1.5.1. Campbell's Developmental Scheme
Campbell's study (1991) is one of the few contributions that
submitted translation evaluation to empirical study (Waddington, 2001). He
investigates the extent to which translation tests, in this case a public
examination, measure translation competence and account for processes involved
in translations' production. He conducted a case study on a sample of
renditions of a text from Arabic into English by non-native speakers of
English. Campbell (1991) found that specific translation strategies and
linguistic devices characterise every performance level. On this basis, he
worked out a scheme of translation competence development composed of three
stages. Each stage is identified through a number of criteria i.e. indicators
of the subject's evolution. In the following description, each notion in
italics is followed by its definition:
Stage 1:
o Substantial decrease in omissions i.e. "the lack,
for a reason or another, of any target language item corresponding in a source
language item" (p. 332)
Stage 2:
o Increase in word length
o Agreement with peers on lexical transfers i.e.
rendering a source text
lexical item by the standard corresponding one in
the target language.
o Decrease in direct translations i.e. translations
that stick to the source text's form.
o Increase in shifts i.e. " a target language item
that is semantically accurate but grammatically different from the source
language item" (p.332).
o Increase in content words, as opposed
tofunction words.
o More nominal style as opposed to verbal
style.
Stage 3:
o Decrease in text length, i.e. increase in text density.
o Increase in variety of vocabulary.
o Accurate spelling.
Campbell declares that this way of measuring evolution may help,
among other situations, in entrance tests to translation courses. It helps
determine the candidate's level according to the outlined
stages. This will help determine whether or not, starting the learning process
from this point, the candidate is likely to attain translation competence given
available time and instruction (Campbell, 1991, p. 340).
The study seems, however, to be based on linguistic features
of the translations on the detriment of features revealing transfer and problem
solving strategies. Overall translation quality, functional considerations,
coherence and other features of higher textual levels are also not considered
(Waddington, 2001).
1.5.2. Orozco and Hurtado Albir's Model
Construction of measuring translation competence acquisition
instruments has been the central concern of Mariana Orozco's doctoral thesis
(2000). The thesis was directed by M. C. Viladrich and A. Hurtado Albir from
the University of Barcelona in Spain. Orozco and Hurtado Albir describe the
suggested model in their article published in the translation journal Meta
in 2001.
This set of measuring instruments aims to account for
translation competence acquisition through three main aspects of performance.
The model then includes three different tests. Each is expected to measure one
element that the two theorists consider as an observable indicator
of
translation competence acquisition. il is hence
considered as a dependent variable. The first element is the way the subject
deals with translation problems. The second concerns translation errors. And
the third is related to general notions about translation. The two theorists
state that all three are observable, has to do with all the stages of the
translation process and involve the student's use of strategies to solve
translation problems. This is why each element can be reliable as an indicator
of translation competence acquisition.
As mentioned above then, one measuring instrument is designed
to test the subject's behaviour when faced with translation problems
(Translation Problems Instrument). A second measures performance with regard to
translation errors (Translation Errors Instrument). And a third measures
translation notions (Translation Notions Instrument). All three are to be
conducted once at the moment students enter translation course and once at the
end of the first year i.e. after eight months. Individual and group evolution
is then measured.
Translation Notions Instrument is a questionnaire whose aim
is to gain some insight into the knowledge students have about general notions
related to translation. The nature of translation, translation unit, and
translation strategies are some examples. This makes sense as the translator's
decision making is determined by the idea he has about the nature of
translation, its objective, its priorities and the like.
Consequently, the test provides an explanatory background to
the process of translation production of each subj ect.
Translation Problems Instrument investigates whether or not
the student detects the problem, and how he deals with it if he does. The test
contains two parts. The first consists in a task of translating a text
including many types of translation problems that are previously identified by
the evaluator. Each problem involves a skill or knowledge the evaluator seeks
measuring. The text is accompanied by a translation brief that contains useful
information about the text, its purpose, and the evaluator's instructions. A
second part of the test is a questionnaire used to provide the information that
the translation task fails to provide.
The researchers adopt Nord's (1996) perception of translation
errors, which states that it arises from an unsolved or an inappropriately
solved translation problem. Translation Errors Instrument provides the students
with a text to translate. Translations are then corrected and errors
classified. Successful solutions, i.e. instances where the student
appropriately solves a translation problem, are also considered. The
researchers point out that the investigators are free to set error categories
as fits their purpose.
This model, state the researchers, is designed to evaluate
the students' written translations into the mother tongue. They should
therefore be modified to evaluate translations into the foreign language.
Furthermore, whatever modifications are brought, the
instruments may apply to all situations where a teacher or a researcher needs
to measure translation students' progress.
| 
