1. Two Short Clarifications

Before I get to Solum’s argument for the fixation thesis and his reservations against the possibility that certain parts of constitutions might have an evolving meaning, I should explain two important distinctions that have some bearing on the topic of this paper.

First, many originalists have come to distinguish between two activities that both fall under what common-language users might mean when they use the term “interpretation” loosely: there is interpretation-strictly-so-called (from now on: interpretation) on the one hand, and construction on the other. Different theorists word the distinction slightly differently, but in general it comes down to this: Interpretation aims at discerning the meaning the text already has, independently of its interpreters. Construction aims at determining what the text means with respect to specific cases.⁸ Construction might well involve normative reasoning on the side of the interpreter. In other words, when an interpreter enters the area of construction, she might use her own judgement of what would be better for the text to mean with respect to a certain case.⁹

Construction becomes necessary when interpretation does not deliver a meaning that is clear enough to conclusively determine what should be done in a certain case. If the constitutional meaning retrieved through interpretation is vague, for example, then judges will have to engage in constitutional construction in order to come to a decision. While interpretation is an empirical inquiry into what the text already means when interpretation begins, construction has normative elements. Many originalists understand themselves to be offering a theory of how the constitutional text should be interpreted.¹⁰ That is: they offer a theory of what it means to retrieve the meaning the text already has at the time of the interpretation, independent of the interpreter. The fixation thesis, too, is a thesis about the meaning that is retrieved through interpretation, the meaning the text has independent of its interpreter. Clearly, the meaning that a text is given through construction might change. But, according to the fixation thesis, the text has only one fixed meaning independently of its interpreters. I, too, will make a claim about interpretation, not construction. I will claim that in addition to the fixed meaning, some parts of the constitutional text also have an evolving meaning that can be determined through interpretation, not construction.

The second distinction I should introduce is one that Solum makes in order to clarify what his fixation thesis is really about: Take the US-American constitution for an example. Solum is aware that we may imagine that the same constitutional text, same words, sentences, paragraphs, was enacted for a second time in 2016, by different people and in a different political climate. This text might then have a different meaning than the text as it was 1787, in the same way as it means something different whether a priest uses the words “you should act according to human nature” or whether a behavioural biologist uses the same combination of words: “you should act according to human nature”. The two constitutional texts would together belong to the same expression-type, sharing the same words, sentences and paragraphs. They would, however, be two different tokens of the expression-type, one originating in 1787, and one today. Solum admits that the meaning of an expression-type can easily be different from time to time and from use to use. The two tokens of the same sentence-type, for example, might contain a word the semantic meaning of which has changed over time.¹¹ However, the fixation thesis is about expression-tokens - that is, in our example, the fixation thesis deals with the meaning of the constitution that was enacted in 1787, not with the meaning of a hypothetical constitution from 2016.¹² When I present my argument, I will therefore have to argue that certain parts of constitutions can plausibly be thought of as always having had two meanings, one of which was evolving. In other words: my point is not the one that Solum rejects - the idea that the constitution may be interpreted as an expression-type. I accept that the constitution is an expression-token, but I will claim that parts of it do and have always expressed a meaning that was changing and evolving.

2. Solum’s Argument

The positive argument Solum presents for his fixation thesis is, according to his own estimate, based on common-sense and intuition. According to Solum, there is no reason to treat the constitutional text any different from any other text. And for texts in general we usually accept his so called “generalized fixation thesis”:

The communicative content of a communication (oral or written, verbal or nonverbal) is fixed at the time the communication occurs.¹³

Solum supports his claim by referring to the way he believes we usually determine the meaning that a text communicates. He summarizes his argument as follows:

The affirmative case for the Fixation Thesis can be articulated via intuitive and commonsense observations about the nature of written communication. If we want to know what a text means and the text was not written very recently, we need to be aware of the possibility that it uses language somewhat differently than we do now. Moreover, meaning is in part a function of context—and context is time-bound. So if we want to know what a text means, we need to investigate the context in which the text was produced.¹⁴

Solum distinguishes the meaning of a text-type, and even the semantic meaning of a text according to a certain set of semantic and grammatical conventions from the meaning that a text communicates. The meaning a text communicates is determined by the text-type the text is a token of, the semantic and grammatical conventions that governed at the time the text was produced and, importantly the context in which the text was produced. According to Solum, when we come across a text or other communicative artefact,¹⁵ our ability to speak the language of that text might not be enough to determine the meaning that the text communicates. This might be so because the text is old, or because it was produced by people in other circumstances than our own, etc. Then we need to use our knowledge about the “conventional semantic meanings of the words and phrases comprised by the sentence and the grammatical relationships between these units of meaning” that were accepted when and where the text was produced.¹⁶ But while this might well get us to the semantic meaning of the text, we might still not know what meaning the text really communicates. The example I gave above is instructive again here: it means something different whether a priest tells you to “act according to human nature” or whether a behavioural biologist tells you to “act according to human nature”, even if they tell you this at around the same time, assuming the same conventions. Therefore, we also take into account the context in which the text was produced. For example, we might inquire by whom the text was produced, in what manner, for what purpose etc.¹⁷

Solum points out that both the conventional semantic meanings and the context of the production of a text are fixed when the text is produced.¹⁸ Together, he believes, they fix the meanings of communicative artefacts.¹⁹ If I read a letter from the 13^th century, and I want to figure out what meaning the letter communicates, then I inform myself about the linguistic conventions of the 13^th century. But this only gives me the semantic meaning of the text. I also need to know about the context in which the text was produced, for example about the person who wrote the letter and the circumstances under which it was written. I expect that, if I use this information correctly, I will determine a meaning for the letter that is the same as when the letter was first written. This meaning is the communicative content of the letter.

It is important here to say a word about Solum’s use of the term “communicative content”. As Solum specifies, the term “communicative content” is supposed to be neutral on theories where that content comes from. Importantly, Solum does not limit what the text communicates to what the author of the text intents to communicate. If the context of the production of the text —the constitutional text in this instance— shows that the text is supposed to be read as the specification that cruel and unusual punishment is unacceptable as it was understood by the general population at the time of the text’s production, then the text communicates the meaning of cruel and unusual punishment as it was understood by the general population when the text was enacted. What the author intents to communicate and what the text communicates are therefore not necessarily the same. Rather, what the text communicates is determined by the semantic content of the text according to the conventions at the time of production and the context in which the text was produced. This context could point to something else than the author’s intentions, like the general public meaning at the time of the production of a text. Solum uses the term “communicative content” instead of the term “meaning” to show that he does not mean to say something about the semantic meaning of a text that it has in virtue just of the words it comprises of and its grammatical structure. Instead he means to say something about the meaning of the text as also determined by the context of its production.²⁰

I have highlighted this feature of Solum’s argument because it figures importantly in my own argument. I will not argue against Solum’s claim that, when we read a text, we use our knowledge about common semantic meanings and grammatical relationships as well as our knowledge about the context of the production of the text in order to determine its meaning. Neither will I reject the idea that the common use and the context at time of production do not change. Indeed, I will follow Solum in his suggestion that in addition to the common semantic meanings, the context of a text’s production guides an interpretation aimed at determining what the text means (what it communicates). However, I will use this very idea about the importance of context to make my point about the possibility that a text might have a changing meaning - or that what a text communicates might change. I will claim that it is the context of their production that shows us that they do. (When I talk about the meaning of a text, I will from now on refer to Solum’s “communicative content”, as determined by both semantic meaning and context).

However, before I can get to that, I would like to ask the reader for a moment’s patience. Solum believes that the acceptance of the fixation thesis will mainly have the effect to redirect the discussion to the constraint principle. He claims that both originalists and non-originalists might think that the attention he gives the fixation thesis is unwarranted because its truth is so easily grasped that it is not interesting.²¹ I do not agree. I believe that a discussion of the constraint principle based on the acceptance of the fixation thesis would put non-originalists at an argumentative disadvantage. Let me shortly explain why.

3. Interpretation and Being Afraid of Judges

Some originalists argue that allowing judges to aim at anything else but the original meaning of the constitution when they do what they call “interpretation” is allowing them to make law behind the mask of applying it.²² For example, Larry Alexander, who argues for an author’s-intent-originalism²³ claims that if the meaning that judges should aim for in their interpretation is not fixed as the original meaning, then judges can freely choose among possible meanings. Thereby they effectively become the authors of a new text instead of the interpreters of the old one. His argument for this, in short, goes like this: if the judge is not restricted in his interpretation of the text by having to aim at the original meaning then she can freely choose another meaning instead. For example, she might want to give the text the meaning that will make it the best possible constitution, trying her hand at a version of Dworkinian interpretation.²⁴ Or she might decide to give the text the meaning it would have simply according to the meanings the words have today. Because there is no way to determine which of these possible meanings is the most appropriate one, the choice is up to the judge. But if the judge chooses which of the many possible meanings she should give the text, then it is the judge who chooses the text’s meaning. This is so especially if the judge decides to “interpret” the text so that it has the best possible meaning - after all, the judge has nothing but her own moral and political commitments to guide her in determining what this best meaning is. As a result, the judge’s act of interpretation becomes an act of authorship. The judge chooses what the text is supposed to mean, according to her own ideas of what it should mean. In Solum’s terms, she produces a new token constitution of the same expression-type as the old constitution, under the guise of determining the meaning of the old constitution.

This, however, means that the judge becomes a law-maker instead of someone who applies the law. In interpreting the constitution, she no longer determines what she has to do in a specific case according to already enacted law. Instead, she takes the occasion of her case to make new law and apply it in one single decision.

The argument that judges become lawmakers if they determine what the constitutional text means unconstrained by its original meaning draws its strength from the fixation thesis. According to the fixation thesis, the constitutional text only has one meaning independently of its interpreters, and that is the original meaning. Attributing any other meaning to it means giving it a new one, and thereby turning it into a different text. Once the fixation thesis is in place, this argument is not only plausible, but it also makes a strong point in favour of the constraint principle. After all, it is a widely accepted common place that judges are put into office to apply law, not to make it. An argument that reveals that all but one kind of interpretation amount to nothing else but judicial lawmaking is a strong argument for the one kind of interpretation that is left. Once the fixation thesis is in place, non-originalist legal theorists therefore find themselves arguing uphill, having to establish one of two difficult points: Either they have to show that interpretation aimed at a meaning the text did not already have when the interpretation began is not the same as putting a meaning into the text the interpreter would like it to have. Or they have to show that judicial lawmaking is, to some extent, desirable. I do not think that either of these tasks are impossible to accomplish. But they do put the non-originalist in a difficult starting position.

1. Of Recipes, Letters, and Folktales

It seems almost a fashion among originalists to make points about constitutional interpretation by referring to other kinds of texts: Solum talks about a 13th century letter and his own lectures,²⁸ Alexander speaks of instructions for assembling toys made in China,²⁹ Lawson of an old recipe,³⁰ Goldsworthy of Shakespeare,³¹ Fish of Finnegans Wake.³² It should therefore not be surprising that I add my own example to the long line of examples, this time one that does not fit with the fixation thesis as easily as the examples chosen by the writers above: folk-tales. What is so different about folk-tales? Different is that if you pick up a book of collected fairy-tales, for example, and you open it to find Cinderella or Little Red Riding Hood, the content of the text you see has not been shaped (or shaped completely) by a determinable group of authors, or at a determinable time, or at a determinable place, or in a determinable political setting. An important part of the context in which Solum’s 13^th century letter was produced was the author of that letter, what she wanted to say, what she assumed to be true, etc. But folk-tales have a different context than letters. The context the text of folk-tales was written in involves that it is a version of a folk-tale. And this context makes a big difference. Allow me to elaborate.

The realm of fairy-tales is divided into two different groups. One kind of fairy-tale, like for example The Little Mermaid, is produced in the same way as the examples originalists favour. The Little Mermaid was conceived and written by Hans Christian Andersen and published in 1837. It is a so-called literary fairy-tale. Andersen’s story has been retold many times,³³ but all the other versions of the Little Mermaid are adaptations from one, central original - Andersen’s story. In writing it, Andersen copied the style of the other kind of fairy tale, the folk-tale.

When you read a folk-tale, like Cinderella or Little Red Riding Hood, then you are always just reading a version. There is no original, like there is with the Little Mermaid. Cinderella, for example, is the main character in hundreds and hundreds of versions of what folklorists call a tale type.³⁴ In these versions, Cinderella changes her character, her ethnic origin, her behaviour, even her ultimate fate.³⁵ Every time someone tells her story, a new version appears.³⁶ The versions are held together through certain reoccurring elements, but no one knows, though considerable effort has been made by folklorists to find out, where Cinderella comes from, when she first appeared, and how exactly her character developed.³⁷ There simply is no original story of Cinderella. Cinderella is not the product of a certain time, or of any author’s imagination. She is, in some sense, the product of the imagination of thousands and thousands of interpreters over literally thousands of years.³⁸ For the most part, these interpreters are not at all aware that they are creating new versions. Instead, they simply re-tell a story they know, inadvertently making changes here and there. These changes accumulate over time until an easily distinguishable new version has appeared.

When an interpreter tries to understand what the Little Mermaid means, the fact that it was written by Hans Christian Andersen is a very important, maybe the most important part of the context that will guide the interpretation. We may assume that Hans Christian Andersen tried to communicate something with his story and chose for this purpose its plot, its characters, the way he would tell it. What he tried to communicate is important for what the story means. However, someone who decides to sit down and document one of Cinderella’s versions does not thereby become the author of her tale in this way. Plot and characters are not of her choosing. The context of the production of this text —unlike others— does not inform us only of the character and situation of its writer. Rather, its context also informs us that the text is a version of a tale-type, and that the tale-type is the product of thousands of re-tellers, living in dozens of historical and political situations over hundreds of years. The result is that the context of every folk-tale-token allows for the version to be read in two different ways, from the very moment it was written down on. First, it can be read as if it was a self-contained product, an original like Andersen’s Little Mermaid. It is possible to ask what the situation of the person writing the tale-version down was, what she likely wanted to say, what she used it for. But it is also possible to ask for the meaning of the tale as a whole, independent of the version - to ask for the meaning of Cinderella in general (or, alternatively, within a certain culture).³⁹ Indeed, the famous folklorist Alen Dundes points out that, after all, Cinderella’s versions are versions of the same story type, found all over the world, retold in Africa and Asia, Europe and America by people of the most different belief systems, religions, experiences etc.⁴⁰ Behind the many meanings, Dundes identifies common, much broader meanings and themes. For example, in his interpretation of the many versions that together constitute Little Red Riding Hood, he demonstrates that the tale is a specific way of dealing with sexuality.⁴¹ In addition to their own, specific meanings, versions of a folk-tale have an underlying, vague meaning, common to all of them. Folklorists have pointed out that this meaning cannot be accessed through the reading of one version alone.⁴² Each version is a mere example of its tale-type. The common meaning remains almost hidden if only one version is considered and can be identified easier the more versions are brought together.

There are then two plausible ways to approach the meaning of any folk-tale version. This is so because the context in which this text was produced contains both, that it was written down by a specific writer, at a specific time; but also that it is a mere version and that it does not wholly belong to this writer. Therefore, on the one hand, you can read it as a product of the time when it was put on paper. You can ask what the version’s re-teller intended to say with it, in the political and social circumstance in which it was written down. This allows you to determine a meaning for it that is fixed. But you can also plausibly read it in its role as a version, a tale-token that is so intimately connected to its type that it is a mere example of it.⁴³ This is a role it necessarily has from the beginning, because it is a folk-tale version. Then you, like Dundes are looking for the meaning of the tale-type it is a version of. There is a sense in which this meaning is not fixed. What sense is this?

The tale-type exists as a function of all its versions, and so its meaning is dependent on the versions that exist and come into existence. Every time a new version appears and an old one is forgotten forever, this meaning might change. The meaning of the tale-type evolves with every new version. Because the context in which every tale-version was produced includes that it is a version of a tale-type, this context warrants that the version can be read as an example of the tale type, carrying, from the very beginning, its evolving meaning. When it comes to the tale-type, the context does fix something, namely that an interpreter reading the tale-version as an example has to find its meaning by aiming at identifying the meaning of the tale-type. But that meaning changes, and so, in an important sense, the meaning of the tale-version is unfixed, changes with the changing tale-type.

To summarize: The context in which a tale-version was produced does guide the interpretation of the version, but it leaves two different directions for this interpretation open: One direction is to find the version’s meaning as an expression of the person who wrote it down, then the meaning is always the same, no matter when the interpretation takes place. The other direction is to find the version’s meaning as an example of a tale-type. The context fixes something here because in each case, the interpreter looks for the same thing, what the tale-type means. But what the tale-type means changes with time, so the meaning the interpreter will identify will not be the same independent of when she undertakes the interpretation. In this sense, this meaning is not fixed.

1. Constitutional Borrowing or Migration ⁴⁷

Constitutional borrowing⁴⁸ occurs when at least part of the constitutional text is a copy or a more or less slightly altered version of (constitutional) law found elsewhere. The migrated or borrowed part might be a whole passage, a sentence, or an idea or structure.⁴⁹ Borrowing and migration happen consciously or unconsciously, voluntarily or involuntarily.⁵⁰ Whenever it occurs, the passage of the text that is borrowed is a version of another text rather than an original. For example, there are a number of rights that are virtually always included whenever a new Bill of Rights is adopted. These rights seem to be versions of the Human Rights we find in international treatises like, for example, the ICCPR (International Covenant on Civil and Political Rights).⁵¹ An example of a right versions of which appear virtually everywhere is Section 12 of the Canadian Charter of Rights and Freedoms, “Everyone has the right not to be subjected to any cruel and unusual treatment or punishment”.⁵² It does not seem too far fetched to say that this right is a version of a right-type, given that other versions of the same right can be found for example in the UDHR,⁵³ the American Bill of Rights⁵⁴ and the English Bill of Rights of 1689.⁵⁵ Just like we have a general idea of what Cinderella is, independent of whether we think of a certain version of this specific tale-type, we have a certain idea of what a right not to be subjected to cruel and unusual punishment is, and we would be able to say rather confidently whether some right-formulation we were presented with would be a version of the right not to be subjected to cruel and unusual punishment.⁵⁶ With the appearance of these versions and the way they were handled, the way this right-type is understood changed.⁵⁷ New versions will change it further. This makes Section 12 more similar to a version of Cinderella than to the Little Mermaid. It is hard to see how one could attribute the invention of the right not to be subjected to cruel and unusual punishment to the ones who wrote it into the Canadian Constitution or even the American Bill of Rights, just like it is hard to see how one could attribute the invention of Cinderella to Walt Disney.⁵⁸ Instead, it appears that they merely wrote down a version of the type, the main content of which they took from other versions. This hardly makes them authors in the sense Andersen was an author, or even in the sense you are the author of your emails.⁵⁹

I believe that we can extend the argument I made about the several meanings of folk-tales to parts of constitutions that can plausibly be claimed to be versions of constitutional types. The Canadian Section 12 can plausibly be read, like Cinderella, as a version of the right-type not to be subjected to cruel and unusual punishment. As such, the context of the production of this section guides its interpretation, but allows the choice to interpret in order to identify the meaning of the right-type.

Of course, not all passages of a constitutional text will be the result of such borrowing. Additionally, even borrowed passages are more than mere copies. They have been picked out intentionally for this constitution and their details and specific wording reflect deliberate choices. Constitutions are written with care and much work goes into every part of them, including the borrowed ones. When a version of a right-type is included in a constitution, the framers of that constitution will often use careful formulation to adapt the right to fit the needs of their country.⁶⁰ Therefore it would be wrong to say that the meaning of for example a right included in a constitution just is the meaning of the right-type. Where explicit choices to diverge from the general understanding of the right have been included in the text, this has to be taken into account. Therefore, even though passages that represent borrowed constitutional ideas remain connected to their types and their evolving meaning, the following objection seems plausible: Even if certain rights can plausibly be understood to be versions of right-types, and can therefore plausibly be thought of as having a changing meaning, this has almost no impact on judicial interpretation.

Adding a right to freedom of speech to a particular constitution might commit us to a right-type whose meaning is determined by the development of all its versions. But this meaning must necessarily stay vague to allow for the many differing and specific implementations of its versions. When judges deliberate on difficult questions about constitutional law, such vague meanings will not be of much help. It is the judge’s job to determine the correct specific meaning. This more specific meaning lies in the details and the specific wordings of the passage as well as its relationship to the other passages that were chosen for this specific constitution. It is a result of the deliberate choices made at the point of the constitutions creation, and therefore fixed. Therefore, all the important questions judges have to answer can only be tackled by referring to the fixed meanings of these passages.

How might one respond to this argument? First, we might note that the meanings that borrowed passages bring with them are not at all as inconsequential as the argument makes them seem. The heated debate over the influence of international law on national judicial decision making might serve as a piece of evidence for this: Why should the practice to consider international understandings be so hotly disputed if such consideration made no difference?⁶¹ Another piece of evidence are situations in which judges are legally obligated to interpret constitutional law so as to render it consistent with prior commitments made when signing international treatises. The meanings imported through these treatises sometimes have a lot of influence on judicial decisions in constitutional law. For example, the European Convention of Human Rights (ECHR), equipped with its own court (ECtHR), is an international treaty that is supposed to protect human rights in Europe. It gives any person whose rights have been violated by one of the participating states the right to bring her case to the ECtHR. Participating countries oblige themselves to give effect to the judgements of the ECtHR, even though it is up to them in what manner this is done. Germany, after signing the treaty, made it a policy to interpret its Basic Law (the German version of a Bill of Rights) in such a manner that it is consistent with the ECHR and the rulings of the ECtHR. For example, the ECHR, but not the Basic Law, includes an individual right to the presumption of innocence. However, a key element of the German constitutional order is the rule of law. Under the influence of the ECHR, that element is being interpreted as implying a right to the presumption of innocence as it is understood by the ECtHR. This means that the court changed its interpretation of what the rule of law implies in order to comply with the international understanding.⁶²

Second, even if rights are not borrowed, or if the borrowed rights are formulated to reflect the specific identity of a country, there is an argument to be made that the context in which some of these rights were included allows us to assume that they have an evolving meaning in addition to their fixed meaning. This argument relies on the idea that parts of constitutions are often included in order to reflect the values that are already present in the national identity of a country’s citizens.

2. Acquiescent and Militant Constitutions

Constitutions often (but not always) mirror the values and norms embedded in the national identity of the people in the constitution’s domain. Their text can plausibly be seen as not so much an original creation as the formulation of content found in the existent commitments, beliefs and values of the people.

Constitutions are usually not written in a vacuum. Instead, they are formulated as the constitutions for a certain group of people, those living in the constitution’s domain. These people have their own identity, values, religions, beliefs. Gary Jacobsohn argues that these identities have to be taken into consideration when constitutional law is made.⁶³ Constitutions are written for their subjects, in response to the people they are constitutions for. Such a response does not have to embrace the identity of the constitution’s subjects. It can just as well reject them. But it is always directed at them.

Gary Jacobsohn distinguishes between acquiescent and militant constitutions. Militant constitutions reject (parts of) the identity of their subjects. They are set out to fundamentally change the national identity of their subjects or to form a new national identity that negates, rather than embraces, its most fundamental elements. The values found in such constitutions are not yet the values their subjects actually endorse. The goal is to replace the values embedded in the national or ethnic identities with those chosen in the constitution.⁶⁴

However, we are much more used to constitutions that are geared to reflect the values and identity of those they are set out to govern. Jacobsohn uses the term “acquiescent” to describe this relationship between the constitution and the national identity of its subjects. Acquiescent constitutions (and constitutional passages) are oriented towards accommodating the national identity or identities of the constitution’s subjects.⁶⁵ The values included in the constitution are supposed to express the values found in the national or ethnic identity of the peoples.

A typical example for an acquiescent passage is section 27 in the Canadian Charter of Rights and Freedoms which expresses the Canadian value of multiculturalism.⁶⁶ But Canada’s commitment to multiculturalism did not come into existence with the Charter. The Canadian government claims that Canada was the first country to adopt multiculturalism as an official policy (in 1971). But even this official act should not be viewed as bringing the value into the set of commitments that, at least in part, define the national identity of Canada. On the contrary, the value was part of the distinctly Canadian national identity well before either of these political decisions were made. It had been theorized, discussed and embraced (though not always under this name) long before legislative policies implemented it.⁶⁷ As a result, it seems implausible to suppose that those who decided to include a passage about multiculturalism in the Canadian Constitution did the same thing that Shakespeare did when he created King Lear or what I do when I write a shopping list. It makes little sense to insist that these authors were the ones who instilled Section 27 with meaning simply because they were the ones who put it in the constitution. It seems much more natural to say that they added a textual passage meant to include in the constitution a reference to a value that was —and is— part of the national identity of Canadians. The framers of the constitution made the conscious choice to include multiculturalism. But that they made this decision does not necessarily entail that their understanding of what multiculturalism is determines what this passage means.⁶⁸ Indeed, it does not even necessarily mean that the passage refers to the understanding of multiculturalism that was prevalent when the passage was included. If the text, from the very beginning, refers to the value of multiculturalism as it is embedded in the national identity of Canadians, if the constitution is supposed to implement into law the values central to this national identity, then the term “multiculturalism”, as used in the constitutional text can be understood as referring to an evolving meaning. This is so because it is used in a text composed for the purpose to integrate values that are part of Canadian national identity into law. Allow me to explain.

As we have just seen, acquiescent passages in constitutions point beyond their authors’ intentions and to the values, ideas, and beliefs that are part of the national and ethnic identities of the constitution’s subjects. To be sure, the framer’s specific choices to include these passages together with their further choices regarding the details of the specific wording employed, contribute to the meaning of the relevant passage(s). But they do not exhaust it. Insofar as the content of these acquiescent passages is taken from the people’s national or ethnic identities, it is from this source that a good deal of their meaning can be derived. However, when interpreters turn to the people’s national and ethnic identities to determine what the value is that was included in the constitutional text, they will likely find that the people’s understanding of it at one point in time is somewhat different than ten or a hundred years earlier. The understanding of the value of multiculturalism that is a part of Canadian’s national identity changes as the way Canadian people think about this value changes. The framers, by including acquiescent passages in the constitutions, cannot help but include, with them, the connections they have to the national and ethnic identities of the people they are supposed to govern. As with folk-tales, the context of production for these kinds of passages can guide the interpretation aimed at identifying the passage’s meaning, but it allows interpretation in two different directions. It is possible to interpret passages like this one in order to find the meaning the framers believed them to have when they implemented them. But it is also plausible to read these passages as ones that are supposed to give legal authority to values embedded in the national identity of the people. Again, if read like this the meaning of the passage is not fixed. Sure, the context does fix something: The interpreter must look for the meaning of the value as it is embedded in the national identity of the people governed by the constitution at any point in time at which she undertakes her interpretation. But the value itself evolves, and therefore the meaning an interpreter might identify at one point in time might be different —have evolved away from— the meaning she might have identified had she interpreted the passage at an earlier point in time.

This thought is no invention of this paper. In fact, a whole theory of constitutional interpretation has been built based on the idea that it is the task of a constitutional bill or charter of rights to reflect the values embedded in the identity of the people it governs as these values evolve. This is the theory of common-law-constitutionalism that has been proposed for example by David Strauss and especially Wil Waluchow, who supplements it with his theory of the so-called Community’s Constitutional Morality (CCM).⁶⁹

According to the theory of common-law-constitutionalism, judges should use earlier decisions as precedents that can guide their decisions. The use of earlier decisions as precedents allows judges to develop the constitutional law in a stable and flexible way.⁷⁰ Waluchow proposes that judges should refer to the community’s constitutional morality when they decide constitutional cases in order to ensure that their interpretation of constitutional law stays in touch with the values that the people the constitution governs are committed to. When it is debatable which decision needs to be reached, judges should decide according to the values and principles underlying the community’s constitutional morality. CCM does not describe an abstract, objective morality. Neither does it refer to the moral or normative opinions the people in a country might actually hold at some specific moment in time.⁷¹ Rather, it refers to those serious normative commitments the community has made through legal and political acts, such as enacting laws or entering international treatises. The content of these commitments can be determined using Rawls’ ideal of the reflective equilibrium; a state in which all normative commitments an entity has made are consistent and ideally also coherent with each other. In order to attain reflective equilibrium, inconsistencies in the normative belief-system have to be eliminated. This means letting go of conflicting normative opinions and adopting new normative commitments only if they fit with the rest of the system. When judges decide a constitutional case, they establish a normative commitment on the part of their community by officially determining the meaning of the relevant part of their constitutional law. As result, they are required to make their constitutional decisions in harmony with the rest of the commitments the community has already made by, for example, enacting laws or entering into binding international treaties.⁷²

I believe that judges who make decisions using CCM access the values embedded in the national identity of their community. At least in working democracies, laws are enacted and treatises are signed by members of the community elected to do so. These individuals are charged with the responsibility to make decisions that represent the will of the community, or at least the majority of it. Before a fashionable normative opinion becomes integrated into the legal commitments of a community, it usually has been tested in public debates, discussed in newspapers and it has earned the majority’s approval. Additionally, one enacted law is only a small voice in the big chorus of a legal system. Therefore, a judge who makes use of the legal normative commitments of a community is well equipped for integrating only those changes into constitutional law which are truly part of the national identity. By attempting to make decisions that will further the community’s attempt to achieve reflective equilibrium, the judge expresses not her own, but her community’s values.⁷³