Implicitation in Legal Translation – A Study of Spanish-Danish Translation of Judgments

Whilst the concept of explicitation has spawned many research projects in the fi eld of Translation Studies, implicitation remains an understudied area. This article addresses that research gap, reporting on the fi ndings of an empirical study on Danish translators’ use of implicitations in their Spanish to Danish translations of an excerpt from a judgment. The aim of the study was to examine, on the one hand, whether the translators used implicitations in their translations, and, on the other hand, whether differences could be observed between experts and non-experts. The data reported here consisted of a Spanish source text and 10 translations into Danish by fi ve experts and fi ve non-experts. The translations were analysed using qualitative methods (consisting of contrastive text analysis) followed by a quantitative synthesis. Overall, the fi ndings revealed that implicitations were very rare and, consequently, no conclusive results could be drawn in relation to translation expertise.


Introduction
In Translation Studies, implicitation refers to the tendency to make implicit in the target text explicitly stated information of the source text (for a defi nition of implicitation as used in this article, see section 3).For example, implicitation occurs when a source text word with a specifi c meaning is replaced with a more general word in the target text (Klaudy/Károly 2005: 15).Thus, implicitation constitutes the opposite of explicitation, that is, the tendency to spell things out that are implicit in the source text by, for instance, replacing a source text word with a general meaning with a more specifi c word in the target text (Klaudy/Károly 2005: 15).Explicitation has been suggeste d to be one of the universals of translation, that is, one of the linguistic features which are typical of translations in comparison with non-translated texts (e.g. by Baker 1996).While research into explicitation is a productive area in Translation studies (see e.g.Blum Kulka 1986, Séguinot 1988, Øverås 1998, Olohan/Baker 2000, Pápai 2004, Englund Dimitrova 2005, Klaudy/Károly 2005, and Becher 2011a, among many others), implicitation has been largely neglected by translation scholars, as we shall see in the literature review below.The fi eld of legal translation is no exception.In such a fi eld, only few empirical studies on implicitation have been conducted so far.Notable exceptions are the studies by Faber/Hjort-Pedersen (2009a, 2009b, 2013) and Hjort-Pedersen/ Faber (2010), in which the authors examined trainee and professional translators' use of implicitation and explicitation in English to Danish translations of an excerpt from a law report.
Thus far, no studies have been conducted on implicitation in the translation of legal texts with respect to the language pair Danish and Spanish.The study reported on in this article seeks to contribute to fi lling this gap in the research.It examines, on the one hand, whether professional Danish translators use implicitations in their translations of a judgment from Spanish into Danish, and, on the other hand, whether differences can be observed between experts and non-experts (for a defi nition, see section 3).
The present paper contributes to the literature on implicitation by providing empirical evidence of professional translators' use of implicitation.Apart from its academic contribution, it may be useful in raising professional and student translators' awareness of their translation practices and decision-making.
The article is organised in the following way.Section 2 explores the phenomenon of implicitation by looking at the literature and previous studies on implicitation.Section 3 presents central defi nitions and concepts, while Section 4 outlines a description of the study in terms of aim, methods and data.Section 5 presents the results of the qualitative analysis and the subsequent quantifi cation, and Section 6 discusses the results and implications of the study and suggests directions for future research.

Literature on implicitation and main results from previous studies
This study draws on theory and research from the fi elds of Translation Studies and legal translation.More specifi cally, it centres on the asymmetry hypothesis, implicitation in legal translation with a focus on the translation of judgments, and implicitation in relation to expertise.In the subsequent sections, each of these topics will be considered.

The asymmetry hypothesis
Vinay/Darbelnet were the fi rst to introduce the term implicitation in 1958 to describe "a stylistic translation technique which consists of making what is explicit in the source language implicit in the target language, relying on the context or the situation for conveying the meaning" (Vinay/ Darbelnet 1958/1995: 342).In the following years, only little attention was given to implicitation by translation scholars compared to the vast amount of literature produced on explicitation, especially after Blum-Kulka's (1986) introduction of her famous explicitation hypothesis suggesting that explicitation may be inherent in translation irrespective of the languages involved.In fact, it was not until Klaudy (2001) introduced her asymmetry hypothesis that translation scholars started taking an interest in the phenomenon of implicitation.According to Klaudy's asymmetry hypothesis, explicitations in one translation direction are often not counterbalanced by implicitations in the other translation direction "because translators -if they have a choice -prefer to use operations involving explicitation, and often fail to perform optional implicitation" (Klaudy/Károly 2005: 14).In consequence, the relationship between explicitation and implicitation is often asymmetric in nature, hence the name of the hypothesis.Klaudy (2001Klaudy ( , 2009) ) operates with four different types of explicitation and implicitation: obligatory, optional, pragmatic, and translation-inherent. Obligatory explicitations and implicitations are caused by structural differences between the source and target languages, and they are obligatory in the sense that they have to be performed to produce a grammatical translation.Optional explicitations and implicitations stem from stylistic differences between the two languages, whereas pragmatic explicitations and implicitations result from cultural differences between the source and target languages.Unlike the fi rst three explicitation and implicitation types, translation-inherent explicitations and implicitations are not motivated by differences between the source and target language, but rather "attributed to the nature of the translation process itself" (Klaudy 2009: 107).Klaudy/Károly (2005) set out to provide empirical evidence for the asymmetry hypothesis, examining professional translators' use of explicitations and implicitations in their translations of reporting verbs in literary texts translated from English to Hungarian and Hungarian to English.The results of the study showed that explicitations were frequent in translations from English into Hungarian, whereas implicitations were relatively rare in translations from Hungarian into English, thus providing evidence in support of the asymmetry hypothesis.Becher (2011a) has criticised Klaudy's asymmetry hypothesis on two main grounds.First, the hypothesis does not specify which types of explicitations and implicitations it covers (if obligatory, optional, pragmatic or translation-inherent), and, second, the hypothesis is not properly motivated.Instead, Becher (2011a: 59) has suggested a slightly modifi ed version of the asymmetry hypothesis which reads as follows: Obligatory, optional and pragmatic explicitations in one translation direction tend to be more frequent than (i.e.not `counterbalanced´ by) the corresponding implicitations in the other translation direction, regardless of the source/target language constellation at hand.
Questioning the very existence of translation-inherent explicitations and implicitations, Becher's version of the hypothesis includes Klaudy's obligatory, optional and pragmatic explicitations and implicitations only.As regards motivation for the hypothesis, Becher (2011a: 60-62) argues that the asymmetric relationship between explicitation and implicitation can largely be explained from two prototypical characteristics of the communicative situation of translation.First, in translation, cultural distance exists between the author of the source text and the recipient of the target text (see House 1997).According to Becher, translators tend to use explicitations in their translations in an attempt to bridge this cultural gap.Becher further suggests that the desire to compensate for cultural differences is, in fact, so strong that it sometimes drives translators to explicitate even when these differences are not relevant.Second, translation is per se associated with a certain amount of communicative risk, that is, the risk of not being understood (Pym 2005).The most important task of the translator is to facilitate understanding between the author of the source text and the recipient of the target text and, consequently, it is imperative to avoid misunderstandings at all costs.Against this backdrop, implicitations are often considered a risky bet because the recipient must be able to infer the implicit information in order to understand the target text.Conversely, explicitations are not associated with the same communicative risk and, as a result, translators tend to explicitate rather than to implicitate because they wish to minimise the risk of misunderstanding (Becher 2011a: 60-62).Becher (2011b) sets out to test his version of the asymmetry hypothesis, examining explicitations and implicitations of connectives in a bidirectional parallel corpus of translations of business texts between English and German.Overall, the fi ndings showed that additions of connectives (i.e.explicitation) in translations from English to German were not counterbalanced by omissions of connectives (i.e.implicitation) in translations from German to English, thus providing evidence in support of the asymmetry hypothesis.However, the results of the study also revealed that with, for example, additions and omissions of pronominal expressions, a symmetric relationship existed between explicitations and implicitations, contradicting the hypothesis.
Denturck's study from 2012 offers empirical evidence contradicting the asymmetry hypothesis in one translation direction.Denturck examined causal conjunctions in French to Dutch and Dutch to French translations of novels.Her fi ndings showed that, while there was no counterbalancing of explicitations in translations from French to Dutch, a considerable number of explicit causal conjunctions were counterbalanced by implicitations in translations from Dutch to French.

Implicitation in legal translation
It has often been suggested that legal translation be accorded a special status because of the sacred (or sensitive) nature of legal texts.What differentiates legal texts from other texts is that they potentially produce legal effects and that such potential legal effect has to be the same in dissimilar legal systems, not only languages.Consequently, approaches to legal translation have historically focussed on literal translation (for a historical overview of legal translation, see Šarčević 2000: 23-53).Because the source text is usually perceived as particularly important in determining the success of a legal translation, the principle of literal translation remained unchallenged in legal translation until the late 20 th century.In fact, it was not until the 1980s, when legal translators were assigned an active role in the drafting of federal legislation in bilingual legal systems (e.g.Switzerland and Canada), that literal translation loosened its grip on legal translation.With the introduction of new bilingual drafting techniques, the entire fi eld of legal translation was revolutionised (Šarčević 2000: 16).
In recent years, several legal translation scholars have argued that legally binding translations (i.e.authoritative translations), such as contracts, prenuptial agreements and wills, must be literal, while translations intended for information (i.e.non-authoritative translations), for instance national laws and court orders, may be freer (Didier 1990: 280, 285, Garzone 2000: 400, Borja Albi 2007: 32).These scholars argue in favour of a freer approach in informative translation because literal translations -using combinations of words and structures similar to those of the source language -may require an additional processing effort on the part of the recipient.In some cases, a literal translation may even be incomprehensible to the target language recipient and thus fail to fulfi l its purpose as a source of information.The key issue in this context is that if the purpose of the translation is information, it is vital that the translation be comprehensible to the recipient.
In monolingual legal systems, the judgment is a non-authoritative translation and, as such, it is translated strictly for information purposes (judgments in bilingual legal systems and from supranational -e.g.EU -or international institutions are authoritative).Whilst the judgment is legally binding in the source language culture, it functions as a metatext in the target language culture, which enables the recipient (i.e. the judge) to understand the foreign judgment.The purpose of a translated judgment is usually to serve as an informational basis for the subsequent issuing of a judgment in the target language culture.As such, the translation helps the judge understand the legal reasoning of the court of the source language culture, and allows him or her to issue a judgment in the target language culture based on the translation.For example, a Spanish company signs a contract with a Danish company, a contract that the latter violates.Unless the contract prescribes otherwise, the Spanish company is most likely to bring the matter before a Spanish court, which subsequently issues a judgment in favour of the Spanish company.However, unless the Danish company has assets in Spain upon which the Spanish company may distrain, this judgment is of little benefi t to them.Therefore, the Spanish company must attain an enforced judgment (i.e. a judgment that may be enforced through the bailiff) from a Danish court.To issue an enforced judgment, the Danish judge must understand the legal reasoning of the Spanish court, which, in turn, requires that the Spanish judgment be translated into Danish.Only after the Danish court has issued an enforced judgment, the Spanish company may distrain upon the assets of the Danish company (Engberg 1999: 84).Other possible recipients of a translated judgment are business executives.For instance, if a foreign business partner is a party in a lawsuit, a business executive may want to understand the judgment to determine its potential consequences for their future relations (Engberg 1999: 84).
Thus, in translations of judgments, the main task of the translator is to facilitate understanding between the author of the source text and the recipient of the target text, rather than to convey the exact content and form of the source text.This is where implicitations may be relevant, and, consequently, it is a reasonable assumption that implicitations will occur in translations of judgments.
Faber/Hjort-Pedersen (2009b) examined implicitation (among other things) in English to Danish translations produced by professional translators and lawyers.Overall, the results of the study showed that the lawyers, who supposedly communicate with experts like themselves, used implicitations more than the translators did.In addition, the fi ndings indicated differences between the two groups in the type of implicitation in which they engaged.While the lawyers mainly used reductions (e.g.leaving out postmodifying phrases of the source text), the translators mainly opted for generalisation (e.g.choosing semantically less specifi c legal terms).
Faber/Hjort-Pedersen (2013) examined how expectancy norms (i.e. the expectations of the translation commissioners) and professional norms (i.e. the expectations of the translator) infl uenced the use of explicitation and implicitation in translation.The participants consisted of six professional Danish translators (experts in English) and fi ve Danish legal experts (experts in law with a suffi cient knowledge in legal English).The participants were asked to evaluate three dif-ferent English translations of the same source text in Danish (one literal, one containing explicitations, and one containing implicitations) according to whether the translations conformed to the norms.The fi ndings revealed that both groups agreed that if the purpose of the translation was unknown, a literal translation was preferable.In addition, translators and legal experts agreed that explicitations were acceptable if the translation was for information purposes.When it comes to implicitations, the two groups differed in opinion.The translators argued that implicitations were acceptable because the target text recipient (i.e.lawyer) was an expert and, as such, was capable of making the necessary inferences.The legal experts, on the other hand, discouraged the use of implicitations, arguing that the target text recipient was provided with too little information.One plausible explanation, at least in part, for the stance taken by the legal experts may be that lawyers are not linguists and, hence, not experts in English.

Implicitation and translation expertise
Some empirical studies have pointed to a link between implicitation and translation expertise.Puurtinen (2003) examined explicitation and implicitation in trainee translators' English to Finnish translations of magazine articles with the purpose of determining whether adding or omitting information changed the opinions expressed by the authors of the source texts.Examples of explicitation included substituting nominalisations with verb phrases and adding connectives to the target text, and implicitation, for instance, involved substituting relative clauses with noun phrases and verb phrases with nominalisations.The fi ndings revealed that implicitations outnumbered explicitations by far and that the use of implicitations often changed the viewpoints expressed by the source-text authors.According to Puurtinen, the implicitations performed largely resulted from insuffi cient language and/or translation skills.
Faber/Hjort-Pedersen (2009a) and Hjort-Pedersen/Faber (2010) propose two contradictory hypotheses regarding implicitation and explicitation in legal translation.On the one hand, because legal texts may produce legal consequences, performing implicitations and explicitations in legal translations is associated with considerable risk, even if they may be necessary in the case of, for instance, system-bound terms.In consequence, the authors hypothesise that implicitations and explicitations are rare in legal translation, particularly in translations produced by trainee translators due to their limited legal expertise.On the other hand, and because legal texts are notoriously complex, they require a considerable processing effort on the part of the translator.Therefore, the authors hypothesise that in the process of trying to understand the source text, the translator leaves traces in the target text in the form of implicitations and explicitations.The assumption is that these traces are particularly evident in translations produced by trainees because they struggle harder with understanding the source text than professional translators do.
Faber/Hjort-Pedersen (2009a) examined the translations produced by eight trainee translators and two professional translators while both groups were thinking aloud.In the analysis, focus was on nominalisations, passives, culture-bound terms (referred to as system-bound terms by Šarčević 2000 and in this study), and elliptical phrases.The fi ndings revealed that both trainee and professional translators used implicitations in the form of reductions, whereas none of the participants used generalisations (for a defi nition of reduction and generalisation, see section 3).However, because the study forms part of a larger project, the authors only offer tentative observations.Hjort-Pedersen/Faber (2010) further explored implicitations in trainee translations, using the same data as in the 2009 study.The results showed that the implicitations of the study generally resulted from the translators' attempts to mask a lack of understanding of the source text.In half of the cases, reductions were preceded by mental implicitations.In the rest of the cases, the participants opted for reductions even though the think-aloud protocols revealed no evidence of preceding mental implicitation.In other words, these implicitations were based on an automatic process.With respect to type of implicitation, the results of the study showed that the translators opted for reduction only, and there were no examples of generalisations.According to the authors, this suggests that the strategy of generalisation does not generally form part of the participants' translation repertoire.

Central concepts and defi nitions
For our purposes, implicitness is defi ned according to Becher (2010: 3) as the non-verbalisation of information that the recipient may be able to infer from the context, his or her world knowledge, etc.Consequently, implicitation occurs when information explicitly stated in the source text is made implicit in the target text.The defi nition of implicitation used in this article only refers to the product of translation, not to the translation process.
In addition, implicitations can take two forms in this study, that is, reduction or generalisation (see Øverås 1998Øverås , Perego 2003 and Klaudy/Károly 2005, among others).While reduction involves leaving out source-text elements in the target text -and is, hence, quantitative in nature -, generalisation involves using elements in the target text, which are semantically less specifi c than those of the source text, and is thus qualitative in nature.
In the study of implicitation, reference is often made to pragmatic terms such as presupposition -which refers to the implicit assumptions relating to an utterance whose proposition must be true for the utterance to make sense (coined by Strawson 1952) -and implicature, that is, implicit assumptions in an utterance which are context-dependent, deniable and not necessarily logical (Grice 1975).However, the study is not interested in examining the implicit assumptions underlying the utterances of the source and target texts, but rather in examining the reductions and generalisations performed by the participants of the study, and, therefore, such pragmatic terms will not be further discussed in this article.
Moreover, for the purposes of this study, the term implicitation refers to reductions and generalisations that belong to the categories of optional implicitation (i.e.resulting from differences in stylistic preferences) and pragmatic implicitation (i.e.motivated by cultural differences) (Klaudy/ Károly 2005: 17).The two categories are interrelated in the sense that because optional implicitation relies on the pragmatic norms of the target language community, it is `pragmatic´ in nature.Conversely, pragmatic implicitation is `optional´ in nature in the sense that it usually does not have to be performed (Becher 2010: 23, on explicitation).In other words, this study focuses on implicitations relating to a specifi c language pair (i.e.Spanish and Danish), which are optional in the sense that a translator may opt for a translation which is closer to the structure and words of the source text (i.e. a literal translation), or a translation which is more explicit than the corresponding source text.In other words, the present study sets out to explore whether translators, when they have a choice, opt for implicitations.
Finally, expertise -or expert performance -is often equated with professionalism in the fi eld of Translation Studies.However, in this study, a distinction is made between professionalism and expertise (Enríquez Raído 2014).Whilst professionalism concerns the translators' ability to earn their living by translating, expertise involves translation processes "that are observed to result in good performance" (Tirkkonen-Condit 2005: 406).This distinction is based on the following premises.First, translation is a skill, and, consequently, it can be learned through formal instruction and practice (Chesterman 2000, Montalt Resurrecció et al. 2008).Secondly, there is wide individual variation between translators and not all professionals reach expert level (Jääskeläinen 1990, Séguinot 2000).In addition, no translator is an expert in all areas (Shreve 2002).During the course of their careers, translators specialise their translation abilities as well as their knowledge of particular subject areas and domain-specifi c terminologies.
For the purposes of this study, the concept of translation expertise is operationalised as the number of years' experience as a professional translator.More specifi cally, translation expertise is, in this study, defi ned as 10 years' experience or more (Shreve 2002, Englund Dimitrova 2005).However, in determining translation expertise, we must take into account domain-specifi c expertise, as mentioned above.Therefore, experts are, in this study, defi ned as translators with 10 years' experience or more who specialise in the translation of legal texts.Conversely, non-experts are defi ned as translators with less than 10 years' experience and/or translators who do not specialise in legal translation.
In establishing expertise, Bernardini (2001) cautions against relying only on external criteria, such as translation experience, as measured by the number of years in the profession.Bernardini argues that we must take into account individual differences between translators as well, as noted above.However, this study is not concerned with defi ning expertise as such, or with translation quality, but with examining the implicitations used in legal translations produced by experts, on the one hand, and non-experts, on the other hand.

Aim, methods, and data
As indicated in the introduction, the aim of this study is to examine, fi rst, whether professional Danish translators use implicitations in their Spanish to Danish translations of an excerpt from a judgment, and, secondly, whether differences can be observed between experts and non-experts.
To examine professional translators' use of implicitations, an experiment involving translation from Spanish into Danish was performed.The data presented here stem from a PhD thesis by Krogsgaard Vesterager (2011).The purpose of that thesis was different from the one pursued in this article in that the former examined whether professional translators opted for a literal or free translation.The data consisted of a Spanish source text and 10 translations into Danish, fi ve of which were produced by experts, and the remaining fi ve by non-experts.The source text, which consisted of 221 words, was an excerpt from the grounds of a Spanish judgment (the source text is appended in section 7).The topic of the source text was dismissal on conduct grounds, that is, dismissal due to misconduct or non-compliance on the part of the employee.The source text was a judgment from an appeal court, referring to the judgment of the fi rst instance court on several occasions in the text.
The translations were analysed by means of qualitative methods consisting of contrastive text analysis, and subsequently a quantitative synthesis was made.Looking at a single case (one source text and 10 translations), the main aim was to examine the implicitations performed by the participants in their target texts in depth, rather than to make generalisations about implicitations in legal translation.However, in studying the particular case, a certain degree of generalisation is inevitable, as has been pointed out by for example Flyvbjerg (2006) and Roald/Køppe (2008).
The analyses focussed on implicitations in relation to nominalisations, passives, system-bound terms, and elliptical phrases because they are characteristic of legal language (see Faber/Hjort-Pedersen 2009a, Hjort-Pedersen/Faber 2010).In the following, these items are referred to as focal points.The source text contained 11 nominalisations, two passives, 15 system-bound terms, and three elliptical phrases.
The set-up and methodology of the study are the same as in Krogsgaard Vesterager's (2017) forthcoming study on explicitations in Spanish to Danish translations of judgments.
Examining translation practices of professional translators while working in their natural setting, this study is naturalistic in nature, although with two reservations.First, the translation task was constructed for research purposes.Second, in two cases, I was referred directly to the translator by the translation agencies.Consequently, two of the translators (no.8 and 10) were aware that they were participating in a research project, but they were not informed of the purpose of the study.The participants were provided with background information in the form of the source text in its full length, and a deadline of two weeks was set for them to solve the translation task.The translators were not given further translation instructions in the form of a brief, but it was possible to obtain one on request.However, none of the participants requested additional information in relation to the translation task.
As already indicated, the participants were divided into two groups on the basis of their practical experience in legal translation, as follows: 1) experts defi ned as translators with 10 years' experience or more who specialise in the translation of legal texts, and 2) non-experts defi ned as translators with less than 10 years' experience and/or translators who do not specialise in legal translation.More specifi cally, translators 1, 3, 4, 5, and 8 belonged to the expert group, whereas translators 2, 6, 7, 9, and 10 belonged to the non-expert group, as detailed in table 1.

Analyses and results
This section presents the results obtained from the analyses of the translations.More specifi cally, section 5.1.presents the results of the qualitative analyses, and section 5.2.outlines the results of the quantitative synthesis.

Analyses of focal points
To determine how the focal points (i.e.nominalisations, passives, system-bound terms, and elliptical phrases) were transferred from the source text to the target text, the 10 target texts were analysed using the framework of contrastive text analysis.According to the analyses, none of the participants opted for implicitations in relation to the elliptical phrases, and, consequently, this focal point will not be discussed in further detail here.In the following, the results of the analyses of the three remaining focal points are presented.

Nominalisations
The analyses show that implicitations of nominalisations are rare.The nominalisations that are implicitated are all in the form of reduction, and they relate to source text redundancy, as exemplifi ed in 1 1 .(… in accordance with this gradualist principle that seeks to establish an appropriate correspondence between conduct and sanction… ) TT2: […] i henhold til naevnte gradualistiske kriterium, der har en tendens til at etablere en passende balance mellem adfaerd og sanktioner, […] (… in accordance with the aforementioned gradualist criterion that has a tendency to establish an appropriate balance between conducts and sanctions …) In example 1, the source text is redundant in the sense that the nominalisations proporción ('proportion') and correspondencia ('correspondence'), both of which are modifi ed by the adjective adecuada ('appropriate'), are near-synonymous -also referred to as a binomial pair (Bhatia/Candlin 2004).Both nominalisations of the source text express the proposition that a sanction must be proportional to the violation committed by the employee.In this context, both translators (who are experts) include only one of the nominalisations in their target texts, as can be seen from my emphasis.By opting for a reduction strategy, both participants eliminate the redundancy of the source text, as can be seen from my translations.Thus, the translators have adapted their target texts to comply with the stylistic preferences of Danish, which is generally less prone to linguistic redundancy than Spanish (Krogsgaard Vesterager 2011), as is typical of Germanic languages compared to Romance languages (Korzen 2005).

Passives
According to the analyses, implicitations of passives are even rarer than implicitations of nominalisations.In fact, only one passive has been implicitated in the ten translations.The implicitation performed is in the form of reduction, as exemplifi ed in 2.
Example 2 (Translation 10, my emphasis) ST: […] que no se justifi ca la sanción de despido apoyada en actos realizados en el clima de tolerancia […] (… that the sanction of dismissal is not justifi ed for actions carried out in a climate of tolerance …) TT: […] at man ikke kan berettige sanktionen om afskedigelse baseret på handlinger i et klima af tolerance […] (… that one cannot justify the sanction of dismissal based on actions in a climate of tolerance …) In example 2, the translator (a non-expert) has opted for a reduction strategy, omitting the verb in the passive voice realizados ('carried out') from the translation.Because of this reduction, the target text is ungrammatical, as can be seen from the translation of the TT into English.Consequently, this translation is associated with considerable communicative risk, that is, the risk that the target-language recipient does not understand the target text, as mentioned in section 2.1.Thus, example 2 reminds us of the risk associated with performing implicitations.

System-bound terms
According to the analyses, implicitations of system-bound terms are also rare.Most of the implicitations that do occur are in the form of reduction, and they relate to source text redundancy, as exemplifi ed in 3.In example 3, the source-text redundancy stems from the noun phrase sanción de despido ('sanction of dismissal').More specifi cally, the head of the noun phrase, sanción ('sanction'), is superfl uous in the sense that a dismissal is per defi nition a sanction, and, consequently, further elaboration is redundant.As can be seen from my emphasis, both translators (one being an expert and the other one a non-expert) have eliminated the redundancy of the source text by opting for a reduction strategy.Thus, both participants have translated sanción de despido ('sanction of dismissal') with afskedigelse ('dismissal') in their target texts.Another instance of implicitation involves generalisation, as exemplifi ed in 4. (… it must be compared with the legal requirement as defi ned in the Spanish Labour Law Act …) In example 4, the translator (an expert) has opted for the translation juridisk krav ('legal requirement'), which is semantically less specifi c than the Spanish term exigencia gradualista ('gradualist requirement'), as may be seen from my translations.In consequence, one may argue that by opting for a more general translation, vital source-text information is left out.However, since this particular excerpt refers to a specifi c provision of the law, the target text recipient will arrive at the same understanding of the target text as does the source-text recipient when consulting the relevant legal provision.As we can see in Table 2, the source text includes 31 occurrences within the categories of the four focal points, making the total possible number of implicitations in the 10 target texts 310.Table 2 also shows us that the total number of implicitations actually performed in the target texts is seven.This means that in only 2 percent (7 of 310) of the cases, the participants opt for implicitation in relation to the focal points.We can also see that only 3 percent (3 of 110) of the nominalisations are implicitated, only 5 percent (1 of 20) of the passives are implicitated, only 2 percent (3 of 150) of the system-bound terms are implicitated, and none of the elliptical phrases are implicitated.Of the implicitations performed, 43 percent (3 of 7) are implicitations of nominalisations, whereas 14 percent (1 of 7) are implicitations of passives, and 43 percent (3 of 7) are implicitations of system-bound terms.If we look at types of implicitation, the results show that reduction constitutes 86 percent (6 of 7) of all instances of implicitations, and generalisation accounts for 14 percent (1 of 7) of the instances.

Quantifi cation
To examine whether there are differences in implicitation patterns between experts and non-experts, we need to take a closer look at the distribution of the results according to the participants' expertise.Table 3 below details the implicitations performed in the 10 target texts distributed on experts, on the one hand, and non-experts, on the other.Table 3. Distribution of implicitations according to expertise According to Table 3, experts undertake 57 percent (4 of 7) of all implicitations, whereas non-experts account for 43 percent (3 of 7) of implicitations in the ten target texts.Thus, experts implicitate slightly more than non-experts do.In addition, the fi ndings reveal slight differences in implicitation patterns between the two groups.While experts implicitate nominalisations and system-bound terms, non-experts also implicitate passives.More specifi cally, experts account for 67 percent (2 of 3) of all implicitations of nominalisations and for 67 percent (2 of 3) of implicitations of system-bound terms.Conversely, non-experts account for 33 percent (1 of 3) of all implicitations of nominalisations, for 100 percent (1 of 1) of the implicitations of passives and for 33 percent (1 of 3) of implicitations of system-bound terms.Although the results of the study indicate slight differences between experts and non-experts, no conclusive results can be drawn from the study in relation to translation expertise because of the very limited number of implicitations performed.

Conclusion and discussion
This article has been concerned with implicitations in Spanish to Danish translations of judgments.The purpose of the study was to examine, on the one hand, whether Danish translators used implicitations in their translation of a Spanish judgment, and, on the other hand, whether differences could be observed between experts and non-experts.To fulfi l that purpose, an experiment involving translation of a judgment from Spanish into Danish was performed.The data consisted of a Spanish source text and 10 translations into Danish produced by fi ve experts and fi ve non-experts.The translations were analysed using qualitative methods followed by quantifi cation.
The analyses focussed on implicitations with respect to nominalisations, passives, system-bound terms, and elliptical phrases.
The fi ndings revealed that implicitations in relation to the focal points were very rare.More specifi cally, the quantitative analyses showed that only 2 percent of the focal points were implicitated in the 10 target texts.Thus, the results of the study are largely consistent with the fi ndings of previous studies (Faber/Hjort-Pedersen 2009a, 2009b, Hjort-Pedersen/Faber, 2010).Drawing on the two prototypical characteristics of the communicative situation of translation, as proposed by Becher (2010), a tentative explanation for these fi ndings may be that translators associate implicitations with unnecessary risk-taking and, consequently, they refrain from using implicitations in their translations.
In addition, the results showed that of the implicitations performed, implicitations of nominalisations represented 43 percent, whereas implicitations of passives and system-bound terms constituted 14 percent and 43 percent, respectively.According to the analyses, none of the participants opted for implicitation in relation to elliptical phrases.As for types of implicitation undertaken by the ten participants, the results showed that reduction accounted for 86 percent of all instances of implicitations, whereas generalisation constituted 14 percent of the implicitations.This is consistent, to some extent, with the fi ndings of previous studies on implicitation in legal translation (Faber/Hjort-Pedersen 2009a, Hjort-Pedersen/Faber, 2010) which suggests that generalisation may not form part of the translators' repertoire.
Moreover, in relation to translation expertise, the fi ndings showed that experts implicitated slightly more than non-experts did.According to the qualitative analysis, the implicitations performed in the target texts mainly served the purpose of eliminating source text redundancy.These reduction strategies were carried out successfully with one exception that resulted in an ungrammatical sentence, reminding us of the risk associated with performing implicitations.Thus, to some extent, the fi ndings contradict those of previous studies, which have suggested that implicitations result from insuffi cient language and/or translation skills (Puurtinen 2003, Hjort-Pedersen/ Faber 2010).Finally, the fi ndings revealed slight differences between experts and non-experts in the items they implicitated.While experts implicitated nominalisations and system-bound terms, non-experts also implicitated passives.However, because of the very limited number of implicitations performed, no conclusive results could be drawn with respect to translation expertise.
This study contributes to the literature on implicitation and thus helps fi ll the research gap by providing empirical evidence of translators' implicitations in legal translation.The limitations of the study point to directions for future research.First, because the study involves a relatively small sample size, the results of the study are, of course, only indicative, and further research involving a larger sample size is needed to confi rm or disclaim them.Second, since only a very limited number of implicitations are performed in the translations, no conclusive results can be drawn as regards translation expertise.Thus, further studies are needed to examine whether there are differences in implicitation patterns between experts and non-experts.Third, the study solely centres on four focal points of the source text.Therefore, an empirical study examining all instances of implicitation would be a valuable follow-up to this study.The next step will be to carry out another experiment, involving a longer source text (based on data by Krogsgaard Vesterager 2011) to supplement the results of this study.In the experiment, I want to examine all instances of implicitations performed by experts and non-experts in order to provide a more elaborate analysis of implicitations, as performed by professional Danish translators, and, in addition, it may lead to a modifi cation and/or a refi nement of the results of this study.

Appendix
The source text used for the translation task and my very close translation of it into English.
De suerte que, con arreglo a dicho criterio gradualista, tendente a establecer una adecuada proporción y correspondencia entre conductas y sanciones, y con criterio individualizador conforme a las peculiaridades del caso concreto, resulta justifi cado el despido para las conductas que supongan una violación trascendente de la buena fe contractual, con lo que no cualquier transgresión de ella, sino solamente la de carácter grave y culpable, es la que tiene entidad bastante para que resulte lícita aquella sanción, habiendo establecido la jurisprudencia incluso que si es cierto que la tolerancia del empresario no genera un derecho al incumplimiento del trabajador, también lo es que no se justifi ca la sanción de despido apoyada en actos realizados en el clima de tolerancia y dentro del margen de la misma (Sª T.S de 24-9-1990 , entre otras).[…] My translation: […] Thus, while a breach of good contractual faith constitutes a non-compliance, this is subject to the gradualistic requirement defi ned in Article 54, section 1 of the Labour Law Act whereby it is an empirical question to identify the extenuating circumstances of the violation, both as regards the subjective element of culpability (i.e. as more or less excusable ignorance, according to Article 433 of the Civil Code, or as more or less surmountable erroneous belief, according to Article 1950 of the aforementioned Code), and the objective element, the seriousness of the violation.
Thus, according to the gradualist criterion aimed at establishing an appropriate proportion and correspondence between conducts and sanctions, and with an individualizing criterion according to the peculiarities of the specifi c case, dismissal is justifi ed for actions that constitute a gross violation of the good contractual faith, and thus not every violation thereof but only one which is serious and culpable justifi es such a sanction, with case law even having established that although it is true that the employer's tolerance does not give the employee the right to non-compliance, it is also true that dismissal is not a viable sanction for actions performed in a tolerant work environment (Supreme Court judgment of 24-9-1990, among others) […]

Table 1 .
Background information on participants(Krogsgaard Vesterager 2017, forthcoming) : […]i henhold til dette gradualistiske princip, hvorved det tilstraebes at etablere passende overensstemmelse mellem adfaerd og sanktion,[…] 1 All examples include the source and target texts with my translations into English and my emphasis.Abbreviations are as follows: ST = source text and TT = target text.If the examples include excerpts from more than one target text, the abbreviations TT1 and ST1 are used.TT1

Table 2 .
Table2below shows how the 10 participants' implicitations are distributed over the focal points and the types of implicitations described in section 5.1.Representation of implicitations by number and type