Research into linguistic interference

3.3.  Brenda Malkiel’s Research into Interference 
Brenda Malkiel‟s corpus consists of students‟ translations from Hebrew into 
English. Nine of the students were native speakers of English (translating into 
their L1) and thirteen students were native speakers of Hebrew (translating into 
their L2). She studies the role of directionality as far as interference is 
concerned and she focuses on the students‟ on their way towards proficiency. 
Malkiel takes notice of their progress in a three-semester time period. She says 
that her “study examines the effect of training and experience on product and 
process, asking whether translation students become better able to withstand 
interference and whether the task becomes less difficult with time” (Malkiel 
2006: 338). Brenda Malkiel takes advantage of Translog to analyse the data 
collected and to see the students‟ progress. The two main forms of interference 
which interest her most concern failure to lexicalize and false cognates. In other 
words, Malkiel deals with interference only on the level of words. Lexicalizable 
strings, as Malkiel terms this concept, are similar to what Thorovský calls 
interference in collocation and semantic interference. In other words, it is the 
case of a concept lexicalized in the source language but not in the target, or 
vice versa. If the concept can be expressed using a single word in the SL but 
there is no adequate single-word equivalent in the TL, the translator has to 
express such a concept using a lexicalizable string. On the other hand, if some 
concept is expressed by a string of words in the SL, but it has a suitable one-
word expression in the TL, the translator should use it. According to 

---

 
34 
Hopkinson‟s claim, the factor causing this type of interference would probably 
be exact syntactic equivalence in lexical interference. The second type of 
interference investigated in Malkiel‟s paper is the occurrence of false cognates. 
Unfortunately, we will not provide concrete examples from Brenda Malkiel‟s 
research because she examines translations from Hebrew. Moreover, instances 
of both of these types of interference have virtually been mentioned in the 
previous chapters. Comparing these two types of interference in students‟ 
translations, Malkiel expresses an important statement: “Whereas there is some 
debate as to whether failure to lexicalize constitutes an actual mistake, the 
consensus is that it is incorrect to translate a false cognate by sound rather than 
by meaning” (Malkiel 2006: 340). Malkiel also inquires into the question of 
students‟ awareness of the difficulty of a text for translation and of potential 
problems that may arise. She obtained the answers concerning this issue from 
questionnaires which she distributed after the students had completed the first 
translation task. 
As far as the results from Malkiel‟s research are concerned, “as predicted, 
performance on the lexicalizable strings and the false cognates significantly 
improved between administrations for both the native English and the native 
Hebrew speakers” (Malkiel 2006: 354). The study revealed that the students 
became aware of possible interferences and they confirmed this fact in their 
questionnaires. The improvement was evident; nevertheless, this did not mean 
that interference disappeared completely from their translations. Brenda Malkiel 
confirmed the claim that the amount of interference was dependent on the 
professional experience of a translator and the quality of translations (from the 
point of view of interferences occurring in target texts) could be improved by 

---

 
35 
training. Results regarding directionality show that “translation students 
translating into L1 are better able to avoid interference than their classmates 
working into L2” (Malkiel 2006: 356). This result only verified the universally 
assumed hypothesis. 
To conclude, Brenda Malkiel studies interference in students‟ translations 
over a longer period of time and thus inquires into the question of experience in 
relation to the occurrence of interference in final products.