Copyright © 1995 Ali Darwish. All Rights Reserved.

Is Redundancy Translatable?

Ali Darwish


Introduction

In recent years, there has been a growing interest in explaining linguistic phenomena within the framework of communication and information theories with explorations into the dynamics of translation and how such dynamics affect the communication process.

In the early sixties, Nida called for employing all the resources of linguistics and communication theory to aid translators (Gentzler, 1993). The Czech scholar Jiri Levý (1969) and Alexander Ludskanov (1974), of the Bulgarian Academy of Sciences, used ideas from communication theory in their translation models. Earlier translation theorists such as I. A. Richards (1953) based their models on the Shannon-Weaver model of communication (Kelly, 1979: 38-39).

Notions such as entropy, redundancy, feedback, and noise have been explored in communication models with language as their focal point or point of departure. As Cherry (1978) and Turk and Kirmann (1989) maintain, redundancy is built into the structural forms of different languages in various ways. So is redundancy translatable?

This paper will examine how redundancy affects translation. It will rely on the contrastive analysis of English and Arabic to show the extent of translatability of redundancy and how redundancy affects translation decisions.


What is redundancy?

For a communication system, or any system for that matter to be effective, it must have a degree of superfluity. For example, a military communication system has several networks. If in war one network is destroyed, communication is maintained by resorting to other networks within the system. Moreover, the communication system in a modern office may consist of a multi-line telephone, a fax, internal and external mail services, online computer communication links such as the internet or electronic mail, to name the main ones. Such a system contains redundant components. For this system to remain effective at all times, these components are essential. Similarly, a human communication system, such as language, must contain redundancy in order to ensure the message is communicated without distortion or fail.

In communication theory, Fiske (1990) defines redundancy as: "... that which is predictable or conventional in a message... Redundancy is the result of high predictability..."(Fiske, 1990: 10) More recently, Fiske (1994) has defined redundancy as:

"...predictability structured into a message or text, in order to facilitate accurate decoding or to strengthen social bonds— or both." (in O’Sullivan et al, 1994:259-261)

Cherry (1978) on the other hand, views redundancy as "a property of languages, codes, and sign systems which arise from a superfluity of rules." This superfluity facilitates communication despite the factors of uncertainty that work against communication. Cherry maintains that human languages have developed a surplus of rules to maintain communication in the event of certain rules being broken. He argues that the rules of grammar and syntax are not inviolable. However, the more rules we break, the lower are our chances of successful communication. As such, language is designed in such a way that the various rules duplicate and supplement one another to ensure that communication is not hampered, interrupted or completely halted. "...We can break some of the rules, but we cannot break them all if we wish to remain within the social community..." (Cherry, 1978: 19)

Campbell (1982) cites the following examples from Professor William Lasher of such violation of rules and reduction of redundancy in the "Black English" of the United States that do not reach the point of breaking down communication.

We was at the ball game last night.
Mary had five card.

In The Grammatical Man, Campbell (1982) maintains that in order to reduce the unexpectedness or surprise effect of the information being sent, to make it more predictable, almost all forms of communication produce additional messages to convey the information intended by the sender. He argues that: "A message conveys no information unless some prior uncertainty exists in the mind of the receiver about what the message will contain. And the greater the uncertainty, the larger the amount of information conveyed when that uncertainty is resolved." (Campbell, 1982: 68)

In this sense, Campbell asserts the relationship between information, probability and predictability. He argues that a written message would be nonsense if it were completely unpredictable. For a message to be comprehensible, it must conform to rules of spelling, syntax and grammar. These rules are a form of redundancy which make certain parts of a message more probable than others and consequently more predictable. Campbell maintains that words might be easily deleted from a message (in English prose, for example) without compromising the integrity of the information it conveyed. Moreover, Campbell remarks that redundancy enables systems to become complex. Such complexity is important because failure is an inherent property of systems. If they are not complex, their failure is fatal or irrecoverable. That is why Campbell quotes von Neumann as saying that failure must not be thought of "as an aberration, but as an essential, independent part of the logic of complex systems." In this sense, redundancy, Campbell asserts, is what keeps a system running in spite of malfunction. It follows that all complex systems contain redundancy. A human body is a complex system that has redundant features — a pair of almost every vital organ: eyes, ears, nostrils, lungs, kidneys, and so on. A modern airplane also has a pair of almost every system — two hydraulic systems, two electric systems, two communication systems — as well as backups.

For Carroll (1964) redundancy is "the property of texts that allows us to predict missing symbols from the context. High-redundancy texts tend to be repetitive and to contain relatively little information per symbol. A text with zero redundancy would be one in which no symbol is predictable from any other, and thus there would be a maximum amount of informativeness per symbol." (1964:56) It follows that in some respect redundancy is contextual. If information is out of context, it will produce a message with low redundancy. This is clearly illustrated in Carroll’s example of a list of words selected at random from the dictionary. The list "would constitute a text with low redundancy." (56)

In their communication model, Turk and Kirkman (1989) highlight an important aspect of redundancy in communication which other models seem to overlook. They identify four points at which data is transferred from one medium to another. At each point, information, facts or ideas are lost. Information is potentially lost: first, when the message is transferred from facts to language; second, when it is transferred from language to written words; third, when it is transferred from written words to language in another mind; and fourth, when it is transferred out of that language into stored information. (13)

Four points at which a message is transferred
(Turk and Kirkman, 1989, 13)

Here, we readily see that redundancy is an important, if not an essential aspect of communication that compensates for any loss of data in the transition between stages.

In a translation system of communication, redundancy imposes constraints on the translation decision making process. These redundancy-induced constraints are explored further in this paper.

According to Littlejohn (1992), language is an example of a Markov process, in which things that come earlier in a chain create a probability that other things will come later. Redundancy in this respect obeys the rules of the Markov process. Finally, Palmer (1986, p:16), defines redundancy as [those] "...parts of the message that can be removed without removing any information."

This latter definition is important since we are trying to establish whether redundancy should be translated when a given text is translated into another language. Berlo (1960) adds another dimension to the definition of redundancy that is relevant to translation. He defines redundancy as:

"...the complement of entropy. As entropy rises, redundancy decreases. As redundancy rises, entropy or uncertainty decreases." (Berlo, 1960:202-203)

Berlo argues that redundancy in communication cannot be looked on as undesirable and that the question is not whether a message is redundant or not, but whether the redundancy is useful in increasing the effect of the message on the receiver.

Sommerhoff (1990) contends that a low redundancy is not always desirable in a message because the greater the redundancy the less will be the effect of any incidental error in the message on the information it conveys. (1990: 312)


Levels of redundancy

Cherry (1978) identifies two levels of redundancy, syntactic and semantic. "Syntactic redundancy implies additions to a text; something more is said or written than is strictly necessary to convey the message. Any individual has an enormous storehouse of his language statistics, habits and conventions, at both syntactic and semantic levels. He stores rules of spelling, word orders, grammar, idioms, clichés; again he stores typical vocabularies and phraseology which are used for specific subject matters, and he can predict to some extent from his knowledge of topics or of the writer’s point of view...We cannot say what elements may be stripped off a given text before the message fails to conveyed to a given recipient." Such inability makes the translation decision problem more difficult to resolve.

Consider the following example of stripping off redundant elements of text.

  1. Help him stand on his own two feet.
  2. Help him stand on his own feet.*
  3. Help him stand on his two feet.*
  4. Help him stand on his feet.*
  5. Help him stand (up).

The following is another example of semantic redundancy.

  1. Make sure the application form is signed.
  2. Make sure the application is signed.
  3. Make sure the form is signed.

Whether we like it or not, the fact remains that the majority of translators (those who effect change, not the elite or the trained) rely on the form and structure of the source text (i.e. the formal constraint). The more uncertainty there is about the information in a text, the closer the translator’s adherence to the formal component of text.


Semantic vs syntactic redundancy

It is interesting to note that in translation between certain languages such as English and Arabic, semantic redundancy is transformed into syntactic redundancy where the redundant semantic elements in one language become syntactic.

Help him stand on his own two feet.
sa’idhu lilwuqufi ‘ala qadamayhi.

The word qadamay(hi) [(his)two feet] is the dual form of qadam [foot].

Additionally, the syntactic limitations of the English sentence "Help him stand on his own two feet." impose semantic constraints on the translation. This is so because the imperative verb help is gender- and number-free in English while in Arabic it is gender- and number-sensitive. A translation into Arabic of the imperative verb form forces the translator to make a decision whether to render it as:

  1. Sa’idhu lilwuqufi ‘ala qadamayhi.
    Help him {second person singular, masculine}

  2. Sa’idihi lilwuqufi ‘ala qadamayhi.
    Help him {second person singular, feminine}

  3. Sa’idahu lilwuqufi ‘ala qadamayhi.
    Help him {second person dual, masculine/feminine}

  4. Sa’iduhu lilwuqufi ‘ala qadamayhi.
    Help him {second person plural, masculine}

  5. Sa’idnahu lilwuqufi ‘ala qadamayhi.
    Help him {second person plural, feminine}

These Arabic sentences have a built in syntactico-semantic redundancy, where the grammatical morphemes indicate gender and number. In translating any of these sentences into English, the syntactico-semantic redundancy is stripped off to a gender-and number-neutral form. Conversely, translating the English imperative into Arabic forces the translator to make decisions about gender and number. The English and Arabic imperatives impose constraints on translation and on the translation decision-making process.

To continue with the same language pair, Arabic also employs a form of semantic redundancy known as hendiadys. Hendiadys is the use of two or more words with an overlapping meaning to complete the overall meaning. In Arabic hendiadys is confined to two overlapping words that go hand in hand. Examples of hendiadys in Arabic are:

al-qada’u wa al-qadar
fate and destiny

al-jahu wa as-sultan
status and power

at-tarbiyah wa at-ta’lim
education and teaching

These word pairs complement each other. Strictly speaking, either word within the pair is redundant and can easily be omitted. But since the communicative value of their constituents is not always equal, it is not always immediately apparent which constituent is primary and which one is secondary , and which one can be dropped. In translating these pairs, however, the redundant element must be dropped, otherwise an awkward translation is produced. This imposes a constraint on the translator who has to decide the primary word.

Another aspect of redundancy in language has to do with tense, the form of the verb that denotes time. Many languages have two categories of time: past and not past. However, in expressing time-action relationship in tense, certain languages divide the temporality of action into various categories of tense. English speakers, for example, conceive of time as present, past and future. Within each tense there are subcategories denoting time-action relationship: simple present, present continuous, perfect present, continuous perfect present, simple past, continuous past, perfect past, continuous perfect past, simple future, continuous future, perfect future, and continuous perfect future. Certain languages do not have this distinction in tense. Arabic for example, has three tense categories: past, present and future although progressive forms are expressed in a variety of ways. The future is realized in present verb form or with the utilization of the auxiliary article sawfa ( or its shortened version sa).

Conceptual redundancy

Another feature of cross-cultural, cross-lingual redundancy is manifest in the system of concepts and how certain concepts are expressed in a concept-term relationship. Certain languages break down a concept into subconcepts. For example, English differentiates between engine and motor, while Arabic does not have such distinction (the word muharrik is usually used for both). In some cases, the translator can get away with the use of one term to refer to both or either concept. However, in certain translational (or transtextual) situations, this is not possible, and the distinction needs to be made. This places a constraint on the translation process. What this really means is that redundancy of concepts can at times be translated, but not always.

The following table compares the English and Arabic systems of concepts of family members (partial listing).

Concept Schema Arabic Arabic Supplemental English English Supplemental
Father abb father
Mother umm mother
Grandfather jadd jadduhu li-abih or
jadduhu abu abih

jadduh li-ummeh or jadduh abu ummeh
grandfather paternal grandfather
maternal grandfather
Grandmother jaddah jaddatuhu li-abih or jaddatuhu ummu abih
jaddatuhu li-ummeh or jaddatuhu ummu ummeh
grandmother paternal grandmother
maternal grandmother
Uncle amm uncle paternal uncle
khal uncle maternal uncle
Aunt ammah aunt paternal aunt
khalah aunt maternal aunt


The economy principle

Th Economy Principle propounds that if one can shorten the text while keeping the message unimpaired, this reduces the amount of time and effort involved both in encoding and in decoding. According to Leech (1983):

"...on the syntactic level, the Economy Principle has a contributory Maxim of Reduction which might be simply enunciated as ‘Reduce where possible’. But reduction should evidently not be recommended where it leads to ambiguity..." (Leech, 1983: 67)

The kind of redundancy which the Economy Principle seeks to eliminate is discussed by both Newmark and Duff. Newmark (1988: 208) discusses redundancy in passing and seems to be happy with Alan Duff’s treatment of the topic.

"The case in favour of eliminating the redundancies of the SL version in translation has been put with a wealth of examples by Duff (1971) which is a refreshing blast in contrast to the stale jargon of translation theory literature... Redundancies hang particularly loosely around clichés, phatic phrases (‘phaticism’) (‘naturally’, ‘of course’, ‘understandably’), repeated implied superlatives (‘basically’, ‘fundamentally’), prepositional phrases (au niveau de, dan le cadre de, ‘in view of the fact that, rhetorical flourishes (‘in the long march of history’), abstract terms (‘development’, ‘evolution’) and sonorous phrases for sound-effect (‘might and main’, ‘ways and means’). "Normally the translator has to use restraint in excising redundant SL features, confining himself to pruning here and there, since if he goes too far he is sometimes likely to find the whole text redundant."

Duff (1971) is concerned with the economy of writing and the time wasted by using superfluous words.

"This concern with redundancy might be considered quibbling. But if one considers how much time is wasted in writing, and then in translating, typesetting and printing these superfluous words, and again how much time is wasted in reading them, it can surely be argued that the translator is doing the public a service by cutting down on rhetorical rhythms.

"It is, of course, difficult for the translator not to be influenced by the number of words used in the source language text...’ (Duff, 1971: 24)

Both Newmark and Duff tackle redundancy as a problem of writing.


Cultural redundancy

Certain cultural systems are embedded in language. For example, in a stratified society such as Japan, the social hierarchy is embedded in the language itself. The language itself denotes the rank, status and social distance of speaker and listener. The usted in Spanish and vous in French are also examples of embedded systems of politeness that denote the kind of relationship between speaker and listener and the social distance and degree of familiarity between them. Such traits may be regarded as cultural redundancies within the context of translation. The translator must decide whether cultural redundancies are essential for the message to be communicated effectively in the target language.

Emotive redundancy

Another kind of redundancy is discussed by Eksell (1993). Redundancy of sentiments, as she calls it, is a cross-cultural idiosyncrasy. Arabic for instance, has often been described by English (and other western) speakers who have come in contact with Arabic through translation, as a "flowery" language. The reason for this misconception is to be found in the failure of many from-Arabic translators to deal with emotive redundancies. Through dogmatic adherence to literal translation and insistence on translating "everything" in the source language, these translators have transferred culture-specific emotive redundancies into other languages and are responsible for such misconceptions. For example, it is customary in Arabic correspondence to open the letter with ba'da at-tahiyyah (after greeting), which corresponds to (Dear...) in English. However, many translators still to this day translate the expression into (After greeting(s)..., or Greetings...) — the latter giving an Arabian Nights flavour.


Rhetorical redundancy

Languages employ various rhetorical techniques to express a message in a particular way and to achieve maximum effect. To achieve emphasis for instance, languages utilize techniques such as repetition, parataxis, positioning, isolation, the use of modifiers, rhythm and rhyme and so on. However, different languages use such techniques differently. For example, in a given context, repetition in one language might not have the same emphatic effect in another. What might be appropriate in one language might not be so in another. Certain languages employ grammatical devices to achieve emphasis, others use semantic devices.

Remember the first time we met each other?
Remember the first time we met?

The cross-language rhetorical conventions have been examined by Carloyn Boiarsky (1995). She argues that rhetorical conventions are informed by cultural customs and are manifested in a variety of ways and that many cultures follow rhetorical conventions that may mislead readers if they are not aware of the cultural traditions informing these conventions. (TCQ, 1995:246-252)


Redundancy and ambiguity

Certain sentence structures have semantic ambiguity inherent in them. This kind of ambiguity is caused by the absence of redundancy. For example, the sentence

He fed her dog biscuits.

is ambiguous because it can be construed as:

He fed her dog biscuits, that is he gave the dog she owns biscuits to eat.

or as

He fed her dog-biscuits, that is biscuits made for dogs to eat. (dog food)

As Harnish (1976) explains, "if we add information to this sentence that is already contained in it, we produce a redundancy... It should be possible to construct a sentence that is redundant on one sense but not on the other - given the knowledge we might have of those senses." (in Bever et al, 1976:330)...

Relativity of redundancy

We can safely say that redundancy is relative to the language in question. What in effect is redundant and superfluous in one language may be necessary and needed in another. The relativity of redundancy may be illustrated by the following example.

If you look after another person who needs special care and attention, you are a carer.

The word another in this sentence is redundant and can easily be omitted without the slightest distortion. However, the sentence might sound abrupt and stilted.

If you look after a person who needs special care and attention, you are a carer.

The intention in the first sentence is to make sure that "you look after another person beside yourself!". If the above sentence is rendered literally in Arabic, for example, the meaning will be distorted because it might be construed as if "you look after a person in addition to another!"

iza kunta ta’tani bishakhsin aakhar yahtaju ila al-’inayah wa al-ihtimam fa anta muqaddim ri’aya.

Here, removing the redundant word in the translation is not a matter of cosmetics or economic efficiency. It is rather obligatory and necessary for communicating the intended message.


Conclusion

This paper has explored the translatability of redundancy and the constraints it imposes on the translation decision-making process. The question whether redundancy should be translated or not must find an answer in the constraints that both the source and target languages impose on the translation decisions, which must include among other things defining the purpose of translation and the information needs of the translation user. Any decisions in this regard should consider whether redundancy increases understanding or acceptance of the translated message.


Top

References

Berger, C and Chaffee, S, editors (1987), Handbook of Communication Science, Sage Publisher, California.

Berlo, D (1960), The Process of Communication, Holt, Rinehart and Winston Inc, New York.

Bever, T C et al, editors (1976), an Integrated Theory of Linguistic Ability, the Harvester Press, Sussex, England.

Boiarsky, C (1995), The Relationship Between Cultural and Rhetorical Conventions: Engaging in International Communication, TCQ, Summar 1995, Vol 4, Number 3, 245-259, Minnesota, Association of Teachers of Technical Writing.

Budd, R W and Ruben, B D, editors (1972), Approaches to Human Communication, Hayden Book Company Inc, Rochelle Park, New Jersey.

Campbell, J (1982), Grammatical Man: Information, Entropy, Language and Life, Touchstone, Simon and Schuster, New York.

Carroll, J B ( 1964), Language and Thought, Prentice-Hall, New Jersey.

Cherry, C (1978), On Human Communication, MIT Press, Cambridge, Mass, third edition.

Duff, A (1981), The Third Language, Pergamon, London.

Fiske, J (1990), Introduction to Communication Studies, Routledge, second edition, London.

Gentzler, E (1993), Contemporary Translation Theories, Routledge, London.

Leech, G N (1983), The Principles of Pragmatics, Longman, London.

Littlejohn, S W (1992), Theories of Human Communication, Wadsworth, California, third edition.

McCroskey, J C (1979), An Introduction to Rhetorical Communication, Prentice-Hall, New Jersey.

Newmark, P (1982), Approaches to Translation, Pergamon, London.

Newmark, P (1988), A Textbook of Translation, Prentice-Hall, London.

Newmark, P (1993), About Translation, Multilingual Matters Ltd, London.

Nida, E (1975), A Framework for the Analysis and Evaluation of Theories of Translation, in Brislin, R. W. (ed) (1975), Translation Application and Research, Gardner Press, New York.

O’Sullivan, T, Hartley, J, Montgomery, M., and Fiske, J. (1994), Key concepts in Communication and Cultural Studies, second edition, Routledge, England.

Sommerhoff, G (1990), Life, Brain and Consciousness, New Perceptions through Targeted Systems Analysis, North-Hollands, Amsterdam.

Turk, C and Kirkmann, J (1985), Effective Writing, second edition, Longman, London.


Copyright © 1995 by Ali Darwish.

All rights reserved. No part of the contents of this document may be copied, reproduced, or stored in any retrieval system, without the express permission of the author.

Please direct all comments on this page to Ali Darwish.


Back to Home Page