Who Speaks for Parliament?: Hansard, the Courts and Legislative Intent

Article 4 / 11 , Vol 40 No. 1 (Spring)

Who Speaks for Parliament?: Hansard, the Courts and Legislative Intent

Two significant Supreme Court rulings from the 1990s have opened the door to using Hansard Debates to divine a parliament’s intent in court cases which challenge understandings of laws. Although the Supreme Court rulings stressed that use of Hansard as a source in legal proceedings should be strictly limited, subsequent lower courts have not always observed these limits. In this article, the author outlines these developments and explains how the more liberal use of Hansard in courts can be problematic. He concludes by cautioning parliamentarians to be mindful of how the words they use during debate may be used by the courts in the future, and urges the courts to consider how some parliamentarians might begin using their speeches in parliament to win in court what they could not in a legislature.

When we’re speaking in our assembly, we have to imagine who the audience will be: constituents, activists, lobbyists, researchers, eventually perhaps historians.

There is one audience that probably does not get enough attention from members, and it should: the courts. Our courts may look at Hansard, sometimes many years after the words were spoken, when they are trying to understand the purpose and meaning of legislation. One legislative speech, even one sentence in a speech, can have far-reaching consequences.

There was a time when the courts would not even look at Hansard, but that rule has been relaxed in recent years. The modern principle laid down in a pair of Supreme Court of Canada decisions is that Hansard can be used in court, but should not be given much weight.

Despite this cautionary rule, my study of recent court cases in Nova Scotia shows that the courts refer to Hansard much more regularly than one would expect.

In this article, I’ll sketch the legal rules about how Hansard is used, why the courts should be more cautious, and the implications for members.

The Legal Rule

On the use of Hansard in courts, there are two key Supreme Court of Canada decisions: R. v. Morgentaler1 in 1993, and Re Rizzo & Rizzo Shoes2 in 1998.

Dr. Henry Morgentaler was charged under the Nova Scotia Medical Services Act with performing abortions outside a hospital. He challenged the constitutionality of the law, arguing it was a criminal law, and therefore outside the province’s authority.

The Supreme Court of Canada agreed with Morgentaler. In reaching its conclusion, the court considered (among many other considerations) whether Hansard evidence is admissible. The court traced the early rejection of Hansard evidence, and the more recent relaxation of that rule:

The former exclusionary rule regarding evidence of legislative history has gradually been relaxed (Reference re Upper Churchill Water Rights Reversion Act, [1984] 1 S.C.R. 297, at pp. 317-19), but until recently the courts have balked at admitting evidence of legislative debates and speeches. Such evidence was described by Dickson J. in Reference re Residential Tenancies Act, 1979, supra, at p. 721 as “inadmissible as having little evidential weight”, and was excluded in Reference re Upper Churchill Water Rights Reversion Act, supra, at p. 319, and Attorney General of Canada v. Reader’s Digest Association (Canada) Ltd., [1961] S.C.R. 775. The main criticism of such evidence has been that it cannot represent the “intent” of the legislature, an incorporeal body, but that is equally true of other forms of legislative history. Provided that the court remains mindful of the limited reliability and weight of Hansard evidence, it should be admitted as relevant to both the background and the purpose of legislation. (Emphasis added.)

The last, underlined sentence is the one most commonly cited with respect to Hansard evidence.

As a result, the court in Morgentaler considered a ministerial statement by the health minister; remarks by the health minister in the budget debate; and second-reading speeches by the health minister, the opposition health critic, and an opposition backbencher. All of this aided the court in deciding whether the impugned law was indeed a criminal law.

The other leading Canadian case on the judicial use of Hansard is Rizzo.

At the heart of Rizzo was a question of statutory interpretation. When a company went bankrupt, did the Ontario Employment Standards Act apply so as to entitle employees to termination pay, vacation pay and severance?

The court concluded that it did. Justice Iacobucci found support for his interpretation in two statements made in the Ontario legislature by the labour minister, Dr. Robert Elgie. He also makes a brief aside, citing Morgentaler, about whether Hansard is admissible at all:

Although the frailties of Hansard evidence are many, this Court has recognized that it can play a limited role in the interpretation of legislation.

The significance of Rizzo is that it takes the idea laid down in Morgentaler, and expands it beyond constitutional cases. The rule now applies to every question of statutory interpretation.

In Rizzo itself, Justice Iacobucci was very restrained in his use of Hansard. There are three noteworthy elements to his approach:

  • Only the bill’s sponsoring minister is quoted.
  • The quotations are brief.
  • The quotations support an interpretation reached by other means.
  • This approach is a model for other courts.
Why the Courts Should Be Cautious

In Rizzo, Justice Iacobucci for a unanimous court noted that “the frailties of Hansard evidence are many” but did not enumerate those frailties. I will list a few that occur to me, based on my 12 years in an assembly.

First, Hansard is a good record, but it is not perfect. I believe that the majority of Hansard does faithfully capture what was said. But very occasionally, I would glance back at what I was reported to have said, and be dismayed at the errors. In Nova Scotia, there is no formal procedure for correcting errors.

Second, Hansard may not capture the sense of what is being said. Like any transcript, the words on the page may be literally accurate, yet miss what the speaker was conveying. Humour, sarcasm, emphasis, tone, body language, gestures, and reactions from the audience are essential to the speaker’s meaning, but they are absent from a transcript.

Third, punctuation and paragraphing can change the meaning of a sentence. The Hansard staff are transcribing oral speeches. They have to guess where the speaker would put a colon, a dash, or a paragraph break. Unless the speaker is explicit, it may not be evident that the speaker is quoting from something or someone else, or where the quotation begins and ends.

Fourth, the fundamental purpose of speeches in the House is partisan. Of course there are exceptions, but there is very rarely meaningful debate in the sense of persuading other members of one’s position. Members’ minds are virtually always made up when they walk through the door. Speeches are intended to characterize the content of a bill for political purposes, not to win anyone over.

But the most fundamental frailty of Hansard evidence is the one alluded to by Justice Sopinka in Morgentaler, quoted earlier in this paper: “The main criticism of such evidence has been that it cannot represent the ‘intent’ of the legislature, an incorporeal body.” An elected assembly is a multi-member body. It is a concept. It cannot have an intention, any more than a neighbourhood or a sporting club can have an intention.

Nevertheless, the search for legislative intent is at the heart of statutory interpretation. Recall Driedger’s modern rule of statutory interpretation, cited approvingly in Rizzo and over one thousand other judicial decisions:

Today there is only one principle or approach, namely, the words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament. (Emphasis added.)

Every province, and the federal jurisdiction, has an Interpretation Act which also encourages a search for the assembly’s “intent” or “objects”.3

The concept of legislative intent makes the most sense when it is taken as a metaphor. The courts imagine the elected assembly as a single person, who is presumed to be knowledgeable about the law; knowledgeable about the subject-matter of the bill; logical, concise, and reasonable. That metaphorical legislator is good at their job and knows what they’re doing.

All of the problems associated with the use of Hansard evidence arise when the courts take the metaphor too far; that is to say, when they take it literally, and start searching for legislative intent in the words of real flesh-and-blood individuals.

After all, who speaks for the assembly? Nobody. That is why the courts will always struggle with their use of Hansard evidence. In most cases, the courts slide over the absence of a spokesperson with two leaps of logic: they take the intention of the minister and call it the intention of the government; and then they take the intention of the government and call it the intention of Parliament. But they are not the same thing.

Of all the members in the assembly, the sponsoring minister is in the best position to speak knowledgeably to the substance of a bill. The minister’s speech on second reading is typically the fullest statement in the House about the purpose of the bill, and about any noteworthy details of policy or drafting. The minister is the spokesperson for the government on that bill. By implication, the minister’s intention is shared by all members on the government benches who will vote for the bill.

Although the minister’s speeches are most likely to be substantive, we still need to be cautious. The intentions of the other members can be all over the map. Some may have a different understanding of the bill. Others may have no understanding at all, and merely wish to vote with their party.

Moreover, political speech is a different beast altogether than sworn testimony in a courtroom. Even sponsoring ministers can have all kinds of motivations for saying what they say. Maybe a deal has been done. Maybe the government is deliberately using ambiguity to win support for legislation. Maybe the sponsoring minister doesn’t really believe that the bill does what he says. Maybe the minister doesn’t understand the bill at all. These and a thousand other scenarios make Hansard a slippery foundation for subsequent judicial decisions.

The Courts Aren’t Cautious Enough

The legal rule is clear enough. The frailties of Hansard are many. The reasons for caution are abundant.

Nevertheless, I had an inkling that the courts looked at Hansard evidence with rather more abandon than the cautionary rule in Morgentaler and Rizzo rule would suggest. To test that inkling, I studied all citations of Hansard in Nova Scotia’s courts for the period 2004-2014. The same sort of study could be done in any other jurisdiction.

My findings were that Nova Scotia’s courts are straying well beyond the restrained use of Hansard evidence suggested by Morgentaler and Rizzo.4

In fact, in most of the cases where Hansard is cited, there is no reference to Morgentaler or Rizzo at all. Perhaps this is the root of some of the difficulties. If one does not remind oneself of the cautionary rule, there may be a tendency to admit Hansard evidence too readily and weight it too heavily. I’ll illustrate that point with two specific Court of Appeal decisions.

In R. v. Carvery,5 the court appeared to open the door to a wide range of legislative evidence.

Justice Beveridge, writing for a unanimous court, was considering whether 2009 amendments to the Criminal Code, known as the Truth in Sentencing Act, justified a quasi-automatic 1.5:1 credit for pre-sentence custody. To find the intent of Parliament, Justice Beveridge turned to the grammatical and ordinary use of the words; the scheme of the Act; and the object of the legislation. It is in this last category that he came to legislative history, and Hansard.

“Legislative history of an enactment consists of everything that relates to the conception, preparation and passage of the legislation,” wrote Justice Beveridge. This is, on its face, remarkably broad. We have gone well beyond a minister’s second-reading speech on the bill. Everything done or said, at any stage of the proceedings, is potentially relevant.

Ironically, such a broadly-stated principle was unnecessary for Justice Beveridge’s decision. The Hansard material he actually used was quite limited. When the case went to the Supreme Court of Canada on appeal, that court used only a single quotation from the sponsoring minister.6 It was, in other words, a model use of Hansard.

A second case illustrates the difficulties that are created when the court ranges widely through Hansard.

At issue in Hartling v. Nova Scotia (Attorney General)7 was the legality of limits imposed on general damages for a “minor injury” suffered in motor vehicle collisions. Three plaintiffs challenged the constitutionality of the law, and they also challenged whether the “minor injury” regulations were authorized by the legislation.

The Chief Justice, writing for the court, uses three quotations from Hansard. The first quotation is from the sponsoring minister’s second-reading speech. This comfortably fits within the model use of Hansard. But the next two quotations are from the third-party Liberals: a second-reading speech and a third-reading speech.

How can an opposition member speak to the intention of a bill that is drafted and introduced by someone else? At the time, the Liberals held only 12 seats in a 52-member assembly. Even if the statements made by Liberal MLAs can be taken as expressing the intent of all 12 Liberals – something that could be problematic in itself – how can they be taken as expressing the intent of the government, or of the legislature?

There are plausible answers to those questions, but Chief Justice MacDonald does not address them explicitly. One must read between the lines. The Chief Justice does mention twice that there was a minority government, but he does not spell out why that is significant.

Because I was there, I know why it is significant. In order for Bill 1 to pass, the government had to attract the support of one of the two opposition parties. The NDP said it would not support Bill 1. That left only the Liberals as a potential partner, but the Liberals were not happy with the original definition of “minor injury.” They believed it to be too broad. Negotiations ensued between second and third readings. Bill 1 was subsequently amended. The Liberals voted for the bill as amended. Thus the Liberal MLA’s speech on third reading was significant because it expressed the Liberals’ view of what the amendments were intended to achieve.

The difficulty is that the Liberal MLA’s speech – indeed, any speech recorded in Hansard – is a political speech, not sworn evidence. A political speech can have multiple motivations, including making one’s party look as good as possible, or making the best of a bad situation. Telling the truth, the whole truth, and nothing but the truth is not necessarily one of a politician’s motivations when speaking to a controversial bill.

To put it gently, there are alternative readings of the facts just as plausible as the Hansard evidence accepted at face value by the Chief Justice.

Implications for Members

Members need to be aware that their words may end up being dissected in a courtroom. Their words in Hansard are admissible in the courts’ search for legislative intention. That is especially true of the sponsoring minister.

The sponsoring minister should therefore be ever mindful of the potential impact of his or her words, especially on second reading. The minister’s words will be taken as stating the intention of the government, and if the bill passes, as stating the intention of the legislature. That is a weighty responsibility. The second-reading speech should be drafted accordingly.

My own experience, however, is that ministers are rarely thinking about the judicial implications of their second-reading speeches. Ministers are more commonly thinking of their political audiences. Indeed, there may be a political imperative to keep key points quiet or fuzzy. That political imperative may run counter to the courts’ need for a detailed explanation of the minister’s intent.

My experience is also that not all second-reading speeches are delivered with care. I have seen second-reading speeches delivered off the cuff. I have seen them be very short. I have seen ministers veer sharply off script. Not all ministers will welcome being told to read a carefully-crafted, lawyerish speech.

When ministers do deliver a prepared second-reading speech, it is typically prepared by expert civil service staff. The civil servants likely know the subject matter, but – and again this is based on my own experience – they are not necessarily aware of how their words may be used in a courtroom.

My study of Nova Scotia court decisions over a 10-year period shows that, despite the cautionary rule in Morgentaler and Rizzo, the words of members other than the sponsoring minister may also be cited. The danger is obvious: members are political people delivering political speeches in a political forum. If they know that their words may have an impact on how the legislation is interpreted, they may well shape their speeches to try to win in court what they could not win in the legislature.


1 [1993] 3 SCR 463, 1993 CanLII 74 (SCC), hereafter Morgentaler.

2 [1998] 1 SCR 27, 1998 CanLII 837 (SCC), hereafter Rizzo.

3 For example, Interpretation Act, RSO 1990, c I.11, s 10; Interpretation Act, RSNS 1989, c 235, s 9(5); Interpretation Act, RSC 1985, c I-21, s 12.

4 For more detail, see Graham Steele, “’The Frailties of Hansard Evidence are Many’: The Use of House of Assembly Debates in Nova Scotia Courts, 2004-2014” (2015) Journal of Political and Parliamentary Law, p. 499-518.

5 2012 NSCA 107 (CanLII).

6 2014 SCC 27 (CanLII), with the substantive reasons given in a companion case delivered the same day, R. v. Summers, 2014 SCC 26 (CanLII).

7 2009 NSCA 130 (CanLII).