Mathew Englander (mathew5000) wrote,
Mathew Englander

Kurt Gödel and Constitutional Amendment

In the Supreme Court of Canada hearings on the Senate Reform reference this week, one of the lawyers was asked how Canada could be transformed into a dictatorship, would it require unanimity of all the provinces, or just seven of them. (In the video, via CPAC, watch the interchange from time 71:00 to 76:20.)

This reminded me of the story about the logician Kurt Gödel studying the U.S. Constitution before his citizenship exam. Gödel (known for Gödel’s Incompleteness Theorem, the most important result in the philosophy of mathematics) concluded that under the Constitution, it would be legally possible to transform the U.S.A. into a dictatorship. His friends Albert Einstein and Oskar Morgenstern accompanied Gödel to the citizenship exam in 1947. As recounted in Rebecca Goldstein’s excellent biography Incompleteness: The Proof and Paradox of Kurt Gödel: turned out that the judge, whose name was Philip Forman, was the very one who had administered the oath of citizenship to Einstein some years before and he ushered the three men into his chambers immediately.
     Einstein and Forman chatted for a while and Gödel, sitting quietly and biding his time, seemed all but forgotten. Eventually, though, Forman got on with the business of the day.
     “Up to now you have held German citizenship.”
     Immediately, Gödel corrected the judicial error: Austrian citizenship.
     Duly corrected, the judge continued.
     “In any case, it was under an evil dictatorship. Fortunately, that is not possible in America.”
     This was just the opening the logician had been waiting for.
     “On the contrary,” he objected, “I know precisely how it can happen here,” and he began to launch into his account of the flawed Constitution. Forman, Morgenstern, and Einstein exchanged meaningful glances and the judge called a halt to Gödel’s exposition, with a hasty, “You needn’t go into all that,” and steered the conversation round to less dangerous subjects.
Goldstein says it is unknown what Constitutional flaw Gödel had in mind. Probably he was simply referring to Article Five, the amending process. If Congress passes a joint resolution, with a two-thirds majority in each house, proposing a constitutional amendment that would cancel all elections and install a president-for-life, and the amendment is then ratified by 38 states, it would take effect as the supreme law of the land. That could happen next week, but it’s rather unlikely. (Though for Gödel, who saw Germany “legally” become a dictatorship after Hitler was elected, maybe it wasn't purely theoretical.)

In Canada, a plain reading of Part V of the Constitution Act, 1982 would allow the elimination of federal elections, turning Canada into a dictatorship, under the 7/50 formula. (The amendment would repeal sections 3 and 4 of the Charter, and provide that members of the House of Commons be chosen by the prime minister and serve at the prime minister’s pleasure.)

The dictatorship hypothetical is extreme but still could be important in the Supreme Court’s reasoning. In questioning counsel at the hearing, some of the judges have pondered whether they should interpret the amending procedures by looking at the underlying principles of the Canadian constitution, and then deciding which amending formula applies to a certain amendment based on how much, in their opinion, the amendment would alter those underlying principles. Some of the provincial attorneys general (most of them, in fact) advocated that approach in argument. Counsel for the federal attorney general argued no, you don't need to go to the underlying principles, the words in the Constitutional amending provisions already take account of those principles, so you should determine the applicable formula based on the text alone.

The question we must confront is this: since the “plain meaning” interpretive approach would let Canada’s form of government become a dictatorship with the consent of only seven provinces rather than all ten, should we therefore reject the plain meaning approach as fundamentally flawed?

The answer is no.

We may think there should be a monotonic relationship between the magnitude of a proposed amendment and the difficulty of the applicable amending formula: that is, it appeals to our sense of order if minor amendments fall under “easier” amending formulas while the most major amendments fall under the most difficult amending formula (i.e. unanimity). But it is dangerous for courts to say, okay, this monotonicity seems to be the underlying logic of Part V, therefore we should disregard the text of Part V when we think it deviates from the monotonicity principle.

If Part V were drafted in open-ended language, then it would be justifiable for the courts to take a contextual approach to interpreting it. For example, if section 41 said “unanimity is required for amendments that alter fundamental characteristics of Canada, such as (a) the office of the Queen...” and so forth, then it would be clear that the specific matters listed in section 41 were just examples of types of amendment requiring unanimity, and it would be up to the courts to decide whether other types of amendment (not mentioned in Part V) also require unanimity.

But that is not the scheme of Part V. In section 41 five matters are listed, amendments in relation to which require unanimity. The list of five matters is not preceded with “such as”, “in particular”, or “for example”. There is just no basis for a court to add to this list matters that it believes are similar.

The drafters of Part V were aware that the number of imaginable amendments is infinite. They could have written Part V to empower the courts to choose which formula would apply to some future amendment proposal that had not been contemplated in 1982. But instead, they consciously drafted the provisions of Part V such that the unanimity formula would be limited (to the five matters set out) but the 7/50 formula would be the “general procedure”, covering all types of amendment not otherwise provided for — including amendment proposals that nobody contemplated in 1982, no matter how radical.

Some of the provincial attorneys general (including Ontario’s AG) cautioned against “formalism” in interpreting the amending provisions. But there is good reason for courts to be formalistic when looking at the amending provisions as compared to other parts of the Constitution. The reason is, constitutional amendment is the last resort for elected officials to overturn decisions of appointed judges. Considering the democratic rights guaranteed in the Charter, what if a future court interprets those provisions in a way so unpopular and unworkable that the public feels a constitutional amendment is warranted to reverse the court decision? Should it then be open to the courts to say, “well, even though the text of the Constitution doesn’t say that the unanimity formula applies to alter democratic rights, we will impose our opinion of what the amending formula should be.” The courts would be making it impossible to change their own interpretation of the democratic rights in the Charter, even if most of the country strongly disagrees with that interpretation.

The 7/50 formula is difficult (albeit not as difficult as unanimity). It is unrealistic to think that seven provinces, comprising 50% of the population, plus the House of Commons, would ever agree on a radical amendment proposal that would get rid of democracy in Canada. Such an amendment is considerably more difficult now, under the 7/50 formula, than it was before 1982, when Canada could legally have been transformed into a dictatorship by an Act of the United Kingdom Parliament.
Tags: Gödel's Incompleteness Theorem, Kurt Gödel, albert einstein, amendment of the constitution of canada, constitutional amendment, constitutional law, oskar morgenstern, philip forman, politics, senate of canada, senate reform, supreme court of canada
  • Post a new comment


    default userpic

    Your reply will be screened

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.
  • 1 comment