The parameter
Every clause does two jobs at once. It states a grammatical relation, saying who did what to whom, and it packages that content for a reader who already knows part of it, marking what is talked about and what is new. Languages differ in how much sentence structure goes to each job.
Li and Thompson set out the classic version of the question in 1976, sorting languages by whether the subject or the topic is the basic structural notion, with four outcomes: subject-prominent, as in the Indo-European languages; topic-prominent, as in Mandarin; both; and neither. Japanese and Korean are usually placed in the "both" group, having subjects and a dedicated topic marker. A topic-prominent language builds clauses as topic plus comment, has little use for a passive, and needs no dummy subject for the weather or for existence.
That classification is a starting point rather than a finding. A second, partly independent variable is how freely word order can be used pragmatically. Russian, Polish and Turkish have case, so order is free to signal what is given and what is new; English has almost none, so order is committed to grammar and focus must be marked by other devices. A third is whether the language reserves a position or particle for focus, which Hungarian and Turkish do and English does not.
How languages vary
- Subject-prominent: the subject is the pivot of clause structure, and packaging is done by voice, clefting and intonation. English, French, German.
- Topic-prominent: clauses are built as topic plus comment, and the topic need not be an argument of the verb. Mandarin, Lahu.
- Both: subject and topic are separately marked. Japanese, with wa for topic against ga for subject; Korean.
- Free order with pragmatic function: case marking releases word order for given-before-new ordering. Russian, Polish, Turkish.
- Dedicated focus marking: a position or particle reserved for focus. Hungarian and Turkish, both using the slot immediately before the verb.
The values overlap heavily, which is why the parameter resists a clean count. No balanced survey of information structure exists comparable to the word order chapters of the World Atlas of Language Structures, so no distribution figure can be given. The free-order pattern travels with rich case marking, and the fixed-order pattern with its loss.
Consequences for translation
The trap here is that keeping the source order preserves the wrong thing. In Russian, moving a constituent to the end of the clause marks it as new information; in English the same position is simply where objects go. Reproducing the Russian sequence in English transmits the words and drops the signal, and reproducing English order in Russian can assert something the source did not. The compensating move is an emphasis change: a cleft, a fronting, an existential, a passive or a lexical marker. Choosing one is compulsory whenever the source used order to do work the target cannot.
The loss is invisible in a target-only read. Nothing is ungrammatical and nothing is mistranslated at word level; the text simply emphasizes the wrong element, consistently. This is the most common undiagnosed fault in accurate translations.
Three practical points follow. Revisers need a comparative read here, since the category cannot be caught monolingually, and it is worth naming in a review brief so that it is looked for rather than felt. Segmentation matters: topic chains in Mandarin and Japanese run across sentence boundaries, the topic stated once and then carried silently, so segment-by-segment work loses the thread and forces the translator to restate a topic the source did not restate. And a choice should be fixed per document about how much fronting and clefting the register tolerates, since English legal prose accepts far less than English narrative.
Examples
JA>EN
- Source: 象は鼻が長い。
- Target: Elephants have long trunks.
- Comment: Japanese states a topic, elephant, then comments on it with its own subject, trunk. The literal shape, "as for the elephant, the nose is long", is not English. English must select one participant as subject, and choosing the elephant turns the trunk into a possession rather than a topic.
RU>EN
- Source: Книгу купил Иван.
- Target: It was Ivan who bought the book.
- Comment: Russian case marking lets the object open the clause, making the book given and Ivan new. English cannot reorder without changing who did what, so the new information takes a cleft instead. "Ivan bought the book" is accurate and says something else.
TR>EN
- Source: Bu raporu ben hazırladım.
- Target: I was the one who prepared this report.
- Comment: Turkish normally drops the subject pronoun, so an overt ben standing immediately before the verb is doubly marked as focus. A plain "I prepared this report" is grammatical and drops both signals.
References
- Li, C. N. & Thompson, S. A. (1976). "Subject and Topic: A New Typology of Language." In Li, C. N. (ed.), Subject and Topic. New York: Academic Press, pp. 457–489.
- Lambrecht, K. (1994). Information Structure and Sentence Form: Topic, Focus, and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press.
- É. Kiss, K. (1998). "Identificational Focus versus Information Focus." Language 74(2), pp. 245–273.
- Chafe, W. L. (1976). "Givenness, Contrastiveness, Definiteness, Subjects, Topics, and Point of View." In Li, C. N. (ed.), Subject and Topic. New York: Academic Press, pp. 25–55.