International Journal of Speech Language and the Law, Vol 22, No 1 (2015)


doi : 10.1558/ijsll.v22i1.24360

The law of England and Wales: translation in transition

Catrin Fflur Huws


This article considers the role of the legal translator/interpreter in bilingual court proceedings. This is not, however, a situation where a non-English speaker gives oral evidence which is translated for the benefit of the court, but rather a situation where a court – the court in question being the Supreme Court of England and Wales – is unable to comprehend the bilingual text of the legislation. The interpreter is therefore called upon to explain the possible differences in meaning between the two versions. This article questions the extent to which the legal system is aware of the significance of these issues, and advocates re-evaluating the training requirements of legal interpreters.


government of wales act 2006, statutory interpretation, expert witness, devolution, bilingual law

1. Introduction

Scene: Cardiff, the near future.

The defendant, Rhiannon Williams, is appearing before a court arguing that she is not in breach of any obligation under the law. The law is a statute passed by the National Assembly for Wales, and has been drafted in Welsh and English. Ms Williams has used the Welsh version of the Act. Under this version of the Act she is able to argue that she has committed no breach. However, the claimant, Richard Harker-Smythe, argues that under the English version of the Act, Ms Williams is liable.

Is this an issue of interpretation? In bilingual jurisdictions, such as Canada, it is not. The court will look at both versions of the text and decide which version is correct, and, more specifically, how to interpret the Act consistently across the two language versions. However, the devolution settlement to Wales introduces a crucial new dimension to the role of the interpreter – the implications of which have not been fully realised. England and Wales is not a bilingual jurisdiction. The higher courts operate solely through the medium of English. The reliance on the legal interpreter is therefore far more significant because the interpreter must explain to the court how the two versions differ and why this difference is important. Accordingly, this article will consider the role of the legal interpreter in situations where the court must adjudicate with reference to both texts.

2. Context: the Welsh devolution settlement

When power was devolved to Scotland (Scotland Act 1998 (1998 c.46)), Wales (Government of Wales Act 1998 (1998 c.37)) and Northern Ireland (Northern Ireland Act 1998 (1998 c.47)) in 1998, it was generally considered (Osmond 2003) that devolution to Wales was the weakest of the three settlements because the National Assembly for Wales was only permitted to legislate on matters that had been specifically devolved to it, whereas Scotland was permitted to legislate on all matters that were not specifically reserved. Even since devolution was expanded to encompass the power to make primary legislation, when the Government of Wales Act 2006 (2006 c.32) came into force, Welsh law may only be made in relation to the 20 devolved matters listed in schedule 7 of the Act (since Part IV of the Act came into force in 2011), and even these contain numerous exceptions.

However, a significant aspect of the devolution settlement in Wales, which is not shared by the other devolved territories, is that legislation enacted by the National Assembly for Wales must be made bilingually, with equal weight being given to the two language versions of the text (Government of Wales Act 2006 s115). What this means is that when the law of Wales is interpreted by the courts, regard must be given to both language versions.

This poses no problem in other bilingual legal systems because, in addition to a bilingual legislature, there is also a bilingual jurisdiction. Accordingly, the courts are able to look at the two versions of a statute and adjudicate as to its proper meaning, explaining and addressing any inconsistencies as required. In the Canadian case of R v Daoust [2004] 1 SCR 217 for example, the Supreme Court of Canada was able to look at both versions of the law, and conclude that, although the English version was the more correct version, the French text was the most appropriate version to use because it encapsulated the only meaning that was common to both versions. This case concerned a situation where the defendants were likely to be liable under the English version of s462.31 of the Canadian Criminal Code, but liability would be avoided under the French version. The Supreme Court was able to examine both versions of the code and conclude (at paragraphs 35 and 36 of the judgment) that:

35 when one of the two versions of a provision of a bilingual statute has a broader meaning than the other, the common meaning of the two versions is normally the one that is derived from the version with a more restricted meaning. This rule is especially relevant in a criminal context, as the accused may, depending on which version he or she reads, form a different conception of the elements of the offence in question.

36 Here, the only possible common meaning is the most restrictive one. This common meaning is easily derived, as the two versions list similar prohibited acts, with the exception of the added phrase which appears in the English version only. Since the English version cannot represent the common meaning, the French version must. Thus, we are restricted to the elements of the actus reus enumerated in the French version.

However, this would not be the case in relation to the interpretation of Welsh law, because, although legislative power has been devolved, the jurisdiction continues to be that of England and Wales. This has two important consequences. Firstly, in the lower courts (particularly in civil cases where the law of Wales is more likely to be applied because the criminal law has not been devolved) the defendant may elect to have his or her case heard in their local court. Thus a court in Worcester, Worksop or Winchester may be called upon to apply the law of Wales to determine the litigation it is asked to adjudicate upon.

A more significant issue, however, is that the appellate courts – the Court of Appeal and the Supreme Court – will have to interpret the law of Wales. Therefore, in these cases, the interpretation of a bilingual statute becomes an important issue. It is envisaged that all interpretation of the law of Wales will need to occur with reference to both language versions (Government of Wales Act 2006 s156), although it is possible that the courts may interpret this requirement as meaning that the court should be able to rely only on its preferred version. This would, however, be contrary to the principle espoused in the Government of Wales Act 2006 that the Welsh and English versions are to have equal weight, and would also undermine the principle regarding the official status of Welsh contained in s.1 of the Welsh Language (Wales) Measure 2011 (2011 NAWM 1). As Beaupré explains:

so long as both statutes are equally authoritative statements of the law and the statute as a whole is to apply equally across the country, you act at your own peril and that of your client whenever you sever one of those versions from the other in your attempt to ascribe meaning to the words of Parliament. (Beaupré 1988: 328)

Where the courts ‘sever one … version from the other’, the peril is of course to justice and the due process of law. This is manifested particularly in the situation described above, where it is alleged that the two versions of the text are inconsistent, as then the question of bilingual interpretation cannot be avoided. Furthermore, where one – or all – the parties to the litigation have relied solely on the Welsh version, the validity of the judgment is likely to be questioned if the court relies only on the English version when adjudicating the matter. Accordingly, in a number of cases, the courts will not be able to avoid referring to the Welsh version of the law as well as the English version.

This is confirmed when we look at Hong Kong’s experience of developing a bilingual legal system. Fung explains that, following the secession of British rule in Hong Kong, the law began to be administered both in English and in Chinese, and parties to litigation began to realise that alleging that the two texts were different could provide them with an argument that would support their case. Accordingly, Fung (1997) anticipated that there would be a proliferation of cases coming before the courts, where it was alleged that the two versions had different meanings. Bilingual lawyers in the UK will be keen, therefore, to scrutinise both the English and Welsh versions of the text either in order to confirm or to restrict their client’s liability, as was encountered in the Daoust case. Here, although the English version of s462.31 provided that:

462.31 (1) Everyone commits an offence who uses, transfers the possession of, sends or delivers to any person or place, transports, transmits, alters, disposes of or otherwise deals with, in any manner and by any means, any property or any proceeds of any property with intent to conceal or convert that property or those proceeds, knowing or believing that all or a part of that property or of those proceeds was obtained or derived directly or indirectly as a result of

(a) the commission in Canada of an enterprise crime offence or a designated substance offence; or

(b) an act or omission anywhere that, if it had occurred in Canada, would have constituted an enterprise crime offence or a designated substance offence. [Emphasis added.](Criminal Code, R.S.C. 1985, c. C-46).

The French version (now amended to reflect the English) stated:

462.31 (1) Est coupable d’une infraction quiconque – de quelque façon que ce soit – utilise, enlève, envoie, livre à une personne ou à un endroit, transporte, modifie ou aliène des biens ou leurs produits – ou en transfère la possession – dans l’intention de les cacher ou de les convertir sachant ou croyant qu’ils ont été obtenus ou proviennent, en totalité ou en partie, directement ou indirectement :

a) soit de la perpétration, au Canada, d’une infraction de criminalité organisée ou d’une infraction désignée;

b) soit d’un acte ou d’une omission survenu à l’extérieur du Canada qui, au Canada, aurait constitué une infraction de criminalité organisée ou une infraction désignée.

The material part of the section is that which is italicised in the English version of the text, which is omitted from the French version. The defendants had not committed any of the specific acts listed in s462.31, but they could be said to have ‘otherwise dealt with’ the goods. Accordingly, under the English version of the Criminal Code, they were liable, but they were not liable under the French version. The Supreme Court, therefore, had to interpret the two statutes and decide a) what the correct meaning was and b) what the appropriate meaning was in the circumstances. In this case it was accepted that the English version most accurately reflected the intentions of the legislature, but that the French version was the one that must be regarded as textually correct, because only the French text contained a meaning that was common to both versions.

What is important to note in the Daoust judgment, however, is that it was the court that was able to decide on an appropriate interpretation of the law by evaluating the meaning of the two texts. However, as regards the interpretation of the law of Wales, this would not be possible. At the present time, no member of the Supreme Court is able to understand Welsh. The situation is a little better in the Court of Appeal as some of the Appellate judges are able to understand Welsh. Furthermore, although the Supreme Court is constituted from at least one judge with expertise in Scots law, and one judge with expertise in the law of Northern Ireland (Constitutional Reform Act 2005 (2005 c.4) s.27(8)) there is no member of the Supreme Court with expertise in relation to the law of Wales.

3. The challenges of translation

The situation described above therefore requires the input of an interpreter, because it will be the interpreter who will need to explain the possible differences in nuance and meaning between the Welsh and English versions of the text. The task of creating two texts whose meaning is exactly the same is onerous – indeed, some would say impossible (Jamieson 2011: 343). The task of explaining those differences is even more significant for a number of reasons.

Firstly, although it is often assumed by monolinguals that a text may be translated exactly, Armstrong (2005) explains how difficult this is in reality. Two languages may conceptualize an idea very differently with the result that a term in one language may include connotations that are entirely absent in the other. Cao (2007a), for example, explains that in the context of the US–China spy-plane incident the difference in meaning between ‘regret’ and ‘apologise’ was highly significant. The English version expressed the US Government’s regret about the missing aircraft, but this was rendered in Chinese as an apology. Similarly, a word may have connotations in one language that are less immediately apparent in the other. For example, the Welsh word for loyalty, ‘teyrngarwch’ is derived etymologically from ‘love of tyrant’ and is therefore conceptually very different from the English notion of loyalty.

Metaphor also creates particular difficulties for the interpreter. For example, in English, proceedings are said to be ‘conducted’. In Welsh, however, they are said to be ‘supported’. However, when interpreting the meaning of these two phrases, the English version includes connotations of the proceedings being controlled by a conducting agent, whereas in Welsh the conceptualisation is more service- focused – there are no implications that the proceedings are being orchestrated, and the agent’s role is seen to be more facilitative.

Armstrong (2005: 13) also explains that there may be differences between the use of hypernyms and hyponyms, with one language having one word for that which may be translated as several different hyponyms in the other language. He explains this with reference to the distinction that exists in French between the words ‘fleuve’ and ‘rivière’, even though both words would be translated into English as ‘river’:

One well known example of an organisation that differs between French and English is expressed in the fleuve-rivière lexical pair … the Seine is a fleuve while a river like the Yonne … is a rivière. It seems then that speakers of French have a hypothesis of rivers that differs from the English one, or if we feel that this term is being stretched too far, we can say that the French language makes explicit a view of river systems that is latent in English, since the latter language does not encode this distinction so compactly.

A third difficulty that arises from the reliance on translation, specifically in the context of Welsh and English law, is that because the production of legal texts through the medium of Welsh is a very new development after many hundreds of years in abeyance (Watkin 2007), inconsistency may be encountered in the codability of the two languages for legal purposes. It is argued (Armstrong 2005: 31) that, through custom and the necessity of practice, speakers of a language develop compact terms that encapsulate very complex ideas. For example, s.1(c) of the Local Government Byelaws (Wales) Act 2012 (2012 NAW 2) includes the word ‘requires’, a term that can only be rendered in Welsh as ‘yn ei gwneud hi’n ofynnol i’. Perhaps English has a longer history of ‘requiring’ particular conduct than Wales, as a more succinct term has evolved in order to describe ‘requiring’. However, its significance with regard to legal translation is that the long history of English law means that more legal terms have developed succinct words to describe legal concepts, meaning that explaining these concepts in Welsh involves explaining that the words used represent an attempt to describe the concept and may therefore only be an approximation of the term used.

The converse is also true, however. The newness of Welsh legal terminology means that Welsh has more flexibility to be able to coin new terms. These may explain a concept in a more straightforward way, without being hindered by the several hundreds of years of legal history that may have caused words to acquire connotations that are no longer relevant. For example, the term ‘custody’ is still commonly, though incorrectly, used in relation to the residence of children. However, the development of terminology in Welsh for this concept meant that different words could be used for ‘the imprisonment of a person suspected of or charged with a criminal offence’ on the one hand (‘yn y ddalfa’), and ‘an order regarding who a child should reside with’ (‘cadwraeth’) on the other. Welsh legal terminology may be developed in ways that do not have connotations derived from existing concepts and therefore may develop terms that represent current law more accurately. Yet, this is also an issue that affects the scope to transpose meaning from one language to another.

A further issue is that many Welsh laws are made in order to be part of a larger legal framework. This may occur either because secondary legislation is made pursuant to a primary Act, or it may be in the form of primary legislation that amends an existing Act. Education, for example, is a devolved matter under Schedule 7 of the Government of Wales Act 2006. However, the primary legislation includes laws that pre-date the granting of primary legislative powers to the National Assembly for Wales. Accordingly, the National Assembly for Wales has made amendments to legislation such as the Education Act 1996 (1996 c.56) and has also enacted secondary legislation pursuant to that Act. In both cases, the legislation must be bilingual. This introduces another dimension into the interpretation of Welsh law. Because the wording of the law – in English at least – must be consistent with the terminology used within the existing law, this then constrains the possible wording that may be used in the Welsh version, and compels choices to be made at the drafting stage regarding the intended meaning. For example, in English the word ‘shall’ may be problematic. ‘Shall’ could mean either ‘must’ or ‘will’ and the Welsh words ‘bydd’ or ‘rhaid’ only convey one of those meanings. The draftsperson must therefore choose whether the legislature’s intention is to convey a current obligation or a future change of behaviour. However, when the matter comes to be interpreted by the courts, the Welsh and English versions appear to be inconsistent because a broader range of meaning is included within the English text than the Welsh text. The possibility is then that the narrower text is to be preferred as encapsulating both versions, as in Daoust above.

These issues all pose significant challenges for the interpreter, whose judgment in this context will be crucial for determining the outcome of the case. The interpreter must therefore act as an expert by assisting the court to adjudicate the matter. This represents a significant expansion of the interpreter’s role, whose function is normally to translate written evidence for the benefit of the court. It also introduces the need for an interpreter into the higher courts, where hitherto their function has mainly been in relation to hearings at first instance. Expert evidence is fairly commonly used in trials at first instance in order to determine causality, liability or the quantum of damages. However, the role of the expert is generally to ascertain fact. However, the bilingual interpreter’s primary role in cases pertaining to the law of Wales is as an adjudicator of law. Accordingly, further consideration must be given to the admissibility of expert evidence into the higher courts for the purposes of presenting evidence on what the law means.

Firstly, there is a need to consider what the nature of the interpreter’s expertise may be. A linguistic expert may be required in order to explain the difference in meaning between, for example, hyponyms and hypernyms, such as the fleuve/rivière distinction described by Armstrong (above) or the distinction between droit and law described by Cao (2007b). A linguistic expert may also be able to explain the semantic meaning of the legislation in terms of, for example, whether the versions are written in the same tense or whether the two versions identify the same subject and object, or whether one version lacks an agent whose identity is made manifest in the other version.

Secondly, the interpreter’s translating expertise must not be overlooked – a linguistics expert is not necessarily an interpreter, and therefore, although the linguistics expert may be able to give an account of meaning in one language, they may not necessarily be an expert interpreter. Therefore, the interpreter must possess the skill of translation, of being able to render into the other language the different possible meanings of the terminology. Ordinarily, an interpreter translates meaning, and therefore a change of terminology may give a more accurate representation of meaning. For example, where there are cultural referents in one language, it may be necessary to render them slightly differently in the other language. However, with the translation of the law, there is a need to ensure that the translation is faithful to the original. The interpreter must also be conscious of the fact that there may be different possible nuances to different translations, and must therefore be able to explain these to the court.

Furthermore, it is probable that the interpreter will also need to have a level of legal expertise in order to be aware of whether differences between two language versions of the text are significant. In cases where the section of legislation to be scrutinised by the court is part of a larger legal framework, this is particularly important because the interpreter will need to be able to explain how the dual-language meaning of the section fits in within the larger framework of amendments, insertions and repeals of legislation enacted both by the Westminster Parliament and by the National Assembly for Wales. For example, the Education Act 1996 (1996 c.56) is, in part, a consolidation of earlier legislation on education enacted between 1944 and 1996, primarily drawing upon the Education Act 1944 (1944 c.31) and the Further and Higher Education Act 1992 (1992 c.13). However, it also introduces new provisions following recommendations made by the Law Commission. Later legislation, specifically the Apprenticeships Skills, Children and Learning Act 2009 (2009 c.22), the Children, Schools and Families Act 2010 (2010 c.26) and the Education Act 2011 (2011) c.21), has amended the Education Act 1996, and this has meant that some sections have been repealed in relation to England, with amendments inserted, while other sections have been repealed in relation to Wales, with different amendments introduced. Accordingly, the interpreter will need the legal expertise to be able to differentiate between the sections applicable to Wales and the sections applicable to England. Furthermore, a large body of subordinate legislation created pursuant to the Act has resulted in further distinctions emerging between the law in England and the law in Wales. In order to explain the meaning of the Education Act 1996, therefore, the interpreter will need to have regard to the ways in which the Act’s interrelationship with these other statutes affect meaning. Therefore, the possibility that the interpreter may also need to be legally trained cannot be discounted.

Clearly, these three requirements involve interdisciplinarity of expertise, and therefore the task of ensuring that an interpreter is an appropriate expert is an onerous responsibility. Nevertheless, Cao explains that this is not insurmountable, because ‘the deep knowledge structures of a translator in a particular field can be enhanced and intensified but does not need to be the same as those of the specialist’ (Cao 2007b: 46). A decision must be made as to who should appoint the expert – should it be the party who is seeking to rely on the differences between one version of the text and the other, or should it be the court itself, as the primary beneficiary of the expert opinion? Accordingly, one important aspect that must be considered is who bears the responsibility – and ultimately the cost of appointing an expert interpreter.

A further issue that arises from the bilingualism of the law is the desirability of the situation where the interpreter assumes a judicial role. Because the court will not be able to evaluate the text for itself, the expert’s opinion acquires paramount importance to the adjudication of the trial. However, the implications for this are highly significant, in that by determining the meaning of a dual-language text, the interpreter assumes the role of the judge. This is something that is likely to raise serious concerns in terms of due process and fairness, because a case will be adjudicated by an interpreter, and also because the procedure for deciding cases under the law of England will be different from the procedure applicable under the law of Wales. The monolingual law of England will be interpreted by the judiciary, whereas the bilingual law of Wales will be initially interpreted by the interpreter, before the court is able to adjudicate on its meaning. This dichotomy calls into question the whole of the law’s authority – if some laws are interpreted according to different processes from others, how is the fairness of the law to be evaluated?

Thirdly, it must be borne in mind that the interpreter’s expertise relies heavily on opinion. In other situations where an expert is called upon, it may be possible to come to a consensus of opinion regarding, for example, the cause of an injury. That is to say, it may be ‘recognised as proper by a competent reasonable body of opinion’ (per McNair J. in Bolam v Friern Hospital Management Committee [1957] 2 All ER 118 at p.122). Although it is possible for interpreters to arrive at such a consensus of opinion, it must nevertheless be borne in mind that, although language speakers may agree on broad meanings, the nuances and connotations of a word or a phrase may be far more idiolectical. Therefore, pragmatic theorists (Canale and Tuzet 2007) may often disagree about whether, for example, Grice’s conversational maxims (Grice 1975) are being followed or flouted, and whether the flouting is intentional or not, and argue that different meanings may be inferred by different listeners. Accordingly, the notion that it is possible to translate the meaning of a legal text into another language may fall short of expectations. The interpreter may translate possible meanings, and offer an opinion as to which is most appropriate, but the difficulty with the law of Wales is that the court is not able to evaluate this advice for itself.

Non-bilinguals may also underestimate the limitations of two languages in terms of mapping directly onto one another (Robinson 2003: 10) and may therefore perceive opinion to be tantamount to truth. Various translators comment about having experienced this form of presumption (Venuti 1995) and, therefore, there needs to be an increased awareness of the scope and limitations of the interpreter’s role.

4. Solutions

What solutions may therefore be proposed? Regrettably, within the existing legal framework, there are no easy solutions. In other bilingual jurisdictions, bilingual capacity has developed over time through the creation of bilingual legislation, and extensive education and judicial training programmes aimed initially at developing the professional fluency level of those who were dual-language native speakers, and then working on developing fluency among those who spoke one official language as their second language. Therefore other bilingual jurisdictions such as Canada and Spain have succeeded in bilingualising their legal system within one or two generations (Legislation Committee of the National Assembly for Wales 2006). However, by creating a legal system where only the legislation must be bilingual, there has been a failure in the United Kingdom to foresee the need to build a mechanism for interpreting those laws. Accordingly, in the longer term, regard must be given to the need to ensure that the judiciary is able to interpret bilingual law without recourse to an interpreter.

In the meantime, however, all that may be achieved is to use an interpreter’s expertise to assist the court in interpreting the meaning of a bilingual text. Accordingly, there is a need to consider the following issues.

Firstly, there is a need to define what the interpreter’s role should be. In what ways may they assist the court without usurping the court’s judicial functions?

Secondly, both the court and the interpreter should be aware of the fact that the interpreter’s role is to offer an opinion. Accordingly, the interpreter must be able to offer a range of possible meanings as well as deliver what might be the consensus of reasonable opinion regarding what the most probable meaning should be. The court should also be made aware of what differences of nuance are likely to be significant. This may be particularly important where both texts are correctly drafted, i.e. free from error. Consideration may also need to be given to whether the court will rely on one interpreter or whether several interpreters may need to be called upon in order to ensure that a consensus of opinion is obtained.

Thirdly, sufficient time must be allowed in the court proceedings to explain the different possible nuances of the two texts. Accordingly, it is necessary to ensure that the judiciary fully understands the implications of favouring one interpretation over another. If a word is translated in one way then its range of meaning may be entirely different from other possible choices that could have been used. Linked to this, there needs to be an awareness of the implications of favouring one version over another. The Canadian approach has hitherto been to favour a shared meaning (cf. R v Daoust, discussed above), and, therefore, the narrower version may be a more accurate representation of the meaning shared between the two versions. On the other hand, following the Daoust case, this approach was very severely criticised by those who felt that the correct version should be the version that most accurately reflects the intention of Parliament (Salembier 2003).

The fourth matter to be considered is the nature of the interpreter’s expertise. There must be a clear understanding of the different skills required – linguistics, translation and law – in order to ensure that the interpreter relied upon is appropriately qualified. If a combination of skills is deemed to be necessary, then it becomes important to ensure that appropriate training is available in order to create a pool of interpreters who possess those skills. There must also be an awareness of the fact that there are significant differences between an interpreter and a translator. Courts may be more accustomed to relying on the services of an interpreter whose function is to interpret what is said in English for the benefit of the court. However, the interpreter’s role in this context is to translate the written text and to explain the differences of meaning between the two versions. Accordingly, there must be a recognition that the courts should not rely on a forensic interpreter for the purposes of carrying out the evaluation, but should instead rely on a translator.

Furthermore, it is necessary for the interpreter to be aware of and to explain how their intervention affects the proceedings. In earlier work, the current author has commented upon the fact that the process of discovering plain and ordinary meaning (Huws 2012) involves the listener in complex decision-making processes that include the elimination of improbable meanings according to context, and the negotiation of assumptions concerning shared meaning. Interpretation of legal texts in one language involves engagement with this type of semantic and pragmatic process. However, where the court is unable to read one version of the text, the interpreter must assume this role. This means then that the interpreter must be aware of the fact that that which may be so obvious to him or her that it goes without saying must be explained for the benefit of the court, as the interpretation given will not only affect the immediate parties to the litigation, but also the interpretation of cases in the future. The interpreter must also be aware of assumptions that he or she may make. Turns of phrase that have become the standard means of expressing a concept, such as, for example, the fact that legal proceedings are conducted through the medium of English but are supported through the medium of Welsh, may have become so normalised within the language that the interpreter does not notice their particular meaning and connotations. The use of a particular metaphor may be so deeply entrenched that its status as a metaphor may be overlooked even by the interpreter, and it is only when the meaning is scrutinised that the different connotations become apparent. For example, in the sentence that follows this one, the phrase ‘packed into a metaphor’ will be used. Although this is a common way of expressing how extensive meaning may be conveyed succinctly, the status of the phrase as a metaphor may be overlooked because a metaphor is not a suitcase or a bag into which things may be packed, and therefore the interpreter must be aware of the assumptions he or she may make regarding the nuances packed into a metaphor.

Another issue to consider is the uneven availability of the two versions of the legal texts. Although and the National Assembly for Wales’s websites contain both language versions of Welsh legislation, concerns have been expressed regarding whether these resources are the most up-to-date representations of the current law (Hughes and Davies 2012). Furthermore, the commercial publishers do not produce the bilingual texts of legislation for Wales (Hughes and Davies 2012) and do not perhaps realise that the information they maintain relating to Welsh law is therefore incomplete. As Melding explains:

We are concerned that it should be possible to ascertain the law as easily in both languages and that is not yet the case. Whilst the website does publish both texts ‘as made’, the Welsh text is not currently updated. That means that the texts that are available do not necessarily correspond, and it is therefore necessary to look separately at legislation amending the Welsh text to ascertain the current state of the law. That is not an acceptable situation, and I am pleased that the Welsh Government is in discussion with The National Archive to ensure that in future texts in both languages are updated at the same time.

Nevertheless, the work of the National Archive is a step ahead of commercial legislative databases, which do not publish the Welsh language texts at all. Clearly, they are driven by commercial considerations, but it does not assist their clients if they are only able to see half of the legislation made by the Assembly. At the very least, they should include a warning that the legislation that they publish has been made bilingually, so that lawyers know that they should also check on the website for the second text. (Melding 2012: 101)

Accordingly, the legal jurisdiction that is England and Wales also needs to be aware of the need to ensure that members of the judiciary are able to access the relevant legal material in the two languages, and perhaps it is because of this lack of accessibility that the issue of bilingual texts was not addressed in the recent case of Local Government Byelaws (Wales) Bill 2012 – Reference by the Attorney General for England and Wales [2012] UKSC 53. Although the Supreme Court was called upon to interpret the meaning of sections 6 and 9 of the (then) Local Government Byelaws (Wales) Bill 2012, they did so only with reference to the English version of the law. However, the primary point at issue in this case was not the meaning of the words in the bill, but instead whether what the bill purported to do was within the legislative competence of the Government of Wales Acts 1998 and 2006, and the National Assembly for Wales (Transfer of Functions Order) 1999. Accordingly, the outcome of the case did not turn directly on the interpretation of the meaning of the Local Government (Byelaws) Wales Bill, merely on the question of whether their general scope was within the powers granted by the earlier law. However, neither of the Governments of Wales Acts nor the National Assembly for Wales (Transfer of Functions Order) 1999 were produced bilingually – as all three laws emanated from the Westminster Parliament. Nevertheless, there were aspects of the Local Government Byelaws (Wales) Bill 2012 that were interpreted by the Supreme Court. For example, at paragraphs 60–63 of the judgment, Lord Neuberger explains:

Section 9 of the Bill would have the effect of enabling the Welsh Ministers to add to (and to subtract from) the scheduled enactments, which would then become subject to the s 6 procedure, rather than the s 7 procedure. As already explained, the crucial difference for present purposes between the two procedures is the requirement under s 7 for confirmation of the byelaw by Welsh Ministers and/or the Secretary of State or other Minister of the Crown (depending on the statutory provision under which the byelaw is made) – see, in particular, s 7(10) to (12).

The Attorney General argument is that s 9 would ‘confer power’ on the Welsh Ministers ‘by subordinate legislation to remove or modify … pre-commencement function[s] of a Minister of the Crown’. Accordingly, he argues, by virtue of s 108(6)(a) of, and para 1(1) of Pt 2 of Sch 7 to, the 2006 Act, the section is outside the legislative competence of the Assembly.

If s 9 is to be interpreted as giving the Welsh Ministers power to add to the scheduled enactments any enactment which gives the Secretary of State or another Minister of the Crown a confirmatory function in relation to byelaws, then I would accept that argument. However, there could be no objection to the section, if the scope of the power it would confer on the Welsh ministers was limited to byelaws made under enactments which currently satisfy one of two requirements. Those requirements are that the enactment concerned (i) identifies the Welsh Ministers and not a Minister of the Crown, as having the confirmatory power, or (ii) identifies a Minister of the Crown as having the confirmatory power, but the removal of that power would be ‘incidental … or consequential’ within the meaning of para 6(1)(b) of Pt 3 of Sch 7 to the 2006 Act. The basis for requirement (i) is self-evident, and the basis for requirement (ii) is the same as that for concluding that s 6 is within the legislative competence of the Assembly.

Although it is perfectly true that there are no express words in s 9 which limit its scope in this way, I am satisfied that it does have such a limited effect. That is because of the simple legal principle, identified by Lord Reed, embodied in the Latin maxim nemo dat quod non habet. Given that the jurisdiction of the Assembly is limited to removing, or delegating the power to remove, functions of Ministers of the Crown when the removal satisfies the requirements of para 6(1)(b) of Pt 3 of Sch 7 to the 2006 Act, the Assembly cannot confer a wider power on Welsh Ministers. Accordingly, the wide words of s 9 must be read as being circumscribed in their scope so as to render the section valid.”

Although this discussion relates more to the general content of sections 6, 7 and 9, the question of whether the legislation was within or without the Welsh Government’s legislative competence is more readily answered with reference to the Welsh version of the text. For example, the Welsh version on s9 does not ‘confer power’ on the Welsh Government. Instead, it gives a more unequivocal declaration of the entitlement to act. Accordingly, the Welsh version provides a far clearer demonstration that reforming Part 1 of Schedule 1 is something that is permitted. Accordingly, it would have been interesting to see whether the Supreme Court would have accepted the Welsh Government’s legislative competence so readily if it had been able to have regard to the Welsh version of the text, or if it had chosen to do so through the involvement of an interpreter. All these issues are matters that need to be resolved in the short term, and before a matter involving the bilingual law of Wales proceeds through the court process.

5. Conclusion

What this article demonstrates, therefore, is that an unrecognised consequence of the devolution settlement to Wales is the need for courts in England to interpret bilingual legislation. Currently they cannot do this without recourse to a legal interpreter, and therefore, consideration must be given to how to define that role, and the extent to which reliance on an interpreter is acceptable in the longer term. If not, then consideration must be given to how England and Wales moves from a monolingual jurisdiction to one that is functionally bilingual.

About the author

Catrin Fflur Huws is a lecturer in Law at Aberystwyth University. She is also the director of the Centre for Welsh Legal Affairs, a research centre dedicated to the Department’s expertise and work on the law as it applies within Wales and on general legal developments of relevance to Wales.


Armstrong, N. (2005) Translation, Linguistics, Culture: A French English Handbook. Clevedon: Multilingual Matters.

Beaupré, M. (1988) Litigating the meaning of bilingual legislation. Advocates Quarterly 9: 32.

Canale, D. and Tuzet, G. (2007) On legal inferentialism: towards a pragmatics of semantic content in legal interpretation. Ratio Juris 20(11): 32.

Cao, D. (2007a) Inter-lingual uncertainty in bilingual and multilingual law. Journal of Pragmatics 39: 69–83.

Cao, D. (2007b) Translating Law. Clevedon: Multilingual Matters.

Fung, S. Y.-C. (1997) Interpretation of the bilingual legislation of Hong Kong. Hong Kong Law Journal 27: 206.

Grice, H. P. (1975) Logic and conversation. In P. Cole and J. Morgan (eds) Syntax and Semantics vol. 3, 41–58. New York: Academic Press.

Hughes, D. and Davies, H. (2012) Accessible bilingual legislation for Wales (Deddfwriaeth hygyrch a dwyieithog i Gymru). Statute Law Review 33(2): 103.

Huws, C. F. (2012) Is meaning plain and ordinary? Are you sure about that? Statute Law Review 33(2): 230.

Jamieson, J. (2011) English in the 2010s: getting up close and personal. Victoria University of Wellington Law Review 42: 343.

Legislation Committee of the National Assembly for Wales (2006) Bilingual lawmaking and justice: a report on the lessons for Wales from the Canadian experience of bilingualism. Site accessed 27 February 2013.

Melding, D. (2012) Foreword: issue on Welsh devolution. Statute Law Review 33(2): 97.

Osmond, J. (2003) Laying the foundations. In J. Osmond and J.B. Jones (eds) Birth of Welsh Democracy: The First Term of the National Assembly for Wales xix–xxx. Cardiff: Institute of Welsh Affairs.

Robinson, D. (2003) Becoming a Translator: An Accelerated Course (2nd edn). London: Routledge.

Salembier, P. (2003) Rethinking the interpretation of bilingual legislation: the demise of the shared meaning rule. Ottawa Law Review 35: 75.

Venuti, L. (1995) The Translator’s Invisibility. London: Routledge.

Watkin, T. G. (2007) The Legal History of Wales. Cardiff: University of Wales Press.

Legal references


Apprenticeships Skills, Children and Learning Act 2009 (2009 c.22).

Constitutional Reform Act 2005 (2005 c.4).

Children, Schools and Families Act 2010 (2010 c.26).

Criminal Code, R.S.C. 1985, c. C-46.

Education Act 1944 (1944 c.31).

Education Act 1996 (1996 c.56).

Education Act 2011 (2011) c.21).

Further and Higher Education Act 1992 (1992 c.13).

Government of Wales Act 1998 (1998 c.38).

Government of Wales Act 2006 (2006 c.32).

Local Government Byelaws (Wales) Bill 2012 – Reference by the Attorney General for England and Wales [2012] UKSC 53.

Local Government Byelaws (Wales) Act 2012 (2012 NAW 2).

National Assembly for Wales (Transfer of Functions Order) 1999 (no 672).

Northern Ireland Act 1998 (1998 c.47).

Scotland Act 1998 (1998 c.46).

Welsh Language (Wales) Measure 2011 (2011 NAWM 1).

Cases cited

Bolam v Friern Hospital Management Committee [1957] 2 All ER 118.

R v Daoust [2004] 1 SCR 217.


  • There are currently no refbacks.

Equinox Publishing Ltd - 415 The Workstation 15 Paternoster Row, Sheffield, S1 2BX United Kingdom
Telephone: +44 (0)114 221-0285 - Email:

Privacy Policy