Comparative-contrastive analysis of Romanian-English translation of verb phrase structures.

Comparative-contrastive analysis of Romanian-English

translation of verb phrase structures

Laura SASU1

Abstract: The underlying study provides analyses of common

translation inadequacies that originate in source language
interference upon target language text. Investigating the occurrence of
detrimental crosslinguistic interference in translation, the linguistic
levels involved and the impact upon meaning/phrasing provides data
relevant for the translating-revising process, thus improving
translation results. Contrastive-comparative analysis applied in
language teaching and translation studies becomes an important
instrument for anticipating specific difficulties that produce
predictable interlingual errors, originating in the source language
transfer to the target language. High risk translation errors originate
in the misperception of semantic equivalence that can be traced back
to merely formal correspondence concerning verb tenses, voice, mood,
expressing non-factuality or factual remoteness. Therefore, contrastive
analysis of source and target language structures benefit translation
theory and practice by predicting or identifying erroneous source
language biased constructions.

1
Transilvania University of Braşov, laurasasu@gmail.com
Laura SASU

Key-words: translation, contrastive linguistics, source language

interference, equivalence.

1. Theoretical framework

1.1. Contrastive linguistics

Contrastive linguistics started out as subdomain of applied linguistics advancing

the hypothesis that studying the influence of the native tongue upon foreign
language acquisition, also referred to as ‘interlingual interference’, might become a
valuable instrument for generating more adequate language teaching methods and
materials. Therefore, the systematic comparative-contrastive analysis of two
languages aims at identifying predictable difficulties either by theoretical forecast
or by conducting methodical ‘error analyses’ on existing samples of language
transfer. Continuing in the tradition of Robert Lado (1957) and Charles C. Fries
(1945), more recent contrastive studies extend the scope of comparative-contrastive
analysis from language learning/teaching to translation studies. Such theoretical
contributions on translation provide “a correct approach to the process of
translation, enriching the translator with translation strategies and techniques”
(Arhire 2014: 185). Anthony Pym’s (2010) theoretical contribution, especially on
concepts regarding equivalence and risk analysis in translation, highlights the fact
that, on one hand, mistranslating certain language structures impact meaning more
severely than mistranslating others and, on the other hand, that there are predictable
interlingual transfer errors for each language pair translation. From this point of
view, verb phrase structures are of utmost importance for accurate transfer of
meaning, while erroneous translations thereof can produce significant alteration of
core messages. Thus, verb phrases can be described as linguistic units that are
likely to produce high risk translation errors by inaccurate transfer of information
regarding tense, time, aspect, voice, factuality or factual remoteness, mostly due to
source language interference in Romanian-English translations/learner language.

1.2. Methodology

The methodology used in this paper implies specific translation error analysis
(formal or semantic mistakes or inadequacies originating in source language
interference upon translation) by providing examples of the incorrect versus
[correct] translation of the source language word/term/phrase (marked as such in
the text).
The erroneous translation examples (that are recurrent in Romanian to
English translation/learner language) are listed first to provide a clear outlook on
 Comparative-contrastive analysis of Romanian-English translation of verb phrase structures.

the actual misphrasing occurring in Romanian to English translation/learner

language.
In the second stage of the analytical process, the correct form for each
example is added in square brackets, marked as [correct form], to pinpoint the
specific difference between the incorrect and the correct form.
The third step implies listing the Romanian verb(phrase) in italics for each
phrase to provide further information on whether the misphrasing originates in
applying a pattern of the source language to the target language or not.
Subsequently, possible causes of specific inaccuracies are investigated by
means of comparative-contrastive analysis concerning Romanian/English language
structures, in order to establish whether these errors can be explained by source
language interference upon target language.
If so, such errors can be coined as predictable errors, and
translating/language learning/teaching become more adequate by prompting the
findings of such analyses as warning signs for predictable traps posed by specific
language structures.

2. Contrastive linguistics in language learning/teaching and translating

Both translation studies and language teaching/learning apply the basic concepts of
contrastive linguistics, ultimately aiming at the correct use of language and the
avoidance of message distortion.
Communicating in a foreign language often implies translation (here - the
translation process itself) both from and into that specific language. Therefore the
target language of this translating process, or ‘learner language’ (Johansson 2008:
112), as it is referred to in contrastive linguistics, is predictably influenced by the
source/native language.

2.1. Meaning and context

Decoding meaning and re-encoding it according to the target language structure

frequently implies (de/re)contextualisation of semantic units that may have distinct
meanings in different contexts. Meaning (and therefore message
decoding/encoding and translation) is context-bound. Translating the meaning of
simple words, like ‘scriu’ or ‘beam’, may have several different results when
inserted into distinct contexts. Source language interference might prompt ‘I work ‘
or ‘rază’ as the first options, but there are several other possibly correct answers,
the most accurate choice depending exclusively on the context:

(1) scriu – I write, I am writing, I have been writing

Laura SASU

(2) beam – rază, fascicul, grindă, buiandrug

Finding suitable contexts for each translation option is further explanatory for the
relationship word/meaning/context/translation/efficient communication. In order to
provide an accurate translation, the word/term/phrase needs to be decoded
(interpreted) in context.

Translation needs to be focused on equivalence in meaning, not on finding

corresponding words, because language is only the interface linking meaning to
specific linguistic units, as follows:

MEANING (concept)<–>LANGUAGE(code)<–>WORD/TERM (linguistic unit).

Therefore, translating a word/term/text is rather about conveying the meaning of A

by B (using the code of a different language), and not about providing a formal
counterpart of A in B.

2.2. Translation

The term translation is used both for the product of the translating process and for
the process itself (Bell 1991:13). The translating process deals with conveying
messages by decoding (meaning) from Source Language and encoding in the
Target Language, aiming at equivalence.

Term A CONCEPT Term B

Source Language MEANING Target Language
decoding MESSAGE encoding

The ideal result of the translation process is complete or total equivalence of the
translation product, when compared to the original. However, “absolute”
equivalence remains a purely theoretical and rather illusory concept, while
concepts such as “poor/good/better” translations are more realistic, despite the
inherent relativity of such defining adjectives.

2.2.1. Formal Equivalence vs. Semantic Equivalence

Poor translations usually result from the preference for formal equivalence over
semantic equivalence.
 Comparative-contrastive analysis of Romanian-English translation of verb phrase structures.

(3) Electronically-controlled pumps measure out drugs for the chronically

ill.

Translating the source language term A ‘drugs’ using the target language term B
‘drogurile’, instead of [medicația] in this context, produces a severe shift in
meaning, since (in current use) the Romanian corresponding term is more
frequently associated to illegal substances rather than medication. As seen in the
example above, quite often the original meaning (message) is significantly altered
or even annulled by preserving the exact sentence structure, word order, choice of
words or means of expression. Translations that are severely biased by the source
language text are usually an artificial sequence of words, difficult to follow and
even more difficult to comprehend. In addition to that, they may include serious
changes of meaning or may come into the proximity of non-language and non-
sense. Examples of such translations are unfortunately quite common nowadays,
being commonly referred to as word-for-word translations.

2.2.2. Literal (word for word) Translation vs. Free Translation

On the other hand, machine translation capabilities improve by the day, but
translation software is still far from being sufficient or reliable for obtaining
accurate translations.

(4) Foremost among the avenues now being pursued are the design of Very
Large Scale Integration (VLSI) and new computer architectures.

The following four translation options emphasize the shortcomings of

literal/word-for-word/machine translation vs. the semantic equivalence attained by
means of free translation.

(a) Înainte de toate (foremost) printre (among) bulevardele (avenues) acum

(now) urmărite (pursued) sunt (are) designul (the design) de foarte mare
scală integrată (very large scale integration) şi noi calculatoare de
arhitectură (new computer architectures). – word-for-word translation: not
acceptable.
(b) Cel mai important dintre căile acum fiind urmărite sunt proiectarea de
integrare Very Large Scale (VLSI) și noi arhitecturi de calculatoare. –
machine-translation: implies major revising.
(c) Printre direcțiile abordate cu precădere în momentul de faţă se numără
proiectarea Integrării la Scară Foarte Mare şi noi modele în arhitectura
calculatoarelor. – acceptable free translation.
Laura SASU

(d) Cele mai actuale (frecventate/vizate/noi) domenii (zone/arii/teme) de

interes includ (vizează/se axează pe/circumscriu) proiectarea VLSI şi noua
(modele noi/moderne/de ultimă generaţie în) arhitectură a calculatorului. –
recommendable free translation.

Analysing the number and variety of changes that had to be made in order to
preserve the meaning and to convey it by understandable, correct, coherent,
adequate means of expression (of the target language) leads to the following
preliminary conclusion: Native/source language interference upon foreign/target
language in language learning and translating is predictable and its negative effects
upon rendering meaning accurately in the target language are, therefore,
amendable. Nevertheless, meaning can be expressed in several ways, so choosing
the most adequate one for each context is what makes translation both a form of
art and a field of science.

3. Verb phrase structures

However, high risk translation errors are often related to interlingual transfer of
language structures. The comparative-contrastive analysis of verb phrase structures
and pattern transfer in Romanian to English translations and learner language
provides a map of the most recurrent and thus predictable types of misphrasing that
occur due to source language interference. Awareness of the likelihood of such
errors to occur (due to the logical validation by the native/source language structure
that is being transferred) becomes a powerful instrument for language
learning/teaching and translation studies, forecasting specific points of difficulty
and opening alternatives for solution identification. The correct form in English is
frequently replaced with an erroneous one due to the Romanian language structure
that is being reproduced by interlingual transfer.

3.1. Tenses

When translating from Romanian into English, the Source Language interference
produces overlapping in the use of tenses of the Target Language. The same form
in Romanian needs to be translated into English in two, three or even four different
ways, depending on the context/meaning. Semantic context-bound translation is
mandatory (considering the semantic implications of time and aspect), whereas the
purely formal translation of the verb leads either to flawed phrasing or to divergent
meaning. Overlapping and/or substitution of several English tenses mainly
originate in the existence of just one corresponding Romanian form. Relevant
information on aspect and time is provided in Romanian by the context (time
 Comparative-contrastive analysis of Romanian-English translation of verb phrase structures.

adverbs), while in English it is often included within the verb (tense-time/aspect).

The following analysis lists the correct form and correspondence, identifies the
predictable mistakes and investigates whether the error is accounted for by source
language structure transfer.

3.1.1. Present Simple [Present Continuous/Present Perfect Continuous]

(5) I write every day. Scriu în fiecare zi.

(6) I write [am writing] now. Scriu acum.
(7) I write [have been writing] for three hours. Scriu de trei ore.

3.1.2. Present Continuous [Present Simple/Present Perfect Continuous]

(8) I am writing right now. Scriu acum.

(9) I am writing [write] a lot. Scriu mult.
(10) I am writing [have been writing] for three hours. Scriu de trei ore.

3.1.3. Present Perfect [Past Simple/Past Perfect]

(11) I have already written the text. Am scris textul deja.

(12) I have written [wrote] a lot yesterday. Am scris mult ieri.
(13) I have written [had written] everything by the time you arrived. Am
scris totul până să ajungi.

3.1.4. Past Simple [Present Perfect/Past Perfect]

(14) I wrote a lot yesterday. Am scris mult ieri.

(15) I wrote [have written] three letters so far. Am scris trei scrisori până
acum.
(16) I wrote [had written] everything by the time you arrived. Am scris totul
până să ajungi.

3.1.5. Past Continuous [Past Perfect Continuous]

(17) I was writing when you arrived. Scriam când ai sosit.

(18) I was writing [had been writing] for three hours when you arrived.
Scriam de trei ore când ai sosit.

3.1.6. Future Simple [Future Continuous/Future Perfect/Future Perfect continuous]

Laura SASU

(19) I will write tomorrow. Voi scrie mâine.

(20) I will write [will be writing] tomorrow at 6. Voi scrie mâine la ora
șase.
(21) I will write [will have written] everything by 6 o’clock tomorrow. Voi
scrie totul până.
(22) I will write [will have been writing] for ten hours when you return. Voi
scrie de zece ore când te vei întoarce.

3.1.7. Future Continuous [Future Perfect Continuous]

(23) I will be writing when you arrive. Voi scrie când vei ajunge.
(24) I will be writing [will have been writing] for hours when you arrive.
Voi scrie de ore întregi când vei ajunge.

English (Tenses) Romanian (Timpuri verbale)

Present simple write Scriu Prezent
Present am writing Scriu Prezent
continuous
Present perfect have written am scris Perfect compus
Present perfect have been Scriu Prezent
continuous writing
Past simple wrote am scrise/scrise Perfect compus,
perfect simplu
Past continuous was writing Scriam Imperfect
Past perfect had written scrisesem/ Mai mult ca
am scris perfectul/perfect
compus
Past perfect had been writing Scriam Imperfect
continuous
Future simple will write voi scrie Viitor
Future will be writing voi scrie Viitor
continuous
Future perfect will have written voi fi scris/ Viitor
voi scrie anterior/viitor
Future perfect will have been voi scrie Viitor
continuous writing

Since there is no one-to-one correspondence between Romanian and English

tenses, erroneous translations include either tenses used incorrectly in particular
 Comparative-contrastive analysis of Romanian-English translation of verb phrase structures.

contexts or correctly formulated messages, flawed by divergent semantic

implications that are inconsistent with the original message.

5. Conclusions

Contrastive analysis applied in foreign language learning/teaching and in

translation theory and practice becomes a valuable instrument for developing the
skills necessary for efficient communication in a foreign language.
The theoretical and practical aspects discussed above are intended to serve
as both notional and illustrative framework for foreign language teaching and
translation practice. The parallel analysis of language structures implies a
comparative-contrastive point of view that is applied to (incorrect/correct)
translation examples, focusing on error analysis, cause identification and synthesis,
ultimately emphasizing the predictability of typical errors in learner language and
translation.
Awareness of the high probability for such inadequacies to occur turns out
to be useful in language learning/teaching activities, as well as in translating-
proofreading-revising processes, since the influence (interference) of the source
language upon the target language translation product is already strong enough to
produce inadequate phrasing when translating from a foreign language into the
native language and, even more so, when translating from the mother tongue into a
foreign language or between two foreign languages.

References

Arhire, Mona. 2014. Corpus-based Translation for Research, Practice and

Training. Iaşi: Institutul European.
Bell, Robert T. 1991. Translation and Translating: Theory and Practice. London
and New York. Longman.
Fries, Charles C. 1945. Teaching and Learning English as a Foreign Language.
Ann Arbor: University of Michigan Press.
Johansson, Stig. 2008. Contrastive Analysis and Learner Language: A Corpus-
based Approach. Oslo: University of Oslo.
Lado, Robert. 1957. Linguistics across cultures: Applied linguistics for language
Teachers. Ann Arbor: University of Michigan Press.
Pym, Anthony. 2010. Exploring translation theories. New York: Routledge.