Erich Steiner

Translation units and grammatical shifts: Towards an integration of product- and process-based translation research

Drawing on corpus-based and process-based approaches, this paper reports on the results of an exploratory study using highly annotated translation corpora in conjunction with key logging, eye tracking, and retrospective verbalizations to identify translation units associated with cognitive effort during a translation task. The annotated corpora are used to analyze the grammatical shifts correlated with the processes of (de)metaphorization that occur during translation. The results of the corpus analysis are then triangulated with process data from an analysis of the performance of two German-English translators while carrying out an inverse translation task. Together the process and product data clearly explicate the nature of the translation units associated with (de)metaphorization during the cognitive processing of the translation.

Cross-linguistic corpora for the study of translations : insights from the language pair English-German

This book is a contribution to the study of translation as a contact variety, and - more generally - to language comparison and language contact with a focus on English and German. Methodologically, the contributors implement an empirical research strategy with further applications in computational linguistics and language technology.

The Eurotra linguistic specifications: An overview

In this article, we outline the contents of the linguistic specifications of the Eurotra machine translation system. We start in sections 1 and 2 from some of the requirements placed by multilingual MT on the overall design of the linguistic components. We then move on to a characterization of the Eurotra interface structure (section 3), the nature of transfer (section 4), and trends towards more interlingual representations within the project (section 5). Thereafter, we concentrate on the contents of the various levels beside the interface structure (section 6) before giving a brief survey of word structure (section 7) and outlining some areas for further research (section 8)The authors of this article are indebted to many other members of the project too numerous to be mentioned here. They wish to record a special intellectual debt to previous members of the Eurotra Linguistics Specification team and, in particular, Doug Arnold, Louis des Tombe and Lieven Jaspaert who did so much to establish sound theoretical bases for multilingual MT (see inter alia Arnold, Jaspaert and des Tombe 1985; Arnold 1986; Arnold and des Tombe 1987). For an extensive version of the overview presented here, see Allegranza et al. 1991. For another recent presentation of Eurotra, see Raw, Vandecapelle and Van Eynde 1988..

Grammatical metaphor in translation – some methods for corpus-based investigations

The focus of this paper will be on investigating textual properties of translations, with special reference to English-German. A number of methods will be critically discussed for empirically testing hypotheses about differences between source language texts and target language texts in cases of translations, as well as between translated and original texts in the same language. We shall then proceed to formulate additional hypotheses about textual properties of translated texts, by modelling in more detail how the process of understanding and unpacking grammatical structures in the process may manifest itself as an independent factor. This will be followed by an analysis of an example of translated text, applying to it some techniques for quantification of lexicogrammatical properties, the aim being to show whether and to what extent there is hope for a closing of the gap between relatively high-level hypotheses and ’raw data’ in electronic corpora.

On the semantics of focus phenomena in Eurotra

In this paper, we discuss issues connected to the phenomenon of linguistic FOCUS or INFORMATION DISTRIBUTION in the sentence in the context of the multi-lingual machine translation project EUROTRA. We shall present some of the arguments why a consideration of FOCUS phenomena is important for the determination of linear order and for semantic interpretation. We shall proceed, in sections 2 and 3 of the paper, to mention the main lines of development in the dicussion of FOCUS phenomena in Computational Linguistics and in Linguistics respectively. Section 4 contains an illustration of a pilot implementation covering some aspects of FOCUS phenomena in EUROTRA-D.

Some Properties of Lexicogrammatical Encoding and Their Implications for Situations of Language Contact and Multilinguality

ZusammenfassungIn diesem Beitrag möchte ich eine multi-funktionale und merkmalbasierte Perspektive auf Situationen von Sprachkontakt, Mehrsprachigkeit und besonders Übersetzung skizzieren, in deren Methodik die Kodierungseigenschaften von Explizitheit, Dichte und Direktheit von Konstruktionen als Dimensionen textueller Variation ins Blickfeld treten.Ich beginne mit einer kurzen Begriffsklärung, in der ich die Begriffe von »Sprachkontakt« und »Mehrsprachigkeit«, wie sie in der Linguistik eingeführt sind, in Erinnerung rufe. Danach folgt die Skizzierung einer im engeren Sinne multi-funktionalen und merkmalbasierten Perspektive hinsichtlich ihrer Methodik und ihrer Erkenntnisobjekte. Im Rahmen einer solchen Perspektive wird Wert zu legen sein auf das System gemeinsam mit der Struktur, auf die Instanziierung ebenso wie auf das System, auf abstraktere (und gleichzeitig stärker empirische) Arten von Kontrast, als sie bisher ins Blickfeld der Forschung traten, und letztlich auf die metafunktionale Modularisierung linguistischer Struktur. Durch eine Referierung bereits vorliegender Arbeiten soll gezeigt werden, wie eine solche Perspektive die bereits bekannten eher strukturbasierten Modellierungen erweitern und bereichern kann.In einem zweiten Schritt wird die skizzierte multi-funktionale und merkmalbasierte Perspektive operationalisiert, wobei wir die Grundzüge einer Korpusarchitektur und eines Schemas zur multilingualen Textanalyse mit besonderer Berücksichtigung englisch-deutscher Übersetzungen entwerfen.Unsere Vorschläge zielen auf die Untersuchung multilingualer Diskurse, nicht so sehr mit Blick auf Entlehnungsprozesse auf der Ebene lexikalischen oder strukturellen Materials, sondern eher im Sinne von Eigenschaften lexikalischer und grammatischer Konfigurationen, die indirekt Folge von Kontaktsituationen sind. Unser Interesse gilt den Gebrauchsweisen systemischer Ressourcen (also von Grammatik, Lexikon, Kohäsion), die von ›muttersprachlichen‹ Diskursen entlang der Dimensionen von Explizitheit, Dichte oder Direktheit abweichen und so zu Diskursen und Texten führen, die auf interessante Weise anders, bemerkenswert, ungewöhnlich und potentiell kreativ sind.

A new view on the process of translation

In this paper we describe a framework for research into translation that draws on a combination of two existing and independently constructed technologies: an analysis component developed for German by the EUROTRA-D (ET-D) group of IAI and the generation component developed for English by the Penman group at ISI. We present some of the linguistic implications of the research and the promise it bears for furthering understanding of the translation process.

Interfacing an English text generator with a German MT analysis

In this paper we investigate a strategy for using an existing English text generator to serve as the generation module in a machine translation (MT) system for translating German into English. The English text generator has been designed to take requests for text construction in a specified interface language and, making reference to domain knowledge as necessary, to produce appropriate multi-sentential English text; this generator is therefore said to be knowledge based. In contrast, the MT analysis result that is used as input to the translation process is based upon semanticaily-interpreted canonical dependency structure (interface structure), there is no direct link to domain knowledge of any kind. The conceptual link that makes the co-operation possible is a particular interpretation of linguistic information as a kind of (linguistic) knowledge base. The characteristics necessary for an interface between two such systems represent a particular challenge, but they also constitute the main interest in the project, implying that projects can be successfully interfaced, providing there are appropriate conceptual links between the systems involved. These links mainly have to do with the embodied notions of linguistic structure. Such links also appear to lead to useful observations concerning the theory of (machine) translation.

Producers-users-customers: Towards a differentiated evaluation of research in Machine Translation

The following remarks are a brief reaction to Somers’ “current research in machine translation”. Because Somers’ article is available to readers of the present volume, I shall not summarize the ideas put forward there. Instead, I shall suggest some views on the evaluation of MT research essentially claiming that Somers’ remarks are relevant and interesting, but would gain substantially by being embedded inside a somewhat wider and more differentiated perspective on the field in question. The beginnings of such a wider perspective might be seen in the two questions (1.1. and 1.2.) which, I suggest, should be put to three communities of people involved in MT (2.1.-2.3.). 1.1. How do you evaluate the history of MT to date? 1.2. Which (changes in) directions do you suggest? 2.1. Researchers 2.2. Translators 2.3. Companies, institutions, and individuals buying MT systems My argument would be that it is only through such a differentiated view that we shall be able to arrive at a realistic assessment and at strategies for MT-research which are not unduely self-interested, nor ill-informed by the exclusion of relevant commmunities.

Some remarks on a functional level for machine translation

Abstract The issues discussed in this paper are differences in phrase structure and difficulties in structural vs. lexical transfer as problems for machine translation. We discuss existing solutions and their problems, specifying advantages of a functional approach. Finally, we discuss one possible way towards the definition of a level of representation of functional information.