Legislative Intent and Ordinary Meaning

ROSS PEY is a PhD candidate specializing in statutory interpretation (including comparative statutory interpretation) at the University of Western Ontario. His thesis maps the current approach to statutory interpretation in the UK, including the nature of bilingual interpretation in Wales. In Edinburgh, he analyzed the interpretation and operation of Brexit legislation. Additionally, Ross retains an interest in tax law (specifically tax avoidance and interpretation) and the conflict of laws. Ross has held visiting posts at the University of Cambridge, Yong Pung How School of Law (Singapore Management University), and the National University of Singapore. Ross is also appointed as the UK national reporter for European Collaborations on the conflict of laws and tax interpretation.

You study the interpretation of bilingual legislation in Wales as part of your larger focus on statutory interpretation. How did you become interested in this specific topic?  More broadly, what is your relationship to the topic of law and language?

I became interested in this topic because, in the UK, statutory interpretation isn’t usually taught properly. We briefly go through it as an introduction to the English legal system. In a lot of places, they still say there are three rules of interpretation: the literal rule, the mischief rule, and the golden rule, and the judges pick one that does justice to the parties at hand. That doesn’t make sense because you start thinking: what do you mean that there are only three rules?  What does the interest of justice mean? Additionally, it can’t be that simple, with only three rules, right? I just thought that was hugely unsatisfactory.

When I was in my first and second years, I did my own reading on it. I found contract interpretation fun, and even though it wasn’t taught in that much detail, I read up on it on my own. In my third year, I decided to do a tax paper. How it was done in Cambridge, and still is, is that you don’t actually have a tax subject per se. Instead, it’s a tax dissertation, in which I wrote about tax avoidance.

Tax avoidance in the UK is still thought of as an application of the normal principles of interpretation to then judge whether taxpayers are inside or outside a chartering provision. The UK has some fairly aggressive tax avoidance cases where the courts apply a purposive approach. My paper was an exploration of the purposive approach in a few tax avoidance cases. If we take the purpose of interpretation seriously, what does it mean? How does it apply? Are there any issues we can see in tax avoidance cases? That was my first big foray into interpretation, and that was sort of set the stage for some of my work now.

Now, my interest in bilingual interpretation came about recently. The constitutional arrangement in the UK is such that the Parliament of the United Kingdom is at the top, and then you have what we call devolved settlements, or devolution settlement arrangements. Scotland is quite autonomous, with its own laws and legal system. Likewise, in Northern Ireland as well. Wales, however, is in a weird position because it’s part of England as one jurisdiction. But via the Government of Wales Act, the devolution settlement for Wales, Wales has devolved competencies in various public law areas. The legislation also allows Welsh legislation to be legislated in Welsh and English. This bilingual legislation was a first in the modern history of the UK. The Law Commission has written up its view on how you ought to consider a bilingual legislation’s object and purpose in the event of a conflict.

As I pointed out in a speech a few weeks ago, I applaud and broadly agree with the Law Commission’s work. For interpretation, it effectively focuses on two issues. One of them was, what’s the approach to interpretation? The other one is the conflict issue.

Now, if you see what goes on in a lot of other Commonwealth jurisdictions, like Malaysia, Hong Kong, South Africa, and even the EU for the matter, you realize that, for bilingual and multilingual statutes, there are way more issues than what the correct interpretation method is and how we deal with conflict. There wasn’t much written about this in the Welsh context because it’s so new, and it’s within legislation and statute, so it’s a niche within a niche. At some point, we need to face the fact that the UK is a bilingual jurisdiction with respect to Wales. We need to get this right for constitutional reasons. How do we go about doing that? That entails an analysis of comparative legislation and comparative statutory interpretation, mainly with Canadian and Hong Kong cases, to say: ‘Do we think it’s right? Do we think it’s wrong? What are the reasons why we think it’s right or a problem?’ I thought that, given my work on interpretation and deconflicting texts, I might be able to have a go at this problem.

Bilingual interpretation is relevant to the Canadian context. Can you explain some of the issues or challenges related to bilingualism that are common in both Canada and Wales? How is comparative interpretation used as a tool?

We can view similarities based on how the jurisdictions implement bilingual legislation. In both jurisdictions, the languages used in legislation are equally authoritative. In Wales, we say that both languages are of ‘equal standing’. In Canada, we say that French and English language versions are ‘equally authentic’. What these labels mean, albeit with different words, is that both language versions represent the law, and you can rely on either language version.

Additionally, there is great similarity in how bilingual legislation is passed. Generally, the modern practice of the Canadian federal Parliament is to draft both language versions concurrently. In Canada, jurilinguists are also involved in co-drafting bilingual legislation. In Wales, the practice is also to debate and pass legislation in both Welsh and English. However, the current practice is not quite like Canadian federal practice, and there is also some amount of translation involved in the drafting process.

 You can see quite quickly that we are going to have an issue, which is what the proper way of interpreting it is. If we say that both language texts are authoritative, on the ground, it means that if you are Welsh or a French speaker, you can just rely on your language’s text. As a matter of law, once we start talking about giving legal advice, trying to write judgments, and trying to interpret legislation, equally authoritative legislation means that you can’t just rely on one of them. In principle, in an equally authoritative text, Parliament is legislating one intent in two languages. At the level of giving legal advice and interpreting, you invariably have to look at both language versions. This is the position in Canada. In Wales, it’s not that sharply written, but that’s what the Law Commission has suggested. This is the position in the EU, nominally – you’re supposed to consider all 23 language versions of EU legislations.

On top of these, there is the issue of bijurality in Canada, exemplified in the position of Quebec. Quebec is not just a jurisdiction that uses a different language, like New Brunswick. Quebec has a completely different legal tradition, the French civil law. When you draft legislation, you need to account for that. Suppose you have federal legislation that regulates contracts of sale. Now, what counts as a contract? If you’re in a common law province, it’s a relationship where there is offer, acceptance, consideration. A civil lawyer would say that there is a contract when there is an offer and acceptance. The presence of consideration is not a consideration (pun intended).

Therefore, with a jurisdiction like Quebec within Canada, it’s not just drafting in two languages; it’s also taking into account the different legal heritage and laws of Quebec as well. We sort of see that in the UK with respect to Scotland. Scotland has a different legal system: Scots Law, a hybrid of civil and common law. Because of the Acts of Union, Scotland has retained its own law. For UK legislation which applies to Scotland, the legislation usually takes into account the differing position in Scotland. The example I used in the talk is the meaning of defamation for the purposes of the Private International (Miscellaneous Provisions) Act 1995,  regarding the choice of law rule for defamation. The Act specified that ‘defamation’ meant libel and slander in England and Wales and Northern Ireland, and verbal injury in Scotland.

Right now, bilingual legislation is a relatively new phenomenon in the UK, as the Government of Wales Act came about in 2000.  Further, the Welsh legislature (the Senedd) has a strong legislative program. They want to consolidate a lot of rules that are within their devolved competency.  When the UK Parliament is drafting legislation with an effect in Wales, it might be the case that there will be very distinct aspects of Welsh law that they have to take into account, but it is obviously going to be more limited to public law issues. It may be that the legislative programme of the Senedd may have an impact on bilingual drafting, as the legislation has to account for both Welsh language and Wales-specific doctrines.  If you take a broader comparative legislative approach to this issue, you realize it’s not just interpreting words. It is also what might be the consequence if you have a sui generis body passing laws and legislating on it.

Is there also something that we can see from Canada that is a model for the UK?

The Law Commission effectively rejected the Canadian approach in Daoust. There, the approach is that if there is a conflict, you apply a series of bright-line rules. If it’s between a narrow and a broad one, you go with the narrow one. If you go with ambiguous versus non-ambiguous, you go with the non-ambiguous one. For step three, you check against legislative intent. The downside of that approach is that it’s a stilted exercise.  Why is it that we go with the narrow meaning first, rather than the broad one?  What do narrow and broad mean? We don’t really know what is ambiguous and non-ambiguous because the Canadian courts are never consistent on that point. Once you break it down into its steps, you find that it’s stilted because it makes a lot of policy assumptions right at the front. Then the court has to check to see if you can rebut those presumptions. In fairness to the Court in Daoust and Bastarache J, the approach makes intuitive sense because it’s regimented – it gives you some sort of an idea of what to do when you meet conflicting language texts.

Where I think I depart from the Law Commission is the resort to the object and purposes of a provision. To give some context, the courts say that when you interpret legislation, you apply the purposive approach. This means that you read legislation with respect to its text, context, and purpose. A side note here is that, if we accept that the purposive approach is the correct approach to interpreting statutes (including bilingual statutes), how does the Daoust approach (selecting the unambiguous or narrow meanings) fit in within this broader framework?

In the Law Commission report, it says that, where conflicts arises, you, as the interpreter, ought to read the conflicting text in a way that best satisfies the purposes of the provision. In the abstract, this has to be right, especially if we accept the purposive approach as the correct approach to interpreting statutes. Where I take issue is that I think that it is a vast oversimplification of the work needed to deconflict a bilingual text.

Let me illustrate this. I think we can accept that, with bilingual legislation, Parliament has one intent but expressed in two languages. You have to read two languages to try to find out as far as we can what the meaning is that’s going to be harmonious to both language texts. Instead of using the stilted rules, this is a broader exercise – we consider everything with special attention to the purposes of the provision and try to articulate a harmonious meaning. This makes intuitive sense if we accept that we are trying to find what Parliament’s intent, expressed in two languages, is.

However, in practice, there are issues like the Daoust case where there’s a drafting error between both language versions. A simpler example I used in my talk is where one language version says 31st of June and the other says 31st of July. You are not going to square it. You can’t say “I’m going to pick somewhere in the middle and say, middle of July”. In really extreme circumstances, you have to make the choice. Of course, you’re going to use ‘text, context, purpose’ to inform that choice, but at that point, we can’t give a harmonious meaning. We just have to prioritize one over the other. You can’t say, “Parliament made an error here. I’m not going to resolve it. Parliament fix it first.” That’s not the practice here. You have to give effect to one of them or both. You need good reasons with respect to ‘text, context, purpose’ to do that.

Moving away from Canada to talk about another area that you’ve worked on, which is mapping the interpretation and operation of Brexit legislation, what was a key interpretive issue or theme that you encountered? Are there different considerations of statutory interpretation for politically charged legislation, such as Brexit?

For the second bit, the answer is no, the courts apply the same approach to interpreting the legislation effecting the UK’s exit from the EU.

With respect to Brexit, the issue has to do with how we approach EU laws that have been retained in the UK after Brexit.

Historically, when the UK joined the EU, then the European Economic Community (EEC), they knew that the EEC (1) is a supranational body and (2) its legislation needs to be given effect in the UK. If you are a monist jurisdiction, like Germany or the Netherlands, it’s easy because anything that happens on the international law plane is also national law.

Unfortunately, the UK is a dualist jurisdiction—international law is only effective in the UK if Parliament incorporates it via legislation. For EU/EEC legislation, there were a few solutions for this. One of them was to say that whenever the EU passes legislation, the UK will pass additional legislation domestically to give effect to that. The problem with that approach is that the amount of legislation you have to pass will be horrendous, given the pace of EEC legislation. The final solution dealt with EU legislation in two ways. For EU regulations, EU laws that are intended to be directly effective at the member state level, the UK Parliament enacted section 2(1) of the European Communities Act 1972,which says that directly effective legislation and EU treaty rights will also be directly effective in the UK. It’s sort of like a tunnel that brings things in. Miller describes this as a “conduit pipe”. Secondly, for directives, EU legislation directed to member states for further implementation, section 2(2) of the same Act allows the government the jurisdiction to draft secondary legislation to implement those directives.

 To implement the UK’s legal withdrawal, the European Union (Withdrawal) Act 2018, sections 2, 3, & 4 (mirroring section 2(1) and 2(2) European Communities Act 1972) essentially stipulates that EU laws which have entered the UK via the 1972 Act will still have effect, even though we’re out of the EU.  In 2023, the UKenacted the Retained EU Law (Revocation and Reform) Act 2023 which repeals certain types of retained EU law.

The main issue that now arises is that we’ve retained EU law through the 2018 Act and the 2023 Act. How do we interpret EU law that has been retained through this mechanism? The operation of the 2018 Act was to make retained EU law a species of domestic law. What that means is, in principle, you apply the domestic purposive approach – text, context, purpose – to interpret some assumptions about national legislation and get an answer. However, it’s weird to apply that for the simple reason that retained EU law (now called assimilated EU law) comes from a very different legislative context, where it was meant to apply throughout the EU. Suppose you have the word ‘contract’ that’s in an EU regulation. You don’t interpret ‘contract’ narrowly. You interpret it to mean broadly speaking, contractual obligations. As with the Canadian federal Quebec example, you want the legislation to apply across different legal traditions. You can’t give those legal terms a narrow national meaning. It has to mean something broader to capture national variation (in EU parlance, this has been described as giving EU law an ‘autonomous interpretation’). There are, of course, more technical rules about interpretation and, of course, different language texts because of how EU rules are drafted.

This is one difficult area. What the courts have done, I think rightly, is to say these EU laws, which have now been domesticated, will be interpreted similarly to how the EU courts would interpret them, unless there’s a good reason not to. The courts absolutely have the jurisdiction to completely reject any CJEU (Court of Justice for the European Union) interpretations right now, and that’s in Sections 5 & 6 of the European Union (Withdrawal) Act 2018. Right now, it’s a weird balance because we’ve left the EU, but because of this legislation and the history and context behind it, it’s hard to run away from giving it nominally an EU-type interpretation and citing the EU cases. As an interpretive matter, I think that’s the more difficult issue, and the courts have to deal with it. Cases, like Lipton, continue to cite EU cases, and they do continue to interpret them based on that. We know in the back of our heads that they can absolutely ignore the EU cases. The point of the 2018 Act was to ensure that there was continuity in the law after Brexit so that prevailing legal structures do not collapse.

Let’s change gears to discuss Uber v. Aslam. In that case, the claimants are claiming they are workers, a very specific definition, and there was a purposive approach to interpreting that definition, first defined by the Supreme Court of the United Kingdom in Autoclenz v. Belcher.  The bill’s purpose itself was discussed, but interestingly, Parliament’s intent was not.  Why do you think they didn’t discuss the intent of Parliament in defining workers?

The modern practice of the UK uses intent but does not refer to it as much as the Canadian courts.  The concept of intent is still important as it connotes what Parliament wants to do and reminds us that we need to give effect to it. On the ground, as a matter of interpretation, it’s not invoked as often now at the highest level as the Canadian cases do. Most often, we say that Parliament’s intent in any given statute is its meaning. We determine meaning, as Lord Borrows instructs, by using text, context, and purpose. We consider the context and a few other interpretive things. So in the UK, intent is seen, as a matter of positive law, as this output of interpretation is the output of legislation.

The Canadian position is slightly different because we still see a lot of references to Parliament’s intent. As a matter of Canadian case law, when the courts refer to intent, they mean the rationale or reasons behind something. In the UK, we would just call it purpose. Randal Graham has this article that says, look, intent means a lot of different things, and one big one is actually statutory purpose. The language of intent also has the ability to confuse. That was the point that was made by Lord Borrows recently. My sense right now is that intent is so important, but there is less emphasis on it because they know that it could confuse the arguments. So that’s why they don’t really use it. We just use purpose to refer to a rationale behind revision or statute rather than using intent. The overall idea is still there.

Statutory interpretation has seen a flurry of changes in recent years, from the revocation of the Chevron doctrine in the United States to recent Commonwealth cases that have affirmed the long-standing but somewhat unexplored principle that statutes are always speaking. What do you see as the future issues of statutory interpretation?

I’m not sure I can predict what the future issues would be.

Generally speaking, most Commonwealth courts have said that the purposive approach is the correct approach to interpreting legislation, despite different labels. The Australians, for instance, call it the contextual approach. It’s the same thing in that when you interpret legislation, you consider text, context, and purpose. You balance them, try to find out as much as you can, and then come to an informed conclusion. Now, there are issues at the edges which do differ. For example, Australia and Canada have a written constitution. Doctrines related to their interaction with statutes will almost be unheard of in the UK, where there’s no written constitution. But these relate to the operation of the purposive approach in a more specific context, rather than an issue with the purposive approach per se.

 There are other specific issues, like the always speaking rule, but this issue has been discussed for a long time in the case law.  There is a debate as to the extent of this principle is or what its conceptual grounding is. But as a matter of positive law, we’ve accepted that it’s a recognized principle for interpretation, including as an element of Canada’s Interpretation Act. In Canadian law at least, it’s not controversial.

With respect to Chevron, there are arguments as to whether or not you defer to the interpretation of an administrative agency. That’s settled in Canada byVavilov – as long as it’s reasonable, it’s fine. On the UK side, we don’t defer to administrative decision makers.

For future issues, I’m not entirely sure. I know that there are still unresolved issues, and one of them has to do with the judicial approach to disabling tax avoidance schemes, where the scheme is set up in a way that purportedly answers to the statutory description.

The issue with tax and a lot of tax structures is that there is some corporate dealing, and it’s done through a series of contracts. By necessity, because you’re dealing with complicated corporate dealings and trying to find out whether or not those things fall under statute, you have to default to regular contracts to see what the structure is that is being created and what they do. 

The difficult question is that right now, the UK courts say we have to look at the facts realistically. What does that mean? Does it mean that we ignore all the contract structures because we see that you’re trying to pay less tax? That can’t be right.  That logic has been applied to really aggressive tax avoidance cases, and you can absolutely see why that logic works. Consider if you have taxpayers trying to avoid tax by just funnelling money around the recipients of the contract without any economic losses, or the money effectively comes back to them, but it’s not reflected in your contracts. It’s absolutely fine to say, realistically, you haven’t made a loss, or you haven’t disposed of things that are required under the statute.

That is more like this specific issue within interpretation that still needs to be sorted out for tax reasons. It is actually broader than that because now the courts have said this is a general principle applicable to all cases. Again, that’s a really specific issue. I can’t think of one that’s a big one. You might say, ‘How about legislative intent?’ Of course, intent is important (and there are live theoretical questions on the nature of legislative intent), but on the ground, it’s not that big of an issue, as we’re trying to find out what the intent is rather than using intent as an a priori factor. Frankly, I’m not sure. ■

Interview by Nicole Buzzelli and Jake Collie. Legal and academic sources referred to in this interview:

LEGISLATION

European Communities Act 1972, UK 1972, c 68, online: <https://www.legislation.gov.uk/ukpga/1972/68/enacted>.

European Union (Withdrawal) Act 2018, UK 2018, c 16, online: <https://www.legislation.gov.uk/ukpga/2018/16/contents>.

Government of Wales Act 2006, UK 2006, c 32, online: <https://www.legislation.gov.uk/ukpga/2006/32/contents>.

Interpretation Act, RSC 1985, c I-21, s 10, online: <https://canlii.ca/t/7vhg#sec10>. 

Private International Law (Miscellaneous Provisions) Act 1995, UK 1995, c 42, online: <https://www.legislation.gov.uk/ukpga/1995/42/contents>.

Retained EU Law (Revocation and Reform) Act 2023, UK 2023, c 28, online: <https://www.legislation.gov.uk/ukpga/2023/28/contents>.

CASES

Autoclenz Limited (Appellant) v Belcher and others (Respondents), [2011] UKSC 41 [Autoclenz v. Belcher].

Canada (Minister of Citizenship and Immigration) v Vavilov, 2019 SCC 65 [Vavilov].

Chevron USA, Inc v NRDC, 467 US 837, 104 S.Ct. 2778 [Chevron].

Lipton & Anor v BA Cityflyer Ltd, [2024] UKSC 24.

Rossendale Borough Council v Hurstwood Properties (A) Ltd and others, [2021] UKSC 16.

R (Miller) v Secretary of State for Exiting the European Union, [2017] UKSC 5.

R v Daoust, 2004 SCC 6.

Uber BV and others (Appellants) v Aslam and others (Respondents), [2021] UKSC 5 [Uber v. Aslam].

UBS AG v HMRC, [2016] UKSC 13.

SECONDARY MATERIALS

Andy Yu, “What Does it Mean to Say that Statutes are Always Speaking?”  (28 May 2025), online (blog): Legislative Intent and Ordinary Meaning  <https://lawandlanguage.ca/2025/05/28/what-does-it-mean-to-say-that-statutes-are-always-speaking/>.

Iliana Auverana, “Jurilinguistic Management in Canada”, Language Update, Qu3:3 (2006), online: Government of Canada <https://www.noslangues-ourlanguages.gc.ca/fr/favourite-articles/jurilinguistic-management-in-canada>.

Lord Burrows, “Some Issues on Statutory Interpretation” (Address delivered at the Statute Law Society Conference, Portcullis House (London, UK), 6 June 2025),  online: Supreme Court of the United Kingdom <https://supremecourt.uk/speeches/lord-burrows-060625>.

Randal Graham, “Good Intentions” (2000) 12 S.C.L.R. (2d) 145.