French

This is the second half of a study on the advisability of having undergraduates perform an exercise in post-editing (PE) on output from machine translation (MT). After a sketch of its chequered history, I take up the various ways in which students could be introduced to MT, and conclude that PE is the most useful, mainly because it makes them aware of MT's limitations and fundamental language properties underlying these limitations. Through a review of the literature on PE practices and challenges, especially a psycho-linguistic study by Krings , I highlight crucial differences between professional PE and our pedagogical experiment. An error analysis of the student PE results (English to French) is then compared with analyses of the MT input, student translations of the same text from scratch, and parallel results from the initial, Frenchto-English experiment. Finally, student comments on the exercise suggest that weaker students especially find it beneficial.

This paper describes an experiment whose goal was to introduce MT via post- editing to a third-year university class in French-to-English translation. Even though most of the students in our French programs do not go on to become professional translators, it seemed a worthy objective to give the2m at least a taste of the challenges and opportunites which MT offers today, given how widespread it has become in the commercial and techni- cal sector. It would also be valuable to know to what extent a brief hands-on encounter with MT would help them better understand the nature of translation. After providing back- ground on the course content, level of students and type of documents chosen for the ex- periment, I briefly look at the question of how to "teach" MT and give reasons for the focus on post-editing. I then outline the experiment itself, including the process of choosing an MT program, the type of texts, the error analysis adopted for evaluating both human and MT outp...

Inalienable possession (iposs) in Mandarin Chinese has traditionally been thought restricted to associative (genitive) phrases where the possessor is juxtaposed to the possessum. In addition to such phrases, this analysis looks at five other possibilities where intrinsically relational nouns arise: zero anaphora; double subjects; passive of bodily effect; discontinuous possessor construction; and retained object construction. In each case, the discussion involves meaning differences with regard to the genitive construction and the semantics of participating nouns. Although body parts and kin terms, the most typical inalienables, are each found in nearly all six structures, the presence of peripherally inalienable terms such as personal effects and even inalienable tokens casts doubt on the possibility of a formal definition for Mandarin iposs. This is one more instance of the non-bi-uniqueness of iposs, a cross-linguistic category that continues to resist a rigorous morpho-syntactic...

In terms of structure and prototype sub-classes, French and Mandarin present virtually no common ground in their treatment of inalienable possession (iposs). Most of French iposs grammar involves body parts or personal traits much more than kin terms. Mandarin iposs is more diffuse: whether kin terms or body parts rank higher on the prototypicality scale depends on the criteria selected. For associative phrases, Chappell/Thompson 1992 and others have found kin terms to have a significantly higher incidence of de omission than body parts. Other structures like the Passive of Bodily Effect (Chappell 1986) and Double Nominatives give body parts the edge.

While syntax is a crucial part of understanding inalienable possession, it must be seen as interacting with lexical semantics and pragmatics. A case in point is the French double dont construction (DOC), which differs from regular genitive relatives in having two NPs which are linked to the antecedent of the complementizer dont. Tellier (1991) claims that the second N P must be either deverbal or inalienable. This paper will examine 3 aspects of Tellier's work that call for further study. First is the argument/adjunct dichotomy which she proposes as a means of distinguishing relational from alienable nouns in the lexicon. It is argued here, on both empirical and theoretical grounds, that such a dichotomy presents too many problems to be adopted for the lexicon in a simple, binary fashion. Secondly, the special status of the definite determiner, which Tellier treats as a flag of inalienability for the 2nd possessum of DOCs, is questioned. The existence of other determiners in that position, as well as the possibility of a generic (i.e., non-possessive) reading for Ns occurring with Ie, suggest a less automatic link between inalienability and Ie. Finally, constraints on possessor construal are examined, an issue going beyond DOCs. Tellier, among others, contends that a relational noun can have a null possessor only in generic contexts, a claim that applies, as she admits, exclusively to isolated 5s, i.e., independently of any discourse or pragmatic context. We can avoid such an extreme requirement by viewing counter-examples to this constraint as pragmatically marked, invariably linked to special contexts such as a medical scenario. Such markedness alleviates the need for explicit local mention of a possessor. These three issues pose significant but not insurmountable obstacles for handling inalienable possession within a G-B framework. patient access to their intuitions, both grammatical and stylistic, my thanks to Pierre Hervy, Caroline Bayard, Madeleine Jeay, and Susanne Crosta. I am also grateful to Diane Massam and Yves Roberge for their insightful comments. Errors and omissions are, of course, my own responsibility.

Linguistic theory and translation theory both deal with language; however, they rarely meet or use each other's results in order to advance their individual areas of research. Linguists often seem to look at translation as either trade or art rather than science, and translators show cynicism about linguistic inquiry ignoring real language data. This paper focuses on one particular area of concern in both linguistics and translation: how to incorporate pragmatics into an explanation of which translation or interpretation is best for a given linguistic expression in a given linguistic and extra-linguistic context? Students in practical translation classes do not appreciate explanations along the lines "this is simply how you would say it in language X" or "this is what the speakers of X would say in this situation". Speakers of X are balancing their knowledge of rules and conventions of language use with pragmatic know-how; they are making choices that transla...

While syntax is a crucial part of understanding inalienable possession, it must be seen as interacting with lexical semantics and pragmatics. A case in point is the French double dont construction (DOC), which differs from regular genitive relatives in having two NPs which are linked to the antecedent of the complementizer dont. Tellier (1991) claims that the second NP must be either deverbal or inalienable. This paper will examine 3 aspects of Tellier's work that call for further study. First is the argument/adjunct dichotomy which she proposes as a means of distinguishing relational from alienable nouns in the lexicon. It is argued here, on both empirical and theoretical grounds, that such a dichotomy presents too many problems to be adopted for the lexicon in a simple, binary fashion. Secondly, the special status of the definite determiner, which Tellier treats as a flag of inalienability for the 2nd possessum of DOCs, is questioned. The existence of other determiners in that...

The orthoconcept was introduced into Hard Science Linguistics as a way to describe properties of a physical reality formed in an observer. An orthoconcept is not a concept in a logical-domain sense but attempts to explain a person’s understanding in physical-domain terms. An orthoconcept thus is a theoretical representation of a property of a person communicating; it does not necessarily correspond to something external to that person. We consider how orthoconcepts arise as a result of a combination of external events and the expectations of the person. The different understandings of a believer and a skeptic at a séance are used to illustrate how this can occur.

This paper describes an experiment whose goal was to introduce MT via postediting to a third-year university class in French-to-English translation. Even though most of the students in our French programs do not go on to become professional translators, it seemed a worthy objective to give them at least a taste of the challenges and opportunites which MT offers today, given how widespread it has become in the commercial and technical sector. It would also be valuable to know to what extent a brief hands-on encounter with MT would help them better understand the nature of translation. After providing background on the course content, level of students and type of documents chosen for the experiment, I briefly look at the question of how to “teach” MT and give reasons for the focus on post-editing. I then outline the experiment itself, including the process of choosing an MT program, the type of texts, the error analysis adopted for evaluating both human and MT output of those texts, ...

Inalienable possession (iposs) in Mandarin Chinese has traditionally been thought restricted to associative (genitive) phrases where the possessor is juxtaposed to the possessum. In addition to such phrases, this analysis looks at five other possibilities where intrinsically relational nouns arise: zero anaphora; double subjects; passive of bodily effect; discontinuous possessor construction; and retained object construction. In each case, the discussion involves meaning differences with regard to the genitive construction and the semantics of participating nouns. Although body parts and kin terms, the most typical inalienables, are each found in nearly all six structures, the presence of peripherally inalienable terms such as personal effects and even inalienable tokens casts doubt on the possibility of a formal definition for Mandarin iposs. This is one more instance of the non-bi-uniqueness of iposs, a cross-linguistic category that continues to resist a rigorous morpho-syntactic...

Abstract. This paper describes an experiment whose goal was to introduce MT via post-editing to a third-year university class in French-to-English translation. Even though most of the students in our French programs do not go on to become professional translators, it seemed a worthy objective to give the2m at least a taste of the challenges and opportunites which MT offers today, given how widespread it has become in the commercial and techni-cal sector. It would also be valuable to know to what extent a brief hands-on encounter with MT would help them better understand the nature of translation. After providing back-ground on the course content, level of students and type of documents chosen for the ex-periment, I briefly look at the question of how to “teach ” MT and give reasons for the focus on post-editing. I then outline the experiment itself, including the process of choosing an MT program, the type of texts, the error analysis adopted for evaluating both human and MT output ...

Cet article étudie les rapports entre humains et animaux mis en scène dans l’œuvre d’Élise Turcotte, notamment dans Pourquoi faire une maison avec ses morts et Autobiographie de l’esprit. Les animaux familiers offrent aux narratrices une consolation et un accompagnement dans la souffrance, ce qui provoque chez elles une identification avec les animaux malades ou morts à qui elles sont confrontées.