Corps de l’article

1. Introduction

Scholars mostly consider the special effect created by literature-text on the reader as the major distinction between literary and non-literary texts; and they argue that such a special effect is the function of the special ‘patterning of language patterns’ (Hasan 1985). We have already attempted to characterize the distinction between literature and non-literature texts at both discoursal and textual levels arguing that literary discourse, apart from the ordinary discoursal strategies, would include three major special discoursal strategies: indirection, indeterminacy and defamiliarization and that these special discoursal strategies would be manifested in the surface text in the form of special textual strategies such as special sound, grammar and meaning patterns (cf. Lotfipour-Saedi 1992a).

Functional approaches to language would believe in a relationship between form and meaning. They maintain that any change in the way of saying is motivated by the factors in the context of situation (cf. Halliday 1985). The way of saying assumes more importance in literature. In fact, the degree to which way of saying becomes important is one of the basic distinguishing factors between literature and non-literature texts. In other words, the more the importance of way of saying in a text, the more literary the text; and, as we have argued before, in highly literary texts, the borderline between form and meaning disappears and the form or way of saying becomes the meaning (cf. Lotfipour-Saedi: 1992a). This special importance of the way of saying in literature-text manifests itself in the form of special literary patterns: i.e. special discoursal strategies (of indirection, indeterminacy and defamiliarization) and the special textual strategies manifesting them. It should be noted that there is no one-to-one relation between the discoursal and textual strategies. In other words, the discoursal strategy of ‘indeterminacy’ is not always presented in a certain textual strategy like ‘metaphor.’ This volatile and non-constant relation between textual strategies and their discoursal values in literary discourse can be attributed to the following two factors:

  1. The reader factor

  2. The special nature of the language patterns

It is due to such reader factors that in literary discourse not only may different readers negotiate different messages from the same text but also the same reader may experience different degrees of aesthetic effect in every reading of the same piece of literary text. Besides, as it was noted above, the discoursal effect of a literary text is not inherent to it but arises from the special language patterns it contains.

In the present paper, we shall examine two discoursal functions of sound patterns, as special textual strategies in poetic text drawing implications for the translatability or perhaps untranslatability of poetry.

2. Expectancy violation and sound patterns

Predictability in textual unfolding or expecting the textual norms/elements which may occur on the basis of what has gone before is an important characteristic of verbal interactions. Anticipating what will come in the text is based either on the receiver’s commonsensical knowledge of the world, of content and formal schemata (cf. Carrell 1983; Swales 1986), or on the writer’s commitment to the reader (mostly realized through certain prospective rhetorical devices in the text), “breaking of which will shake the credibility of the text” (Tadros 1994: 70). This feature would make an optimal level of redundancy available to the discourse receiver thereby reducing the amount of cognitive effort demanded on his part and enhancing the discourse processing operation.

The language user’s ability to predict some aspects of the forthcoming text can be said to derive from his overall communicative competence, which can be explained within the framework of his knowledge of the speech events and scripts involved (cf. Hymes 1972 and Schank and Abelson 1977). This is the case in normal and non-literary discourse. But the situation differs in literary discourse. Literary foregrounding mostly arises from the violation of the rules of everyday language use. Constant language norms would hardly be established in literature text. Besides, any norm, which may be established in violation of the rules, would be transient in nature. In such a volatile ‘universe of discourse,’ predictability is reduced to a minimum, a situation that demands higher attention and more intensive cognitive effort on the part of the receiver and makes the message less determinate in nature. Thus, one can argue that the degree of literariness of a text is conversely proportional to its degree of predictability. This aspect of literary text we name expectancy violation, by which we mean: at every stage of its textual unfurling, the literary text violates the expectancy of its readers. To exemplify the point, we look at a sonnet (in Persian) from the Persian poet Hafez. All lines in this sonnet end in the simile meaning ‘like the candle.’ The reader, having read the first few lines and on the basis of his familiarity with the poetic tradition in Hafez’s sonnets, would expect the same ‘explicit comparison’ or simile to occur at the final position of all the next-coming lines. But the poet has opted to violate this reader-expectancy by introducing a novel dimension of the candle schema for comparison at every simile-point; and in so doing he has attempted to activate more reader imagination, thus contributing to the literary effect of the text.

As another example (of the notion of expectancy violation), we look at the rhyming elements preceding the above-named simile in every line in the same sonnet from Hafez:

 Rhyming elements

Simile

1. khoob-^an-am

cho sham

2. rend-^an-am

cho sham

3. gery-^an-am

cho sham

4. suz-^an-am

cho sham

5. penh-^an-am

cho sham

6. ashkb^a-an-am

cho sham

7. suz-^an-am

cho sham

8. nogs-^an-am

cho sham

9. god^az-^an-am

cho sham

10. barafsh-^an-am

cho sham

11. eyv-^an-am

cho sham

12. bensh-^an-am

cho sham

These rhyming elements consist of two forms: ‘-^an’ and ‘-am.’ These two forms, despite their surface uniformity throughout the poem, perform different functions as follows:

-am:

  1. Predicative ‘be’: first person singular

  2. possessive morpheme: first person singular

  3. verbal affix: first person singular

-^an:

  1. plural morpheme

  2. adjectival affix

  3. part of the stem

The patterning of these elements in the twelve lines of the sonnet can be displayed as follows:

  • Line

  • 1

  • 2

  • 3

  • 4

  • 5

  • 6

  • 7

  • 8

  • 9

  • 10

  • 11

  • 12

  • -^an

  • A

  • A

  • B

  • B

  • C

  • B

  • B

  • C

  • B

  • C

  • C

  • C

  • -am

  • 1

  • 1

  • 1

  • 1

  • 2

  • 2

  • 3

  • 1

  • 1

  • 3

  • 2

  • 3

The above patterning can clearly demonstrate the amount of ‘expectancy violation’ or ‘norm-resetting’ introduced into the body of the poetic text. The effect of this strategy is further enhanced by the fact that the textual points accommodating these elements are graphologically and phonologically isomorphic, thus creating expectations as to their semantic and syntactic homogeneity, which are violated upon their deeper perception by the reader.

As another example of the notion of expectancy violation in poetic texts, we look at the initial phrases of the stanzas of the poem Follow Thy Fair Sun, Unhappy Shadow by Thomas Campion (1567-1620):

Stanza

Initial phrase

1

Follow thy fair sun

2

Follow her whose light

3

Follow those pure beams

4

Follow her

5

Follow still

Here, despite the fact that the presence of the same element ‘follow’ at the beginning of each stanza contributes to the textual cohesion of the poem and creates expectations as to the identity of the elements coming after that, changes in these elements would violate such expectations.

Such violations of the expectations cumulatively gained by the reader would certainly prolong his cognitive handling of the text. But this prolongation of perception in literary discourse, unlike the case of ordinary non-literary discourse where factors impeding the smooth processing of the text would be considered undesirable textual factors, is considered a desirable literary strategy which contributes to the literary value of the text (cf. Shklovsky 1965).

3. The textual function of sound patterns

What distinguishes a text from non-text (a set of unrelated sentences) is its texture, which consists of different components and textual strategies the discussion of which is beyond the scope of the present paper. The type and number of textual strategies included in a text have a direct effect on its degree of comprehensibility (Lotfipour-Saedi 1992b) and proportional with these textual strategies and depending on the links such strategies create among the lexical elements in a text, the discourse processes are activated for the computation and making of a meaning. In poetic text, many special textual strategies are employed in addition to the ordinary ones, which would strike unpredictable and extraordinary links among the lexical elements in the text. In the present section of the paper, we examine the textual function of sound patterns. One of the usual textual strategies is lexical reiteration (cf. Halliday and Hasan 1976), where the producer repeats a concept throughout a text thereby reinforcing its textual cohesion and affecting the relevant discoursal and cognitive processes. Besides, on the basis of the research carried out in cognitive psychology, the exact repetition of elements in a text would reduce their cognitive effect; but if the repetition of a concept is accompanied with ‘elaboration,’ i.e. innovative and abnormal strategies, the text will have deeper and better cognitive and discoursal effect (cf. Bransford 1979). We think the sound patterns in a poetic text would represent innovative and abnormal patterns for the reiteration of concepts and would thereby reinforce the texture of the text. For example, in the sonnet analyzed from Hafez in this paper, the sound similarities constituting its rhyming patterns would act as special lexical reiteration devices reiterating the concepts indicated by the lexical elements which act as the localities of such patterns. In other words, within the framework of a poetic text, upon the linear reception of the elements of a sound pattern, every element would reiterate the concept represented by the foregoing element in the active memory of the receiver. That is, upon the reception of the second element, the first element and the concept represented by it are reiterated; on the reception of the third element, the first and second elements, and on the reception of the fourth element, the first, second and third elements are repeated again. This process would continue until the end of the text. Thus, depending on the position of an element in the text, the number of times it is repeated throughout the text would vary: the number of times a concept is repeated and reiterated would be proportional to its physical distance from the end of the text.

The linear arrangement of a text is considered as a limitation for its discoursal processing (cf. Lotfipour-Saedi 1992b), a process which, due to its nature, would require all the textual elements to be accessible to the cognitive system of the receiver for it to be set in motion. But there are certain impediments for this, among which one can name the linear arrangement of the textual indices and the limitations of human short-term memory capacity which cannot hold more than a few chunks of information for more than a few seconds (cf. Bransford 1979). But certain textual strategies such as ‘reiteration’ would be employed to compensate for such limitations. Among the special strategies which we have discussed to be the characteristic of literature-text, we may name special ‘defamiliarized’ reiteration strategies, which, we believe, would manifest in some poetic texts as sound patterns. Such patterns would enable the reader of a poem to have access to the whole text and to process the whole discourse in its totality despite the linear reception of its components by the receiver. Such sound patterns, apart from acting as abnormal and special reiterating devices, would also create links across their lexical localities constituting an extra text in addition to the conventional one. Such a non-conventional and defamiliarized text would set its own special discourse process in motion in light of the links it crystallizes among concepts activated by the lexical localities involved. One can speculate that the sound similarities existing in the sound patterns in a poetic text, apart from activating the concepts of their lexical localities to the active memory of the reader, would also initiate a creative interaction among the mental images activated by such lexical items. Stipulated by the expectations arising from such sound similarities, the reader would endeavour to establish possible meaning relationships among the concepts involved by resorting to creative imagination procedures; and one may argue that the amount of such imaginative procedures would be directly proportional to the degree to which the concepts and images involved are diverse and heterogeneous in terms of the semantic fields they belong to (cf. Goudarzi 2000). As we have discussed before, the literary effect can be argued to be a function of such imaginative procedures activated by the special and defamiliarized textual strategies. Thus, the sound patterns in poetic texts perform essential cognitve, literary and discoursal functions.

4. Implications for translation of poetic text

Translation equivalence in literature-text should naturally be characterized in terms of the equivalence of the literary effect, which, as argued above, is the function of the special patterning of language patterns. Among such patterns, the sound patterns in poetry and their textual as well as expectancy-violation functions were discussed. The fact that every language is a unique system and that ways of expression in different languages are non-isomorphic would indicate that translating poetic texts is much more demanding than merely expressing what is said in the source language text in rhymed patterns in the target language. The sound patterns introduced into the body of the target text as the translation equivalents of a source poetic text should perform the same discoursal and textual function as those in the source language text in terms of the mode and quantity of reiteration, the degree of expectancy violation and the depth and type of imaginative procedures in which they engage the reader as a result of their lexical locality. This is a demand which, due to the unique character of every individual language and their non-isomorphic ways of saying the same thing, sound very difficult, if not impossible. Even if the same prepositional meaning can be reconstituted across languages in translating a poetic text, ‘relaying’ (cf. Hatim and Mason 1997) the discoursal function performed by the source language sound patterns into the target language would thus demand delicate strategic manoeuvrings on the part of the translator (cf. Lotfipour-Saedi 1996).