation. This skill reviews a translation, a translation choice, or a translation-studies analysis; it does not produce the finished translation or subtitle file, nor does it make the publication decision.

Anti-patterns to flag

• A subtitle review that checks only the linguistic dialogue track while ignoring the acoustic and visual codes it must cohere with, missing an unfamiliar on-screen symbol left unexplained or a mismatch between subtitle and image (P019).
• A review that judges a translation's fidelity without first placing it on the overt/covert cline, faulting a covert translation for not letting readers "eavesdrop" or an overt one for reading too smoothly as an original (P021).
• A subtitle that drops hedges and pronoun-of-address ambiguity for brevity and is read as a simple omission, when the concise style has actually re-coded a face-threatening character as merely familiar (P052).
• A camp or ironic register flattened into one neutral mainstream term, or its face-threat judged phrase by phrase against a fixed rule rather than against the culture that produced it (P053).
• A shift from active to passive, or a nominalization that quietly reassigns responsibility, passed over because only lexical choices — not transitivity or modality — were checked (P064).
• A register or genre mismatch labelled an "error" without distinguishing whether it is covertly erroneous (a dimensional slip) or overtly erroneous (a denotative or target-system mistake) (P065).
• A target text whose cohesive ties are copied straight from the source's reference and conjunction patterns rather than rebuilt for the target language, leaving it locally correct but globally incoherent (P066).
• An implicit pragmatic link or culturally loaded presupposition carried over unexplained for target readers who lack the background to reconstruct it (P067).
• A translator's own evaluative stance smuggled in through an untracked shift in hedges, attitudinal language, or pronoun choice that the review never names as an interpretation (P068).
• Subtitles that run over two lines, exceed the character or six-second limit, or ignore a camera cut, or a culture-bound reference handled without applying any of Pedersen's seven strategies (P069).
• A face-threat judged by a fixed severity scale that ignores the speakers' social distance and relative power, or a subtitler who relays the character's line but drops the scriptwriter's separate signal to the audience (P085).
• A legitimate compensation or explicitation strategy misread as an error because the reviewer applied a rigid one-to-one register model, or an English-oriented discourse framework imposed uncritically on an unrelated language pair (P090).
• A subtitle that quietly normalizes an informal sociolect into conventional grammar or drops a reported-speech subjunctive's tenor marker without compensating for it elsewhere, or a text-only analysis that misses a tone only the soundtrack carries (P091).

References

See ../../references/descriptive-translation-principles-index.md for the full principle catalogue grouped by skill, and ../../references/descriptive-translation-evidence-notes.md for how these principles are grounded and kept faithful to the sources.

Provenance

Derived from P019, P021, P052, P053, P064, P065, P066, P067, P068, P069, P085, P090, P091, grounded in the distillation-only sources (Munday, Introducing Translation Studies; Venuti, ed., The Translation Studies Reader; Toury, The Nature and Role of Norms in Translation). The frontmatter provenance block lists the exact principle and claim ids, which resolve into principles/principles.yaml and analysis/claims.jsonl.