THE TEXTUAL ROLE OF REPETITION IN THE TRANSLATION OF POLISH AND HUNGARIAN LEGAL TEXTS

Th is article presents research on written legal texts with a focus on the cohesion of such texts by analyzing the function of lexical repetition. Th e author indicates the possibility of using Hoey and Károly’s method of researching repetition patterns in texts in the process of translating Polish and Hungarian legal texts. In this analysis Polish, and Hungarian contract texts serve as a search base. Because contracts in both languages are structured into similar units, so called clauses, the author chose it for a base category of analysis. Th e author used three structures to search for lexical repetitions: intrasentential, intersentential and discourse structure. Because of the specifi c genre, contract clauses were used for analysis instead of popular linguistic units such as the sentence. Th erefore, the discussion here concerns intra-clausulal, inter-clausulal or legal discourse structures. Th e author states that the number and quality of repetition in Polish und Hungarian contracts is comparable. However, the number of lexical repetition appear to be smaller in Hungarian texts.

In recent years, the textual nature of translation has become a popular research topic.According to Neubert (1996: 87, 91) "the textual perspective has now been fully integrated into the agenda of translation studies." Neubert indicates two aspects of textuality which are connected with translation studies: 1. Linguistic utterances are always part of larger communicative events, and 2. Individual texts can be grouped into classes of texts that refl ect the communicative habits of speakers of a particular language.
Neubert rephrase these insights for two translation purposes: 1. Translation in the real world always has to do with whole texts 2. Translation is always connected to how other "similar" texts have been translated.
For a comprehensive analysis of text, it is necessary to examine the various levels of language.Some researchers emphasize the need to analyze the interaction of three levels of language: the intrasentential structure, the intersentential structure, and the discourse structure (compare: Connor andKaplan: 1987:2, Károly: 2002:16).According to Károly (2002), no theories thus far have been able to analyze all of these levels.One of the reasons for this could be the variety of text types and genres.In this analysis, a specifi c genre was analyzed and Polish and Hungarian legal texts were chosen as a search base.
In our study of legal text, we apply two issues as follows: 1. Th e meaning of written legal texts is consistently infl uenced by the whole legal system of the country from which the text originates in addition to other legal systems.Th is means that a translation of texts cannot be completed without taking into account the complete legal context.
Sometimes units of texts indicate the type of context which should be applied.For instance, Polish and Hungarian written contracts usually contain a unit which indicates the source of law regulation which should be applied if there are issues not mentioned in the units of contract.One such example is that of Polish or Hungarian statutory text.For instance: Unit of Polish contract: W sprawach nieuregulowanych postanowieniami niniejszej umowy mają zastosowanie przepisy kodeksu cywilnego.
Unit of Polish contract: A szerződésre a magyar jog szabályait kell alkalmazni.
2. In the study of legal text, it can be useful to compare parallel texts."Parallel texts are texts produced by users of diff erent languages under nearidentical communicative conditions.[...] Parallel text fi les [...] are part and parcel of the material and mental equipment of the competent translator.Th is equipment is a vast database storing enormous experience.It is the key to an extensive knowledge of how texts are structured in the (text) world of diff erent (communicative) cultures" (Neubert 1996:101).In translating legal written texts, the comparison of content and structure in parallel texts can be especially helpful when the legal systems of the two languages are similar.An example of this is the tradition of the continental legal systems in Europe and the Code Napoleon.Polish and Hungarian legal systems are both based on the continental tradition.Moreover, for fi ft y years the two legal systems were applied as systems of Soviet satellite states.Th at is why parallel texts of the two languages are oft en surprising with their similarity of content and structures in spite of diff erent (infl ected and agglutinative) language systems.
From our "legal text translation" perspective, one of the interesting issues discussed by Neubert (1996: 99) is the term units of translation.According to this perspective, units of translation are "the smallest source items of patterns susceptible of being rendered into the target text." Th is fl exibility means that they may consist of single words or whole texts.In our research, we applied the term unit to help with the process of analysis of Polish and Hungarian written contract texts.Comparing parallel texts, we stated that contracts in both languages are structured into the same kind of units, called clauses.According to the defi nition of Oxford Dictionary of Law, "a clause is a subdivision of a document.A clause of written contract contains a term of provision of the contract.Clauses are usually numbered consecutively" (Martin, Law: 2006).Units of translation can also reside in larger textual structures called macrostructures."Source texts very oft en show a particular macro structural order or distribution that is dependent upon certain conventions of a text type.Th ese conventions governing how textual chunks are coordinated and/or superordinated in the source culture may diff er signifi cantly from the usual way texts are structured by the target community" (Neubert 1996: 99).
Written contracts using a special macro structural order and certain conventions of a text type can be treated as a special genre of legal texts.Text researchers oft en use the two terms macrostructure and microstructure of a text.However, for our analysis of written contracts, we applied the third term used in the context of text structure: a mezostructure (van Dijk: 1980, Beaugrande-Dressler: 1981).Th is term is understood as the middle level of text which, in contracts, would be the part of the text called the clause.A clause is a unit which contains one or more sentences, but the sentences within the clause are more strongly connected to each other.
One of the most important issues in text research is coherence as well as cohesion.According to Halliday and Hasan (1976: 4) cohesion is "the relation of meaning that exists within the text, and that defi ne it as a text".Th e most frequent, and thus the most important category of cohesive ties, is the lexical category.Halliday and Hassan show that over forty percent of all cohesive ties in texts submitted to analysis were constituted by lexical ties.Grabe and Kaplan (1996: 55) defi ne cohesion as "the means available in the surface forms of the text to signal relationships that exist between sentences or clausal units in the text" (compare: Károly 2002: 64-65).Károly states, that they explain the importance of lexis from a more cognitive perspective.Th eir main argument is that "lexical entries used in text construction provide the basic meaning and inference signaling from which syntactic structures, semantic senses, and pragmatic interpretations are produced" (Károly 2002: 66-67).Our research of written legal texts is focused on cohesion of text by analyzing the function of lexical repetition.We compare the models of repetition in texts of two languages to obtain information about cohesion as well as obtain theoretical, methodological and practical issues for translation.Károly (2002) indicates, that most of the research on cohesion draws on Halliday's and Hasan's (1976) work, especially in connection with their cohesion analysis model of explicit cohesion markers to help infer meaning.Károly focuses especially on Hoey's (1991) work and modifi es a particular aspect of Hoey's original taxonomy to propose a refi ned version of the repetition model.She proposes an analytical tool for the study of lexical repetition in written argumentative discourse, which contributes to three main areas of study: cohesion analysis, repetition research, and English written text analysis.
One of the most important features of Hoey's system is his analysis of repetition cohesion which "not only itemizes cohesive features but also observes how they combine to organize text" (Károly: 2002:73).According to Hoey, repetition forms bonds which create nets of bonds, thus organizing text.In later studies Hoey (1995) also includes intertextual bonding in his system.Károly mentions a numerous aspects of Hoey's (1991) research.One aspect mentioned is that the greater part of cohesion is the product of lexical relations rather than grammatical relations.He perceives these lexical relations as various forms of lexical repetition.He emphasises the non-structural nature of his method, and reinterprets the description of text as "structure" and "culturally popular patterns of organization".In analysing texts, Hoey dissects each sentence of the given text and indentifi es items which are related.An amount of connection helps to distinguish between "central and marginal sentences" in the text.Central sentences make multiple connections with other sentences and thus play a crucial role in the development of theme in a text.Th e ones which show fewer connections and consequently contribute less to the development of the theme are called marginal sentences (compare Károly 2002:79).
Th us Károly, basing her research on Hoey's system, also indicates areas of weakness and proposes solutions.For instance, Hoey's criteria for the selection of lexical relations under the category of repetition remains unclear.Th e question remains of why certain lexical relations are excluded, such as antonyms which designate opposite rather than similar meaning.Diagrammatic representation does not indicate the quality of bonds, which can give misleading information about the distance between the bonded pairs.Károly improves these categories to cover more lexical relations and apply standard terminology (e.g.synonymy, hyponymy).It is also crucial that she establishes a hierarchy of lexical relations based on semantic closeness and demonstrates the place of each category within this hierarchy.Károly applied Hasan's (1984) category of instantial relations and grouped all instances of relations under this category which are created by textual context, i.e. are text-bound lexical relations.Károly (2002:113) states that "the potential of Hoey's (1991) analytical method is greater than what is shows in its present form.[...] In order for the revised analytical tool to be sensitive enough to predict subjective reader judgement of the organizational quality of texts, it is important to include measures related to the length, strength, and position of bonds." Th e framework of analysis applied by Károly is a partly revised version of Hoey's (1991) repetition model for the study of the text-organizing function of lexical repetition.Th e main categories of repetition are: same unit repetition (simple and derived repetition) and diff erent unit repetition, including synonymy (simple and derived), opposites, hyponymy, meronymy and instantial relations.Th e table below shows the summarized categories proposed by Károly (2002:104) and their distribution within the text: Hoey's principal aim of his research was to highlight sentences that are central to the development of theme and assist in the formation of readable summaries of text.Th e aim of the study by Károly was to propose and test a theory-based analytical tool, which would be capable of predicting subjective reader appeal in a reliable and valid manner.In relation to this our analysis is connected with the process of translation, especially with the translation of legal texts.Károly (2002:117) noticed that "as repetition is expected to play a similar role in other text types as well, the application of the revised taxonomy may be extendable to other genres too, except narration, on the same grounds as espressed by Hoey (1991) as well." According to the legal dictionary defi nition (Martin, Law: 2006), contract is "a legally binding agreement.Agreement arises as a result of off er and acceptance, but a number of other requirements must be satisfi ed for an agreement to be legally binding." Data analysis in Polish and Hungarian contracts on the basis of Károly's method reveals: a number of basic measures: the number of sentences and clauses in texts and the number of sentence interactions within texts repetition type combination of links and bonds to organize text, including strength of connection between the sentences linked by repetition (compare Károly 2002: 131).

Analyzed text of Polish contract
Th e repetition matrix of text shows us repetitions within the text.Th e number 0 represents the title of text.Also, connections within a title are accounted for.Th e method of counting repetition connections across the text are as follows: the 0 sentence (title) connects with the 0 clause, followed by the fi rst clause, second clause, third clause, etc. Th e table contains repeated units from two analysed clauses.Exceptions to that rule are: the sentence 0, in which the window contains repetitions beyond 0 sentence, and the windows 1-1, 2-2, 3-3 etc., which show connections within one clause.0 -repetition of diff erent personal (*) data within the same established pattern 0-4 content of clause (**) is very similar to the defi nition of contract in statute, so it is not treated as a repetition of the text.It is treated as a repetition within legal discourse.We do not deny that it is possible to fi nd more repetition in analyzed text with such as approach, but concentration in this article was placed on repetition within the text.
On the basis of the table above various conclusions were drawn:

Table of repetition
Th e table below contains a number of repetitions within the title, clauses, and text.According to this information, we fi nd three repetitions in the 0 sentence (title), three connections of 0 sentence within fi rst clause, four connections between fourth clause and second clause, two repetitions in sixth clause, etc.  2 3 4 5   3 3 1 3 1   4 4 3 4 7 6   5 2 1 2 2 0 0   6 4 1 2 4 4 0 2   7 1 1 3 2 1 0 2 0   8 3 1 2 2 2 2 3 1 1   0 1 2 3 4 5 6 7 8 Basic conclusions deducted from fi gure 1 and fi gure 2: • the number of clauses in texts (with a title): 9 • the number of sentences in texts: (with a title): 10 • the number of signs in texts: 1158 • the number of sentence interactions within texts (with a title): 101 Repetition type Based on table 1 and 2, we could also ascertain the type of repetitions used and the frequency of their appearance.

Same unit repetition
Th e table below contains a number of same unit repetition as well as the proportion of same unit repetition and diff erent unit repetition.Th e fi rst number in the table shows the number of same unit repetition.Th e second number gives the number of all repetitions in the analysed clause.For reasons of better clarity, we have not distinguished derivated and non-derivated forms.

Diff erent unit repetition
In our analysis, we took into account categories such as synonyms and opposites as well as other categories.When one category is in a repeated unit, we can fi nd similar words but diff erent grammatical forms, such as the noun kupno and the participle kupujący.Th e other category represents the repetition of the same established pattern and widely understood repetition within legal discourse.For reasons of better clarity, we have not counted opposites between seller and buyer or sell and buy.
Meaning of signs used in the fi gure 3: G -grammar change, Hn -hyponym, O -opposite, Hm -homonym (or partly homonym), S -synonym, P -the same pattern, D -repetition within legal discourse.
Th e number before a sign signals the frequency of appearance.For example, signing 2 G means two repetitions with a grammar change (for example: 0 (title) -sprzedaż -sprzedający, kupno -kupujący).0-2 1 Hr -Hn means that beyond 0-2 (title and second clause) connection there is one hiperonym -hiponym relation (pojazd -samochód).7-3 S, O -shows the repeated word stay in relation to synonymity and oppositity to its pair.To sum up, in the text we could fi nd 45 diff erent unit repetitions -28 relations of hyperonyms and hyponyms; -5 repetitions with grammar changes; -1 repetition with the same pattern; -1 repetition within legal discourse; -6 relations of synonym; -2 relations of synonym but opposite; -1 partly homonym; -1 homonym.

Diff erent unit repetition
To sum up, we can see, that the number and quality of repetition in Polish and Hungarian contracts is comparable.However, the number of lexical repetitions appears to be smaller in Hungarian texts.Th ough more research is required, we deduced that in making a text more coherent, the role of grammar is made more signifi cant in Hungarian texts than in Polish.Our second thesis, which calls for replication, is that in legal texts there are more of the same unit repetitions than in other kinds of texts.Both in Hungarian and Polish there is a manner, in order to make the style of text better that one should avoid same unit repetitions.However, that manner is not applied to legal written texts because of the importance of clarity.
Coming back to the Connor and Kaplan's (1987:2) approach, we can say that we used three structures to search for lexical repetitions: intrasentential, intersentential and discourse structure.Because of the specifi c genre, contract clauses where used for analysis despite other popular linguistic units such as the sentence.Intra-clausal relations are shown in the tables in windows such as 1-1, 2-2, 3-3, etc. Inter-clausal relations are shown in other windows, such as 1-2, 1-3, 1-4, etc.In our research, we did not focus on legal discourse structure.However, every clause of contract can be a kind of repetition and can be found in the content of a statute.
Because of the specifi c genre analyzed, it would not be useful to highlight the sentences that are central to the development of theme.We can see that almost every clause has at least one connection with another clause of the text, especially with the title of a text.Th e basic and most frequently used words of texts are: buy, sell and car.Each one of these can be found in the title, and in nearly each clause.Th erefore, every clause has the same importance.Th is statement agrees with the principle of law that the degree of importance of each word in legal text is the same.
It should be noted, that the results of our analysis may only be regarded as provisional, and further research is needed.However, we can say that research of repetition is useful not only for linguistics but also for translatology.It is imperative to better understand the meaning of translated text, taking into account both the contents inside the text and their connections to wider discourse.

Figure 3 :
Figure 3: Number of same unit repetition in text All amount of same unit repetition: 56

FigureFigure
Figure 5: Th e repetition matrix of Hungarian contract

Figure 7 :
Figure 7: Number of same unit repetition in text: All amount of same unit repetition: 45. 0 3-6