n this paper, an attempt will be made to examine the intricate nature of synonymy in an attempt to investigate its problematic nature in relation to translation. Emphasis will be made on whether or not translation is a form of synonymy. Types of synonymy will be analyzed and then examples from both English and Arabic will be provided to examine the overlap between one form of synonymy and another. Conclusions will be drawn at the end of the analysis and implications will be provided for further studies.
- What is Synonymy?
Although the notion of synonymy has been regarded in the past two decades as one of the most significant linguistic phenomena that influenced the structure of the lexicon, not much attention has been paid to this notion in the fields of lexicography, psychology or even computational linguistics (Edmonds and Hirst 2002). Whatever the reason, whether it be philosophical, practical or of expedience, synonymy was thought of as a non-problematic issue in linguistics or translation, because we have either synonyms with meanings that are completely identical and hence easy to deal with, or we have non-synonyms, in which case they can be treated as just different words (ibid: 106). The notion of near-synonyms, Edmonds and Hirst argue, shows that it is just as complex as the notion of polysemy, and that it inherently affects the structure of the lexical knowledge.
So what is this notion that is called synonymy? Synonymy is a kind of semantic relation among words. Technically, it occurs when two or more linguistic forms are used to substitute one another in any context in which their common meaning is not affected denotatively or connotatively. For example, words such as healthy and well, sick and ill, quickly and speedily, quickly and rapidly may be viewed as examples of synonymy, simply because they share most of the characteristics with one another.
Synonymy does not mean sameness, as this form of synonymy does not exist in monolingual or multilingual settings.
In an article entitled Translating Cultures: a Light-Hearted Look at the Pitfalls of Communication through Translation, Shaw (2003) states that human beings can differentiate between the nuances and/or fine distinctions of meanings between one object and another. Shaw exemplifies this by saying that, within our own language, a show can be a play, a drama, a musical, or a movie. The word show can even be a display of talent, i.e. a talent or a variety show. Shaw argues that, later on in life, we learn the real significant differences between angry, upset, bothered, ticked off, furious, and ballistic (as in "he went ballistic when they criticized his friends"). This same distinguishing process takes place as we learn a second language and, at the same time, learn that words have values and such values have unique and different semantic units.
Along the same line, Hjorland (2007) believes that synonymy is a kind of semantic relation. That is, words or phrases are synonymous only if they have the same meaning. However, there are cases where words or phrases may have subtle meanings and may therefore give rise to different word associations. For example, the Word Net database (2006) differentiates between different kinds of meanings for the word "computer" (cited in Edmonds and Hirst 2002:107). The first meaning is given as a "machine for performing calculations automatically". Here is a list of the different meanings the word computer entails:
Information processing system
From a non-contextual point of view, Merriam-Webster's New Dictionary of Synonyms (1984: 24) provides the following accurate definition of synonymy:
A synonym, in this dictionary, will always mean one of two or more words in the English language which have the same or very nearly the same essential meaning. Synonyms, therefore, are only such words as may be defined wholly, or almost wholly, in the same terms. Usually they are distinguished from one another by an added implication or connotation, or they may differ in their idiomatic use or in their application.
The above definition is a bit loose, as it does not distinguish between full or complete synonymy and near or partial synonymy. Synonymy has been defined as both full and partial synonymy, ignoring the subtle differences between one word and another. This inadequate definition, or rather the way some may regard synonymy, is a bit confusing to translators, particularly those who believe translation to be a form of synonymy.
- Types of Synonymy
Since many linguists believe that true or complete synonymy does not exist in any language (Quine 1951; Cruse 1986:270), attempts were made to classify synonymy into types. According to Quine (1951), there are two kinds of synonymy: complete synonymy and partial synonymy. Complete synonymy is regarded as words having identical meaning components. In more specific terms, words are complete synonyms if and only if they share all ingredients with one another. According to Quine, this kind of synonymy does not exist simply because it is impossible to define, and the meanings of words in monolingual or multilingual settings are constantly changing. Therefore, words may share most of the constituents with one another, but not all the constituents. As for partial synonymy, it is when words share most of the necessary components or constituents. For example, the words finish and terminate may share most of the characteristics with one another, but they are still different in some respects. The word finish suggests the final stage of doing something, whereas terminate suggests reaching a limit. It may suggest an end to a previous formal rendezvous. Since complete synonymy does not exist in monolingual settings, let alone across languages, partial synonymy has been emphasized. This dichotomy between complete and partial synonymy has added salt to injury in dealing with the notion of equivalence in translation or whether or not translation is a form of synonymy.
Based on the above discussion, I believe there is clear confusion as to what constitutes synonymy. That is, some treat synonymy as words sharing several characteristics with one another (Nida 1969: 73). Others suggest that this is regarded as a form of partial synonymy (Edmonds and Hirst 2002:107). I would like to suggest here that in order to be reasonable and clear, synonymy should be classified as follows:
Classifications of Synonymy
The above diagram shows that, for two words to be synonymous, they have to be identical and share all essential components and thus capable of being used to substitute one another in all contexts without any noticeable difference in their meanings. This kind of synonymy does not exist, without any doubt, between two text versions of the same language or source texts, let alone texts across languages.
- Translation and Synonymy
My point of departure here is to suggest that translation is not a form of synonymy, simply because words may have semantic values that are not translatable into other languages. For example, although words such as lie, falsehood, untruth, fib, and misrepresentation may be used to substitute one another in most contexts within the same language, they cannot be used to substitute one another in all contexts. According to Edmonds and Hirst (2002: 107), these are regarded as near or partial synonyms. The explanation is given by Edmonds and Hirst as follows:
Indeed, near-synonyms are pervasive in language; examples are easy to find. Lie, falsehood, untruth, fib, and misrepresentation, for instance, are near-synonyms of one another. All denote a statement that does not conform to the truth, but they differ from one another in fine aspects of their denotation. A lie is a deliberate attempt to deceive that is a flat contradiction of the truth, whereas a misrepresentation may be more indirect, as by misplacement of emphasis, an untruth might be told merely out of ignorance, and a fib is deliberate but relatively trivial, possibly told to save one's own or another's face (Gove 1984). The words also differ stylistically; fib is an informal, childish term, whereas falsehood is quite formal, and untruth can be used euphemistically to avoid some of the derogatory implications of some of the other terms.
From a different angle, the Arabic words hisaan, faras, jawaad, agarr, stand for the English word horse. Although these words can be used interchangeably in most contexts (since they all refer to the word horse), they are not interchangeable in all contexts. If we take the words for horse, we may find the following meanings that are synonymous and used in a context related to that word:
- The word hisaan has the components of horse and male.
- The word faras has the components of horse and male or female.
- The word jawaad has the components of a particular horse, which is fast, male or female.
- The word agarr has the components of a particular horse, which has a white patch on its forehead and male or female.
The plural form of any of these forms is khayl (horses), though (1) and (2) can have their distinct plurals as hisaan/ahsina and faras as furus/afraas, respectively. The above synonymous words have more than one semantic component in common. All of them have the component horse and male and female components. Only (1) has the component male alone, while (2) and (4) share the component male or female. We can also find that (1) and (2) have no distinctive qualities as horses, other than the components mentioned. However, (3) is characterized by agile movement and fastness and (4) by a special white patch on the forehead, which naturally contrasts with the overall dark color of the horse. How can the translator render these words in translation with their shared meanings into other language, without any loss or gain of meaning? This is an area where more research needs to be done.
In actuality, however, (1) and (2) can be used to substitute one another, without posing serious syntactic or semantic difficulties. I believe translators will have no difficulty transferring any of these two forms into English as horse since the words denote species and gender. Although (3) denotes a race horse, it can also be used to refer to horse in the general sense, with some loss of meaning in its associative meaning, i.e. fast horse. As for the word in (4), translators have to make it clear when transferring the meaning of this word into English as horse, that it denotes a horse of a particular color. If translators choose to be more faithful to the (SL) text, they can resort to paraphrase, in which case the word jawaad can be translated as a race horse, and agarr as a horse with a white patch on the forehead. If one agrees with Nida that, when dealing with synonymous words, we must look at the different componential features of the meanings of these synonyms and "select only those meanings which compete in the same semantic fields" (Nida 1969: 64), then we can be sure that the Arabic words for horse mentioned above are near synonyms. Such words show certain overlapping areas of meaning which 'compete in the same semantic field.'
Also, Arabic words such sayf, muhannad, husaam, among other words or expressions, stand for the English word sword. The word sayf is a neutral word, denoting the English word sword. Although the words muhannad and husaam share all the characteristics with the word sword, they connote additional characteristics. For example, the word muhannad refers to a sword in its sheath or scabbard, case, indicating that the sword has not been used yet. The word husaam refers to a sword that is pointed or sharp. It also suggests meanings of straightforwardness or uprightness. The neutral Arabic word sayf does not allude to such connotations. The question now is whether or not these words can be used to replace one another in all contexts without any loss or gain of meaning. In other words, are all these synonyms substitutable for one another in all contexts?
From a linguistic perspective, Nida (1969: 73) defines synonymy in language as "words which share several (but not all) essential components and thus can be used to substitute one another in some (but not all) contexts without any appreciable difference of meaning in these contexts, e.g. love and like. Peter Newmark (1981:101) takes a position similar to that of Nida declaring very clearly "I do not approve of the proposition that translation is a form of synonymy". Susan Bassnett-McGuire explains synonymy and the complexities associated with it in more detail. She points out that even apparent synonymy does not yield equivalence, "hence a dictionary of so-called synonyms may give the word perfect as a synonym for ideal or vehicle as a synonym for conveyance but in neither case can there be said to be complete equivalence, since each unit contains within itself a set of non-translatable associations and connotations" (Bassnett-McGuire 1980: 15). Furthermore, Bassnett-McGuire (1980:29) argues that "equivalence in translation should not be approached as a search for sameness, since sameness cannot even exist between two (TL) versions of the same text, let alone between the (SL) and the (TL) versions.
Anna Wierzbicka, on the other hand, examines the problem of synonymy and translatability by analyzing the deep structures of a language in terms of what she calls semantic primitives. Discussing the problems involved in translating the English color words and kinship terminology into other languages, she arrives at the conclusion that utterances in various languages differ, not only in their surface structures, but in their deep structures as well. Wierzbicka (1980: 67) maintains that "those different deep structures are always expressible in languages, which are mutually isomorphic; they are all isomorphic with respect to the universal lingua, that is to the language of semantic primitives. For this reason, deep structures of sentences in different languages (different as they may be in themselves) are always mutually translatable".
As pointed out earlier, synonymy does not mean sameness, as this form of synonymy does not exist in monolingual or multilingual settings. Synonymy can be described in terms of exact replacement and interchangeability. That is, words can be described as synonymous if and only if they replace each other in all contexts without any change in either the cognitive or emotive import.
Moreover, equivalence may be regarded as an appropriate criterion that proves to be an adequate form of translation. What is meant by equivalence here is the fact that every linguistic unit (below the level of the sentence) has a characteristic distribution. If two (or more) units occur in the same range of contexts, they are to be distributionally equivalent (or have the same distribution). It is extremely important however to make sure that these two equivalent lexical items are synonymous if and only if there is no change in the meaning within the whole text.
Bassnett-McGuire, S. (1980). Translation Studies. NY: Methuen & Co. Ltd.
Cruse, D. A. (1986). Lexical Semantivs. Cambridge: Cambridge University Press.
Edmonds, O. P. and Hirst, G. (2002). "Near-Synonymy and Lexical Choice". Computational Linguistics, Vol.28, Number 2: 105-144.
Gove, Philip B., Editor. (1984). Webster's New Dictionary of Synonyms. Merriam-Webster, Springfield, MA.
Lyons, J. (1995). Linguistic Semantics. An Introduction. Cambridge: Cambridge University Press.
Nida, E. and Taber, C. (1969). The Theory and Practice of Translation. Netherlands: E.J. Brill.
Newmark, P. (1981). Approaches to Translation. Oxford: pergamon.
Quine, W.V.O. (1951). "Two Dogmas of Empiricism". Philosophical Review, 60: 20-43.
Wierzbicka, Anna (1980). Lingua Mentalis: the Semantics of Natural Language. Sydney/New York: Academic Press.