Legislative Intelligibility and the Rule of Law

At some point in our education, we learn that we live in a nomocracy, a society governed by the rule of law. Although the phrase trips lightly off the tongue of judges and deans at call ceremonies, it is in fact a complex and enormously important concept that underpins our system of government and separates Canada from not only North Korea but from more junior democracies, like Thailand and Egypt, prone to coups d’état.

Apart from meaty ideas about fundamental justice, arbitrary decision-making and the independence and interrelationship of the branches of government, the rule of law also describes certain basic principles about the legislature and the legislation it produces:

Laws should treat those they govern equally.

Laws should be forward speaking and avoid retroactive effect.

Laws should be published and freely available to those they govern.

Laws should be clear, comprehensible and certain.

This last principle is worthy of special consideration in the age of the litigant without counsel. It seems to me that central idea here is that legislation should be intelligible on its face, or, more specifically, that the average citizen, possessed of average intellect, average education and average fluency in an official language, should be able to read and understand the text of a particular piece of legislation as well as its application to her circumstances. If I’ve got this right, then a corollary principle logically follows:

Persons governed by a law should not be required to obtain legal advice to follow the law.

After all, the average citizen should be able to access and understand the laws of the land regardless of her means or inclination toward lawyers, particularly in respect of laws carrying penal, economic or social consequences.

In my view, however, much of our present legislation and regulation fails this test spectacularly. Although I’m not particularly preoccupied with legislation governing banks and corporations, I am concerned about the rules governing civil proceedings and the law on domestic relations, personal taxation, government benefits and criminal conduct, and it seems to me that a major goal of law and justice reform over the next decade should be improving the accessibility of the legislation and regulations most affecting individuals.

The principles barriers to the intelligibility of a particular law are, in my view, are excessively involuted sentence structure, the use of language other than in its plain or ordinary sense, and recursive drafting techniques that require the reader to interpret one section by reference to another. Take, for a simple example, s. 12(1.2) of the Canada Pension Plan concerning the right of persons 65 and older to opt out of the deduction of premiums from their income:

If a person does not revoke — in respect of an employer — an election in the prescribed form and manner, the contributory salary and wages referred to in paragraphs 8(1)(a) and 9(1)(a) do not, for the purposes of those paragraphs, include income from that employment. However, they may — in respect of that income — make an election under subsection 13(3) and pay the contribution required under section 10 within one year after their balance-due day.

This subsection uses both commas and em-dashes to mark phrases and subphrases, and cannot be understood without reference to four other sections of the act. Or, take s. 56.1(2) of the federal Income Tax Act which provides direction on the correct calculation of the amount of spousal support payments to be deducted from or added to one’s taxable income:

For the purposes of section 56, this section and subsection 118(5), the amount determined by the formula

A – B



is the total of all amounts each of which is an amount (other than an amount that is otherwise a support amount) that became payable by a person in a taxation year, under an order of a competent tribunal or under a written agreement, in respect of an expense (other than an expenditure in respect of a self-contained domestic establishment in which the person resides or an expenditure for the acquisition of tangible property, or for civil law corporeal property, that is not an expenditure on account of a medical or education expense or in respect of the acquisition, improvement or maintenance of a self-contained domestic establishment in which the taxpayer described in paragraph (a) or (b) resides) incurred in the year or the preceding taxation year for the maintenance of a taxpayer, children in the taxpayer’s custody or both the taxpayer and those children, if the taxpayer is

(a) the person’s spouse or common-law partner or former spouse or common-law partner, or

(b) where the amount became payable under an order made by a competent tribunal in accordance with the laws of a province, an individual who is the parent of a child of whom the person is a legal parent,



is the amount, if any, by which

(a) the total of all amounts each of which is an amount included in the total determined for A in respect of the acquisition or improvement of a self-contained domestic establishment in which the taxpayer resides, including any payment of principal or interest in respect of a loan made or indebtedness incurred to finance, in any manner whatever, such acquisition or improvement


(b) the total of all amounts each of which is an amount equal to 1/5 of the original principal amount of a loan or indebtedness described in paragraph (a),

is, where the order or written agreement, as the case may be, provides that this subsection and subsection 60.1(2) shall apply to any amount paid or payable thereunder, deemed to be an amount payable to and receivable by the taxpayer as an allowance on a periodic basis, and the taxpayer is deemed to have discretion as to the use of that amount.

Now, to find this subsection, the taxpayer cannot search for the word “spousal,” even though that’s the word used in reference to spousal support in the Divorce Act; instead, the Income Tax Act uses “support amount” and the archaic term “maintenance.” Moreover, the taxpayer wouldn’t find s. 56.1 from reading s. 56 concerning non-employment sources of income to be included in the taxpayer’s taxable income, as s. 56.1 isn’t even mentioned in s. 56.

Assuming the taxpayer has discovered that s. 56.1 is relevant to her tax liability, the taxpayer must then figure out how s. 56.1 applies. This would require the taxpayer to:

  • correctly parse a complicated series of nested paragraphs and subparagraphs;
  • guess that “self-contained domestic establishment,” a stellar illustration of the benefits of plain language, probably means house or apartment;
  • find a case or a textbook explaining “tangible property,” “civil law corporeal property” and “custody,” none of which are defined by the act;
  • find the definition of “medical expense” at s. 118.2(2), the definition of “common-law partner” at s. 248(1) and the definition of “child” at s. 252(1)
  • determine when a “taxation year” begins and ends, as the term doesn’t seem to be explained in the act or its regulation;
  • verify the dictionary definitions of “spouse” and “allowance;” and, having figured all this out,
  • do some math.

With the greatest respect to the extraordinarily erudite and punctilious drafters at the federal Department of Justice, legislation of this ilk is neither clear nor comprehensible. (I admit to cherry-picking some delightful examples of legislative bafflegab, however the amount of Canada Pension Play benefits are critical for many seniors and errors in the calculation of income tax potentially carry serious penal and financial consequences.)

I’ve discussed some of my concerns about complexity in the context of family law matters elsewhere, however the issue of complexity is truly multidimensional and has a significant potential impact on the extent to which Canada is a nation truly governed by the rule of law, in both its formal sense and in the ability of its citizens to access justice. In my view, any reform aimed at improving access to justice will be incomplete and inadequate unless we ensure that our legislation is exoteric and likewise accessible.


  1. This is a very timely commentary.

    I’m not a lawyer and I don’t have the benefit of a university education, but I’ve always been interested in how language works: how it is used and how it is misused.

    Though I accept that a certain level of complexity is probably inevitable in legislation, I long ago concluded that the worst instances of abuse of language in legislation are deliberate. There are two in particular that I identified in actions I’ve brought in court and before the CJC, which is in receipt of another complaint submission from me dated December 18:


    I would maintain that each of these instances is very interesting and very revealing. A course I regret I was unable to take years ago was to proceed with a lawsuit against the government, calling upon the expert testimony linguists to prove that the abuse I had identified of both language and the legislative process was deliberate.

  2. I am a big fan of clear language in documentation that the public are going to be using on a regular basis. So I find myself pained in that I have to disagree with you here.

    First, accessibility can be injured by many things. One of the things that injures accessibility is cross-referencing, and the use of obscure terminology. Another thing that injures accessibility is length. if you get rid of cross references and defined terms, you necessarily make the text considerably longer.

    With regard to laws, unfortunately things should be made “as simple as possible and no simpler.”

    I’m not sure that your prescription would help accessibility.

    I also don’t think there is any suggestion that making the laws more accessible, even if possible, is going to eliminate the need for lawyers. I think the depth and breadth of the law and the principle of specialization mean that it will always be valuable to employ someone who already knows rather than learning yourself. If your suggestion is that it should be possible for the average person to understand the laws that apply to them without a lawyer, that is only helpful if they also know which ones they are. which is just to say that I don’t think the accessibility of specific laws is the main threshold.

    But, and this is the part that surprises me to be arguing, because I am a fan of plain language: I will go further and say that the laws should be made less comprehensible for people, and more comprehensible for computers.

    Am I entitled to be able to predict what the law will do in a given situation? Yes. Am I entitled to be able to predict what my word processor will do with certain inputs? Yes. It doesn’t serve my purposes of I can’t predict its reactions.

    Am I therefore entitled to be able to understand the code that runs my word processor? No. As long as I can test what I want to do, it should be fine.

    The solution to complexity in the law is not to try and reduce it, but to give it to something that is good with dealing with complexity.

    If laws were drafted in a syntax designed to allow for computational analysis of their outcomes, and if those codes were open source, and could be examined and tested by anyone, then we could build tools that allowed people to test how the law would react to given scenarios.

    Crazy, right?