1. Introduction

In any account of interlingual communication, translation is used as a generic term. Professionally, however, the term translation is confined to the written, and the term interpretation to the spoken (Newmark, 1991: 35). If confined to a written language, translation is a cover term with three distinguishable meanings: 1) translating, the process (to translate; the activity rather than the tangible object), 2) a translation: the product of the process of translating (e.g. the translated text), and 3) translation: the abstract concept which encompasses both the process of translating and the product of that process Bell (1991: 13). The term 'translation' used and discussed throughout this paper is confined to the written language, and refers to both the product and process of translating.

The definitions of translation suggested above imply that producing the same meaning or message in the target language text as intended by the original author is the main objective of a translator. This notion of 'sameness' is often understood as an equivalence relation between the source and target texts. This equivalence relation is generally considered the most salient feature of a quality translation.

2. Problems of Equivalence

The principle that a translation should have an equivalence relation with the source language text is problematic. There are three main reasons why an exact equivalence or effect is difficult to achieve. Firstly, it is impossible for a text to have constant interpretations even for the same person on two occasions (Hervey, Higgins and Haywood (1995: 14). According to these translation scholars:

before one could objectively assess textual effects, one would need to have recourse to a fairly detailed and exact theory of psychological effect, a theory capable, among other things, of giving an account of the aesthetic sensations that are often paramount in response to a text (Hervey, Higgins and Haywood (1995: 14).

Secondly, translation is a matter of subjective interpretation of translators of the source language text. Thus, producing an objective effect on the target text readers, which is the same as that on the source text readers is an unrealistic expectation. Thirdly, it may not be possible for translators to determine how audiences responded to the source text when it was first produced (ibid, p. 14). Miao (2000) gives a specific example of the impossibility of the equivalence relation:

If an original was written centuries ago and the language of the original is difficult to comprehend for modern readers, then a simplified translation may well have greater impact on its readers that the original had on the readers in the source culture. No translator would hinder the reader's comprehension by using absolute expressions in order to achieve equivalent effect (Miao, 2000: 202)

Because the target text can never be equivalent to the source text at all levels, researchers have distinguished different types of equivalence (Lauscher, 2000: 151). Nida (1964) suggests formal and dynamic or functional equivalence. Formal equivalence focuses attention on the message itself, in both form and content. It requires that the message in the target language should match as closely as possible the different elements in the source language (p.159). Dynamic equivalence is based on the principle of equivalent effect, where the relationship between the receptor and message should be substantially the same as that which existed between the original receptors and the message (p. 159). Newmark (1981) makes a distinction between communicative and semantic translation. Like Nida's dynamic equivalence, communicative translation also tries to create the effect on the target text reader which is the same as that received by readers of the source language text. Koller (1997) proposes denotative, connotative, pragmatic, textual, formal and aesthetic equivalence. Munday (2001) describes these five different types of equivalence as follows:
1.Denotative equivalence is related to equivalence of the extralinguistic content of a text.
2.Connotative equivalence is related to the lexical choices, especially between near-synonyms.
3.Text-normative equivalence is related to text types, with texts behaving in different ways.
4.Pragmatic equivalence, or 'communicative equivalence', is oriented towards the receiver of the text or message.
5.Formal equivalence is related to the form and aesthetics of the text, includes word plays and the individual stylistic features of the source text (p. 47).

Baker (1992) classifies various problems of equivalence in translation and suggests some strategies to deal with them. Adopting a bottom-up approach, she begins with simple words and phrases and continues with grammatical, textual and pragmatic equivalences.

3. Strategies to solve problems of equivalence

As has been mentioned above, problems of equivalence occur at various levels, ranging from word to textual level. The equivalence problems emerge due to semantic, socio-cultural, and grammatical differences between the source language and the target language. These three areas of equivalence problems are intertwined with one another. The meaning(s) that a word refers to are culturally bound, and in most cases the meaning(s) of a word can only be understood through its context of use.

Due to semantic, socio-cultural, grammatical differences between the source language and the target language, loss and addition of information in translation cannot be avoided. Basnett-McGuire (1991) states that once the principle is accepted that sameness cannot exist between the two languages, it is possible to approach the question of loss and gain in the translation process (p.30). Bell (1991: 6) suggests a similar point that 'something' is always lost or, one might suggest, gained in the process, and according to Nida (1975), "all types of translation involve 1) loss of information, 2) addition of information, and /or 3) skewing of information" (p. 27). To conform to the stylistic demands and grammatical conventions of the target language, structural adjustment in translation is inevitably needed. These possibilities are expanded below.

3.1 Addition of information

Information which is not present in the source language text may be added to the target language text. According to Newmark (1988: 91), information added to the translation is normally cultural (accounting for the differences between SL and TL culture), technical (relating to the topic), or linguistic (explaining wayward use of words). The additional information may be put in the text (i.e. by putting it in brackets) or out of the text (i.e. by using a footnote or annotation). Such additional information is regarded as an extra explanation of culture-specific concepts (Baker, 1992) and is obligatory specification for comprehension purposes. Native speakers of Batak Tapanuli language (the native language of Batak community in North Sumatra), for example, have the word marhusip which literally means 'to whisper'. If the word marhusip is used in the context of discussing marriage within the community in question, its meaning is more than 'to whisper'. It refers specifically to a situation where family members of the bride meet family members of the groom to talk about the dowry. In the meeting, family members of the bride whisper with one another while deciding the amount of dowry they ask from the groom. Family members of the groom also do they same thing while deciding whether to accept or reject it. In this context, the word marhusip may be translated into 'to whisper', but additional information to clarify the meaning of marhusip is needed to help target readers understand its underlying concept.

Addition of information for specification purposes is also required "if ambiguity occurs in the receptor language formation and if the fact that greater specificity may be required so as to avoid misleading reference" (Nida, 1964: 227). It would be misleading, for example, if the word men in Tannen is an apologist for men is translated into para pria in Indonesian. The reason is that it does not actually refer to men in general but to American men in particular, who became the focus of Tannen's study on male-female interactions. It can be argued that translators should add the word Amerika to the Indonesian version to avoid ambiguity or to avoid a misleading interpretation of the outcomes of the study by Indonesian readers (Nababan, 2003).

Amplification from implicit to explicit status is another factor that requires additions. In relation to this, Nida (1964) states that 'important semantic elements carried implicitly in the source language may require explicit identification in the receptor language' (p. 227). In a given context, the meaning of the sentence, This rule is to round to the nearest even number, is implicitly stated and can easily be understood by readers of the original text (See Nababan, 1989 and 1999) If translated into Indonesian, an addition of information of suatu angka yang berada pada dua batas kategori (a number lying between two categories) and alteration of word class (the active verb membulatkan into the passive verb dibulatkan) are required to achieve grammaticality and produce an explicit meaning for Indonesian readers. It is by convention the Indonesian transitive verb membulatkan, as the equivalence of to round, needs an object. In such case, that sentence should be rendered into:

Target sentence:
Menurut aturan pembulatan ini, suatu angka yang berada pada batas dua kategori dibulatkan ke angka genap terdekat.

Back-translation:
According to the rule, a number lying between two categories is rounded to the nearest even number.

Addition of information may also be required due to the shift of voice and the alteration of word classes to avoid misinterpretation (Nida, 1964: 227). The word cut in I cut my finger is an active voice. If translated into Indonesian, the word class should be changed into a passive one, tersayat (was cut) and the addition of oleh pisau (with knife) is needed if a native speaker of Indonesian means that he or she did it by accident. There are also cases where two languages use a different class of words and a different level of utterances to denote the same meaning. The adjective adjustable in I have an adjustable chair is changed or translated into an adjective clause yang dapat disetel (which can be adjusted) in which the addition of yang (which) is obligatory to achieve grammaticality.

3.2 Deletion of information

Baker (1992: 40) refers to deletion as "omission of a lexical item due to grammatical or semantic patterns of the receptor language" (Baker, 1992: 40). She states further that

this strategy may sound rather drastic, but in fact it does no harm to omit translating a word or expression in some contexts. If the meaning conveyed by a particular item or expression is not vital enough to the development of the text to justify distracting the reader with lengthy explanations, translators can and often do simply omit translating the word or expression in question (Baker, 1992: 40).

There are cases where omission is required to avoid redundancy and awkwardness (Nida, 1964: 228) and this strategy is particularly applied if the source language tends be a redundant language. The category of plural in English is both morphologically conditioned (e.g. child/children, mouse/mice), and phonologically conditioned (e.g. book/books, box/boxes, pen/pens). In some circumstances, a plural noun is also preceded by a determiner showing plurality (some books, three pens). If the 'double' expression of such category is reflected in Indonesian, redundancy will occur. It is by convention that the category of plural in Indonesian is lexically formed by repetition of the noun buku-buku (book-book) or by adding a noun quantifier such as beberapa (some) or tiga (three). Once a given noun is in the plural form, the quantifier has to be deleted. On the other hand, once there exists a quantifier denoting plurality, the noun in question should be in the singular form or the repetition of the noun should be avoided.

As implicitly stated by Baker (1992: 40) above, deletion may also refer to pieces of content rather than restructuring for grammatical purposes. Such a deletion of expressions or information is debatable in relation to the translation of academic texts, however. Anyone who writes an academic text, for example, will not include unimportant information in his or her writing. Similarly, anyone who reads such a text should consider that all information in the text is important. Translators are not an exception; they should read the text as the original reader or a non-translator reader reads it. That is to say that this notion of information deletion should not be used as 'an excuse' to hide the inability of translators to understand and transfer message of the original text.

3.3. Structural adjustment

Structural adjustment is another important strategy for achieving equivalence. Structural adjustment which is also called shift (see Catford, 1965) or transposition (see Vinay and Darbellnet, 1977) or alteration (see Newmark, 1988) refers to a change in the grammar from SL to TL (Newmark, 1988: 85). Similarly, Bell (1991: 6) states that to shift from one language to another is, by definition, to alter the forms. The alteration of form may mean changes of categories, word classes, and word orders. Structural adjustment, according to Nida (1964: 226), has various purposes, including: 1) to permit adjustment of the form of the message to the requirements of structure of the receptor language, 2) to produce semantically equivalent structures, 3) to provide equivalent stylistic appropriateness, and 4) to carry an equivalent communication load.

Newmark (1988: 85-87) divides the shift of forms into four types. One type of shift is the change from singular to plural or in the position of adjective. The position of an adjective in English, for example, may occur before a noun (i.e. a difficult text) or before and after a noun (i.e. a difficult text available in the library). An adjective in Indonesian always comes before a noun. Therefore, a difficult text and a difficult text available in the library should be translated into sebuah teks sulit (a difficult text) and sebuah teks sulit yang tersedia di perpustakaan itu (a difficult text which is available in the library or a difficult text available in the library) respectively. A second type of shift is required when a SL grammatical structure does not exist in the TL. In English, for example, cohesive devices such as however and nevertheless may be put at the beginning or in the middle of a sentence. In Indonesian, such cohesive devices always occur at the beginning of a sentence. The third type of shift is the one where literal translation is grammatically possible but may not accord with natural usage in the TL. The English sentence The man to whom she is talking on the phone lives in Jakarta can be translated literally into Laki-laki kepada siapa dia sedang berbicara di telepon tinggal di Jakarta. This literal translation is accurate in content but doesn't sound Indonesian . To conform to natural usage in Indonesian, the structure of the sentence should be adjusted into Laki-laki yang sedang berbicara dengannya di telpon tinggal di Jakarta (The man who is talking to her on the phone lives in Jakarta). The fourth type of transposition is the replacement of a virtual lexical gap by a grammatical structure (see Newmark, 1988: 87).
In addition to the types of alteration described above, alterations of word classes (i.e. shifts from one class of words to another or from word level to phrase or clause level) are also required due to grammatical differences between the source and target languages. The preposition with in I am married with three young girls is changed into a conjunction dan (and), and the verb mempunyai needs to be added in Indonesian. The prepositional phrase in red in The woman in red is my wife is altered into an adjective clause yang berbaju merah (who wears the red clothes).


References

Baker, M. 1992. In Other Words: A Coursebook on Translation. London: Sage Publication.
Bell, R.T. 1991. Translation and Translating: Theory and Practice. London: Longman.
Bassnett-McGuire, S. 1991. Translation Studies. New York: Methuen & Co. Ltd.
Catford, J.C. 1965. A Linguistic Theory of Translation. London: Longman.
Hervey, S., Higgins, I., and Haywood, L. M. 1995. Thinking Spanish Translation: A Course in Translation Method: Spanish into English. London; New York: Routledge.
Koller, W. 1995. “The concept of equivalence and the object of translation studies”. Target, 7 (2), 191-222.
Miao, J. 2000. “The limitations of equivalent effect”. Perspectives: Studies in Translatology, Vol. 8. No. 3, 197-205.
Munday, J. 2001. Introducing Translation Studies. London; New York: Routledge.
Nababan, M.R. 2003. “Translation Processes, Practices and Products of Professional Indonesian Translators. Unpublished Ph.D. Thesis. Schools of Linguistics and Applied Language Studies, Victoria University of Wellington, New Zealand.
_________. 1999. Teori Menerjemah Bahasa Inggris. Yogyakarta: Pustaka Pelajar.
_________.1989. “Analisis terjemahan buku Research Methods and Analysis: Searching for Relationship karya Michael H. Walizer dan Paul, W. Wienir ke dalam bahasa Indonesia oleh Sadiman dan Hutagaol”. Unpublished Thesis. Surakarta: Universitas Sebelas Maret.
Newmark, P.1991. About Translation. Great Britain: Longdunn Press, Ltd.
__________. 1988. A Textbook of Translation. New York: Prentice-Hall International.
__________. 1981. Approaches to Translation. Oxford: Pergamon Press
Nida, E. 1975. Language Structure and Translation. Standford, California: Standford University Press.
______. 1964. Towards a Science of Translating. Leiden: Brill.
Vinay, J.P. and Darbelnet, J. 1965. Stylistique Comparee du Francois et de L’angalis. Paris: Didier.