Accounting for style in machine translation. DiMarco, C. & Hirst, G. In Third International Conference on Theoretical Issues in Machine Translation, Austin TX, June, 1990. abstract bibtex A significant part of the meaning of any text lies in the author's style. Different choices of words and syntactic structure convey different nuances in meaning, which must be carried through in any translation if it is to be considered faithful. Up to now, machine translation systems have been unable to do this. Subtleties of style are simply lost to current machine-translation systems.
The goal of the present research is to develop a method to provide machine-translation systems with the ability to understand and preserve the intent of an author's stylistic characteristics. Unilingual natural language understanding systems could also benefit from an appreciation of these aspects of meaning. However, in translation, style plays an additional role, for here one must also deal with the generation of appropriate target-language style.
Consideration of style in translation involves two complementary, but sometimes conflicting, aims:
- The translation must preserve, as much as possible, the author's stylistic intent — the information conveyed through the manner of presentation.
- But it must have a style that is appropriate and natural to the target language.
The study of comparative stylistics is, in fact, guided by the recognition that languages differ in their stylistic approaches: each has its own characteristic stylistic preferences. The stylistic differences between French and English are exemplified by the predominance of the pronominal verb in French. This contrast allows us to recognize the greater preference of English for the passive voice:
- (a) Le jambon se mange froid.
(b) Ham is eaten cold.
Such preferences exist at the lexical, syntactic, and semantic levels, but reflect differences in the two languages that can be grouped in terms of more-general stylistic qualities. French words are generally situated at a higher level of abstraction than that of the corresponding English words, which tend to be more concrete (Vinay and Darbelnet 1958, 59). French aims for precision while English is more tolerant of vagueness. (Duron 1963, 109).
So, a French source text may be abstract and very precise in style, but the translated English text should be looser and less abstract, while still retaining the author's stylistic intent. Translators use this kind of knowledge about comparative stylistics as they clean up raw machine-translation output, dealing with various kinds of stylistic complexities.
@InProceedings{ dimarco7,
author = {Chrysanne DiMarco and Graeme Hirst},
title = {Accounting for style in machine translation},
booktitle = {Third International Conference on Theoretical Issues in
Machine Translation},
address = {Austin TX},
month = {June},
year = {1990},
abstract = {<P>A significant part of the meaning of any text lies in
the author's style. Different choices of words and
syntactic structure convey different nuances in meaning,
which must be carried through in any translation if it is
to be considered faithful. Up to now, machine translation
systems have been unable to do this. Subtleties of style
are simply lost to current machine-translation systems.</p>
<P> The goal of the present research is to develop a method
to provide machine-translation systems with the ability to
understand and preserve the intent of an author's stylistic
characteristics. Unilingual natural language understanding
systems could also benefit from an appreciation of these
aspects of meaning. However, in translation, style plays an
additional role, for here one must also deal with the
generation of appropriate target-language style.</p> <P>
Consideration of style in translation involves two
complementary, but sometimes conflicting, aims:</p>
<UL><LI> The translation must preserve, as much as
possible, the author's stylistic intent --- the information
conveyed through the manner of presentation. <LI> But it
must have a style that is appropriate and natural to the
target language. </UL> <P> The study of comparative
stylistics is, in fact, guided by the recognition that
languages differ in their stylistic approaches: each has
its own characteristic stylistic preferences. The stylistic
differences between French and English are exemplified by
the predominance of the pronominal verb in French. This
contrast allows us to recognize the greater preference of
English for the passive voice:</p> <UL> <LI> <I>(a)</I> Le
jambon se mange froid.<BR> <I>(b)</I> Ham is eaten cold.
</UL> <P> Such preferences exist at the lexical, syntactic,
and semantic levels, but reflect differences in the two
languages that can be grouped in terms of more-general
stylistic qualities. French words are generally situated at
a higher level of abstraction than that of the
corresponding English words, which tend to be more concrete
(Vinay and Darbelnet 1958, 59). French aims for precision
while English is more tolerant of vagueness. (Duron 1963,
109).</p> <P> So, a French source text may be abstract and
very precise in style, but the translated English text
should be looser and less abstract, while still retaining
the author's stylistic intent. Translators use this kind of
knowledge about comparative stylistics as they clean up raw
machine-translation output, dealing with various kinds of
stylistic complexities.</p>}
}
Downloads: 0
{"_id":{"_str":"534282740e946d920a001ad3"},"__v":10,"authorIDs":["545761622abc8e9f3700033e","5457fff52abc8e9f370009b3"],"author_short":["DiMarco, C.","Hirst, G."],"bibbaseid":"dimarco-hirst-accountingforstyleinmachinetranslation-1990","bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Chrysanne"],"propositions":[],"lastnames":["DiMarco"],"suffixes":[]},{"firstnames":["Graeme"],"propositions":[],"lastnames":["Hirst"],"suffixes":[]}],"title":"Accounting for style in machine translation","booktitle":"Third International Conference on Theoretical Issues in Machine Translation","address":"Austin TX","month":"June","year":"1990","abstract":"<P>A significant part of the meaning of any text lies in the author's style. Different choices of words and syntactic structure convey different nuances in meaning, which must be carried through in any translation if it is to be considered faithful. Up to now, machine translation systems have been unable to do this. Subtleties of style are simply lost to current machine-translation systems.</p> <P> The goal of the present research is to develop a method to provide machine-translation systems with the ability to understand and preserve the intent of an author's stylistic characteristics. Unilingual natural language understanding systems could also benefit from an appreciation of these aspects of meaning. However, in translation, style plays an additional role, for here one must also deal with the generation of appropriate target-language style.</p> <P> Consideration of style in translation involves two complementary, but sometimes conflicting, aims:</p> <UL><LI> The translation must preserve, as much as possible, the author's stylistic intent — the information conveyed through the manner of presentation. <LI> But it must have a style that is appropriate and natural to the target language. </UL> <P> The study of comparative stylistics is, in fact, guided by the recognition that languages differ in their stylistic approaches: each has its own characteristic stylistic preferences. The stylistic differences between French and English are exemplified by the predominance of the pronominal verb in French. This contrast allows us to recognize the greater preference of English for the passive voice:</p> <UL> <LI> <I>(a)</I> Le jambon se mange froid.<BR> <I>(b)</I> Ham is eaten cold. </UL> <P> Such preferences exist at the lexical, syntactic, and semantic levels, but reflect differences in the two languages that can be grouped in terms of more-general stylistic qualities. French words are generally situated at a higher level of abstraction than that of the corresponding English words, which tend to be more concrete (Vinay and Darbelnet 1958, 59). French aims for precision while English is more tolerant of vagueness. (Duron 1963, 109).</p> <P> So, a French source text may be abstract and very precise in style, but the translated English text should be looser and less abstract, while still retaining the author's stylistic intent. Translators use this kind of knowledge about comparative stylistics as they clean up raw machine-translation output, dealing with various kinds of stylistic complexities.</p>","bibtex":"@InProceedings{\t dimarco7,\n author\t= {Chrysanne DiMarco and Graeme Hirst},\n title\t\t= {Accounting for style in machine translation},\n booktitle\t= {Third International Conference on Theoretical Issues in\n\t\t Machine Translation},\n address\t= {Austin TX},\n month\t\t= {June},\n year\t\t= {1990},\n abstract\t= {<P>A significant part of the meaning of any text lies in\n\t\t the author's style. Different choices of words and\n\t\t syntactic structure convey different nuances in meaning,\n\t\t which must be carried through in any translation if it is\n\t\t to be considered faithful. Up to now, machine translation\n\t\t systems have been unable to do this. Subtleties of style\n\t\t are simply lost to current machine-translation systems.</p>\n\t\t <P> The goal of the present research is to develop a method\n\t\t to provide machine-translation systems with the ability to\n\t\t understand and preserve the intent of an author's stylistic\n\t\t characteristics. Unilingual natural language understanding\n\t\t systems could also benefit from an appreciation of these\n\t\t aspects of meaning. However, in translation, style plays an\n\t\t additional role, for here one must also deal with the\n\t\t generation of appropriate target-language style.</p> <P>\n\t\t Consideration of style in translation involves two\n\t\t complementary, but sometimes conflicting, aims:</p>\n\t\t <UL><LI> The translation must preserve, as much as\n\t\t possible, the author's stylistic intent --- the information\n\t\t conveyed through the manner of presentation. <LI> But it\n\t\t must have a style that is appropriate and natural to the\n\t\t target language. </UL> <P> The study of comparative\n\t\t stylistics is, in fact, guided by the recognition that\n\t\t languages differ in their stylistic approaches: each has\n\t\t its own characteristic stylistic preferences. The stylistic\n\t\t differences between French and English are exemplified by\n\t\t the predominance of the pronominal verb in French. This\n\t\t contrast allows us to recognize the greater preference of\n\t\t English for the passive voice:</p> <UL> <LI> <I>(a)</I> Le\n\t\t jambon se mange froid.<BR> <I>(b)</I> Ham is eaten cold.\n\t\t </UL> <P> Such preferences exist at the lexical, syntactic,\n\t\t and semantic levels, but reflect differences in the two\n\t\t languages that can be grouped in terms of more-general\n\t\t stylistic qualities. French words are generally situated at\n\t\t a higher level of abstraction than that of the\n\t\t corresponding English words, which tend to be more concrete\n\t\t (Vinay and Darbelnet 1958, 59). French aims for precision\n\t\t while English is more tolerant of vagueness. (Duron 1963,\n\t\t 109).</p> <P> So, a French source text may be abstract and\n\t\t very precise in style, but the translated English text\n\t\t should be looser and less abstract, while still retaining\n\t\t the author's stylistic intent. Translators use this kind of\n\t\t knowledge about comparative stylistics as they clean up raw\n\t\t machine-translation output, dealing with various kinds of\n\t\t stylistic complexities.</p>}\n}\n\n","author_short":["DiMarco, C.","Hirst, G."],"key":"dimarco7","id":"dimarco7","bibbaseid":"dimarco-hirst-accountingforstyleinmachinetranslation-1990","role":"author","urls":{},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"www.cs.toronto.edu/~fritz/tmp/compling.bib","downloads":0,"keywords":[],"search_terms":["accounting","style","machine","translation","dimarco","hirst"],"title":"Accounting for style in machine translation","year":1990,"dataSources":["n8jB5BJxaeSmH6mtR","6b6A9kbkw4CsEGnRX"]}