A wobbly balance? A comparison of proportionality testing in Canada, the United States, the European Union and the World Trade Organization.

The law and the newspapers are full of instances in which two different, conflicting societal interests are being traded off--or balanced--against one another by the courts. Every democracy is faced with such choices, and in a given moment the choice may be simple (the decision to suppress publishing military secrets in wartime), or complex (the decision to suppress anti-war protests). Choices of this kind--constitutional choices--are so identified with the protection of civil rights that in the vernacular, the adjective "unconstitutional" is understood as intolerable state action encroaching on the rights of a person. Tests over constitutionality are among the most celebrated matches between a person and the state, generating rich public discourse. For instance, do the courts go too far when they protect the rights of unpopular persons, such as the right of Nazis to parade though a neighborhood of Holocaust survivors? (1) The answer, of course, depends on how courts balance competing interests.

However, in the last several decades, balancing has ceased to be the exclusive preserve of national human rights law. International trade courts, applying the General Agreement on Tariffs and Trade (GATT) (2) and other European Community treaties, now use proportionality testing to balance national interests and the objectives of free trade. Yet, among the thorny questions of the free trade debate whether jobs are lost, environmental or health standards weakened, or developing countries fairly treated, etc.--it is rarely asked whether trade courts balance competing interests appropriately. Public attention is focused on the results of trade adjudication, or the language of the trade agreements, but seldom the process of how trade courts balance competing interests.

Why does something as subtle as the process of balancing matter? Plainly, it matters because the process can render the law highly malleable. The same court, in the same country, faced with the same substantive law, can reach one judgment, and, shortly thereafter, reach its polar opposite. Upon close inspection, the only real difference may be how the court balanced the competing interests. The rise and fall of the "separate but equal" doctrine of racial segregation in the United States is a famous example of this and suggests that to understand the balancing process it is best to examine a few illustrative examples. (3)

This study is a cross-jurisdictional survey of judicial balancing and proportionality testing--the primary juridical tool which courts use in balancing. This study has two purposes. The first purpose is expository: to describe in detail the legal principles and judicial application of balancing and proportionality testing in four different tribunals (the Supreme Court of Canada; the United States Supreme Court; the ECJ; and the panels of the WTO). This exposition sets the stage for the paper's second, discursive purpose: to propound theories of how the tribunals can articulate superficially similar proportionality tests, yet ultimately achieve vastly different results.

PART I--WHAT ARE BALANCING AND PROPORTIONALITY?

Balancing is a judicial exercise aimed at reconciling two legitimate and conflicting interests, by ranking them hierarchically, and determining the permissible extent of conflict between them. The object of any balancing exercise is to arrive at a proportional result: affirming the superior interest, yet allowing the inferior interest to coexist to the extent that it remains compatible. The balancing inquiry may be approached in different ways, but it always requires the judge to take certain logical steps: (4)

(1) There must be some ascription of weight or value to the interests concerned,

(2) A determination must be made as to whether certain interests may be traded off to achieve other goals. Some interests (e.g. the right to life), may be so highly valued that we would not compromise them for the sake of other interests (e.g. the right to shout "fire" in a crowded theater).

(3) If a trade off is appropriate, a decision must be made as to whether the intrusion on the superior interest is proportionate. The proportionality test itself may be phrased a number of ways. For example:

(a) Are the means adopted to achieve an objective rationally connected to the objective, and if so, are the means the least restrictive which could be adopted in the circumstances? Or,

(b) Are the means adopted to achieve an objective congruent with the importance of the objective, and are they necessary for its achievement? Or,

(c) Are the means adopted suitable and necessary for the achievement of the objective, and do they not impose excessive burdens on the interests which are adversely affected? Or,

(d) What are the relative costs and benefits of the means chosen to attain an objective?

(4) Having decided which proportionality test to use, the judge must decide on the stringency or the intensity of judicial review. As a general rule, if the state infringes a highly valued interest, a more stringent formulation of the proportionality test in step (3) will be used.

Balancing and proportionality testing aspire to be rigorous, almost quasi-scientific in nature, for good reason. Although the law is shot through with discretion, openly acknowledging that fact makes lawyers and judges uneasy--discretion reeks of subjectivity. Lawyers and judges prefer to distance themselves from discretion, using tests that suggest logical purity, intellectual legitimacy, and legal "truth". (5)

However, just because proportionality testing appears to be quasi-scientific does not mean that it is actually so. There remains abundant, covert discretion secreted away within the steps of proportionality testing. Some academics have cautioned about the judicial latitude that exists when characterizing competing interests, arguing that judges can preordain the outcome of a case by picking either a broad or a narrow characterization of the interests to be balanced. (6) Some scholars believe this latitude is enough to question whether the public should have trust in such judicial decisions at all. (7) Other scholars disagree, and while acknowledging certain permutations in different courts, they find in proportionality testing a subtle proof of bedrock legalism and truth--no less than "the ultimate rule of law". (8)

This paper falls between these two views, though nearer the former. Proportionality testing, despite the quasi-scientific incantations that surround it, is no better than open, notorious and bare discretion in finding legally true results. It is the sugar coating that renders bitter discretion palatable to lawyers and judges. But the process of proportionality testing forces lawyers and judges to formulate their judgments, rendering the outcome more reasoned than tendentious and more true. Even if proportionality testing can be manipulated by a judge to produce the desired result, the distortion to the legal truth is evident in the written judgment.

PART II--PARAMETERS OF THE COMPARISON:

When comparing four jurisdictions, with four separate bodies of substantive law and four tribunals, a few simplifying assumptions are required.

First, this comparison is restricted in the areas of law it surveys. Even within a single jurisdiction, balancing is done differently depending on the legal interest at stake. In Canada, the courts balance an infringement of the right to life quite differently than the right to freedom of expression--both would confound the comparison. Thus, for the Supreme Courts of Canada and the United States, I focus on the law of free expression; for the ECJ and the Panels of the WTO, I focus on the law of free trade of goods. To show how proportionality testing can be manipulated by judges, it is more important to go deep rather than broad in this comparison. Though the number of cases presented is small, they are looked at in great detail.

Second, it is relevant to briefly set out the fundamental features of the tribunals.

Along jurisprudential lines: The Supreme Courts of Canada and the United States are largely human rights-based, and decide cases involving fundamental freedoms. The ECJ is largely, but not exclusively, concerned with international trade-based disputes. The WTO Panels adjudicate trade-based disputes alone.

Along historical lines: The Supreme Courts of Canada and the United States descend from the precedent-driven, adversarial, and verbose English law tradition. The ECJ stems from a less precedential, less adversarial, and terser European civil law tradition. The WTO Panels are unique and almost without tradition, as they stem from a global trade regime that dates back only to 1947.

Along political and institutional lines: The Supreme Courts of Canada and the United States are the permanent judicial institutions of national governments. The ECJ is a permanent judicial institution of a federal European government. The WTO Panels are ad hoc administrative tribunals unaffiliated with any government.

It is these distinctions--of structure, history and approach--that lie at the root of the method by which these tribunals make use of balancing and proportionality testing. This argument requires a detailed exposition and dissection of how the tribunals have used proportionality testing, which is the subject of the next four sections.

1. The Supreme Court of Canada (SCC):

The major instrument guaranteeing human and civil rights in Canada is the Canadian Charter of Rights and Freedoms, which is part of the national constitution. (9) The Charter is relatively new, and marks a departure from the old Diceyan model of Canadian parliamentary supremacy.

Section 2(b) of the Charter provides for the right of freedom of expression:

2. Everyone has the following fundamental freedoms: (b) freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication;

This freedom is qualified by section 1 of the Charter, which provides an explicit limitation across the Charter:

1. The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

Section 1 provides that in balancing a limitation on a fundamental freedom, the Court must make two inquiries: a) that the limit be "prescribed by law"; and b) that it be a "reasonable limit ... demonstrably justified in a free and democratic society." It is the latter consideration that commands more judicial attention. (10)

The s. 1 balancing of interests is particularly important in freedom of expression cases because the threshold of what qualifies as expression is so low. In Canadian law "expression" means any utterance or activity, excepting those which are violent. (11) The SCC does not religiously distinguish between measures that purposefully, rather than incidentally, stifle expression. (12)

Early in the Charter's history, despite the clear wording of s. 1, the Court deviated from applying its words directly. Instead, the Court outlined the dominant approach to s. 1 in the leading case of R. v. Oakes. (13) As Oakes reads:

   To establish that a limit is reasonable and demonstrably justified
   in a free and democratic society, two central criteria must be
   satisfied [by the state]. First, the objective, which the measures
   responsible for a limit on a Charter right or freedom are designed
   to serve, must be 'of sufficient importance to warrant overriding a
   constitutionally protected right or freedom' ... The standard must
   be high in order to ensure that objectives which are trivial or
   discordant with the principles integral to a free and democratic
   society do not gain s. 1 protection. It is necessary, at a minimum,
   that an objective relate to concerns which are pressing and
   substantial in a free and democratic society before it can be
   characterized as sufficiently important.

   Second, once a sufficiently significant objective is recognized ...
   [the state] must show that the means chosen are reasonable and
   demonstrably justified. This involves 'a form of proportionality
   test' ... Although the nature of the proportionality test will vary
   depending on the circumstances, in each case courts will be
   required to balance the interests of society with those of
   individuals and groups. There are, in my view, three important
   components of a proportionality test. First, the measures adopted
   must be carefully designed to achieve the objective in question.
   They must not be arbitrary, unfair or based on irrational
   considerations. In short, they must be rationally connected to the
   objective. Second, the means, even if rationally connected to the
   objective in this first sense, should impair 'as little as
   possible' the right or freedom in question ... Third, there must be
   a proportionality between the effects of the measures which are
   responsible for limiting the Charter right or freedom, and the
   objective which has been identified as of 'sufficient importance'.
   (14)

The Oakes proportionality test has proved so ubiquitous in the Court's Charter jurisprudence that scholars lament it having "the character of holy writ". (15) While it would be incorrect to equate the Oakes test with s. 1 stricto sensu, nearly all Charter judgments cite it or variants thereof. (16)

The approach of the Oakes test has the great advantage of explicitness. Its discrete steps help the Court state exactly how a measure runs afoul (or not) of the proportionality inquiry. Interestingly, there has never been a case where the third step of the Oakes test was decisive to the outcome. (17) It is usually the first proportionality inquiry, known as the "minimal impairment" test, which is fatal. Accordingly, judges and litigants tend to manipulate the intensity of judicial review at this first step. (18)

Certain principles have evolved in the s. 2(b) line of cases that shed light on how the court calibrates the intensity of judicial review under s. 1 and the Oakes test. The most important principle stems from the rule that Charter interpretation is marked by a teleological, or "purposive approach". The intensity of judicial review varies, becoming deferential when the expression in question accords with the notional purpose of s. 2(b), and unforgivingly strict when it does not. The court has articulated three "core" societal values which free expression is meant to promote: i) the pursuit of truth; ii) the fostering of social and political decision making; and, iii) the cultivation of diversity in modes of individual self-fulfillment and endeavour, which benefits society as a whole. (19) The further an expressive act strays from these core values, the less valuable it is, and the less worthy of constitutional protection.

The teaching of deeply anti-Semitic lessons to children attending public school is an example of expression far removed from these core values. In R. v. Keegstra, a teacher who lectured his students that the Holocaust was a historical fabrication of "subversive", "child killer" Jews, and graded them favorably for absorbing this repugnant lesson, was prosecuted under a criminal statute prohibiting hate propaganda. (20) Finding that Mr. Keegstra's utterances "stray[ed] some distance from the spirit of s. 2(b)" (21), and that Parliament's prohibition was rationally connected to the pressing and substantial objective of preventing harm caused by hate propaganda (22), the Court examined whether a total criminal prohibition was the minimally impairing means available for stanching hate propaganda. Dickson C.J.C. reasoned:

   In assessing the proportionality of a legislative enactment to a
   valid governmental objective ... s. 1 should not operate in every
   instance so as to force the government to rely upon only the mode
   of intervention least intrusive of a Charter right or freedom. It
   may be that a number of courses of action are available in the
   furtherance of a pressing and substantial objective, each imposing
   a varying degree of restriction upon a right or freedom. In such
   circumstances, the government may legitimately employ a more
   restrictive measure ... furthering the objective in ways that
   alternative responses could not. (23)

The state may actually employ more impairing means than strictly necessary, provided these means are more efficacious in the attainment of a pressing and substantial objective. The proviso is an important one: the state goes too far if it employs a more impairing means that is not demonstrably more efficacious. (24) This reasoning is compelling and was even welcomed by the minority in Keegstra, who held that the criminal prohibition was disproportionate because prosecuting Holocaust deniers gives them an aura of martyrdom and notoriety, making prosecution less efficacious than a non-criminal means of dissuading hate propaganda.

In addition to the core values principle, the Court is also willing to attenuate the Oakes proportionality test when the impugned law possesses certain features. This attenuation is evident where commercial expression is at issue; where the threshold at which regulatory action should be invoked is ambiguous and a matter of some discretion; where the state is concerned with the protection of vulnerable groups; and where the state has enacted social regulatory legislation aimed at "mediating between the claims of competing groups", as opposed to criminal legislation where the state is "the singular antagonist of the individual". (25) These criteria arise as questions of legislative fact, because legislators have attempted to balance the competing interests and subsequently codify them. In such cases, the Court has always been reluctant to second-guess the legislator's own chosen balance.

All the above factors arise in the Irwin Toy case, where a toy company sought to overturn a Quebec regulation restricting television advertising directed at children under thirteen years of age. (26) The company argued that the regulation unreasonably limited its right to freedom of expression, which in this case was freedom to advertise towards children.

The majority of the court struggled with whether a line--protecting those under thirteen years of age, but not over--was too arbitrary a basis on which to sustain a limitation of rights. Undoubtedly, drawing such a line was more a matter of discretion than precise scientific knowledge, nor was it clear that it was "pressing and substantial" to pass a law for only those below the age watershed. Dickson C.J.C. reasoned:

   Where the legislature mediates between the competing claims of
   different groups in the community, it will inevitably be called on
   to draw a line marking where one set of claims legitimately begins
   and the other fades away without access to complete knowledge as to
   its precise location. If the legislature has made a reasonable
   assessment as to where the line is most properly drawn, especially
   if that assessment involves weighing conflicting scientific
   evidence and allocating scarce resources on this basis, it is not
   for the court to second guess. (27)

What this passage articulates is a rule for dealing with scientific uncertainty: In Canadian law, the court will defer to the state if it can be shown that the state made a "reasonable assessment" in legislating protection at the threshold it did. Taken together with the rule in Keegstra, the state can legislate at any reasonable threshold, so long as with each increment taken from the right of expression, the legislative objective is furthered in ways that alternative responses could not. This approach grants extraordinary deference to the state. Dickson C.J.C. reiterated this reasoning in his minimal impairment analysis and listed additional principles upon which judicial deference would lie:

   [I]n matching means to ends and asking whether rights or freedoms
   are impaired as little as possible, a legislature mediating between
   the claims of competing groups will be forced to strike a balance
   without the benefit of absolute certainty concerning how the
   balance is best struck. Vulnerable groups [i.e. children] will
   claim the need for protection by the government whereas other
   groups and individuals will assert that the government should not
   intrude ... Thus, as courts review the results of the legislature's
   deliberations, particularly with respect to the protection of
   vulnerable groups, they must be mindful of the legislature's
   representative function ...

   In other cases, however, rather than mediating between different
   groups, the government is best characterized as the singular
   antagonist of the individual whose right has been infringed. For
   example, in justifying an infringement of legal rights [arising in
   the criminal process] ... the state, on behalf of the whole
   community, typically will assert its responsibility for prosecuting
   crime whereas the individual will assert the paramountcy of [his
   rights] ... In such circumstances ... the courts can assess with
   some certainty whether the 'least drastic means' for achieving the
   purpose have been chosen, especially given their accumulated
   experience in dealing with such [criminal law] questions ... (28)

It is easiest to capture Dickson C.J.C.'s meaning by reading the final line of each of the two passages together. Courts have "accumulated experience" when it comes to matters such as criminal prosecutions, and that entitles them to use an intense, undeferential standard of judicial review. But where courts are navigating through matters where they are inexperienced, like advertising directed to children, "they must be mindful of the legislature's representative function" and defer to it. (29) Parliament has the difficult task of accommodating the interests of vulnerable persons, reaching conclusions in a sea of scientific uncertainty, and balancing competing social interests. These conclusions signal the Court to attenuate the stringency of its judicial review and blunt the minimal impairment test. The less that the Court knows about a subject matter, the less aggressive it will be regarding the minimal impairment test.

The SCC takes a purposive approach to calibrating judicial deference--faithful to the Oakes test and religiously passing the cases through the different steps of proportionality testing. All in all, this approach amounts to a coherent and logical package--but that does not mean it is legally true. Upstream of proportionality testing, the purposive approach to defining Charter rights is certainly very malleable. Skeptics wonder: How anyone can objectively ascertain the true purpose for the existence of a Charter right? (30) Even supposing the true purpose that lies behind a Charter right were somehow decipherable, judges still have far-ranging latitude to decide on the appropriate intensity of judicial review. No matter how rigorous and methodical it seems, the Oakes test leaves lawyers and judges a lot of wiggle room.

However, there is little to the contention that the SCC sets out to defeat proportionality testing or to expand its discretionary wiggle room. The teleological principles that qualify and attenuate the minimal impairment inquiry are sound ones. Even if differences arise over their proper application, it is also true that the Court consistently returns to the same principles. (31) The ubiquity of the Oakes-like test in the Canadian jurisprudence is reassuring, even if it is not the final word on true law.

2. The United States Supreme Court:

The United States shares a number of liberal, legal traditions with Canada: the rule of law, judicial review, and a constitutional Bill of Rights. But where the Canadian Parliament is familiar with sovereignty, America's Congress lives in a republic accustomed to checks and balances; where the Canadian Charter hedges its rights with s. 1, the American Bill of Rights speaks in absolutes.

The first amendment of the Bill of Rights provides that, "Congress shall make no law ... abridging the freedom of speech or of the press." The due process clause of the fourteenth amendment extends this constitutional guarantee to state and local governments as well. (32) Taken together, freedom of speech in the American Bill of Rights is not very different from free expression in the Canadian Charter.

However, there is a large difference between the U.S. and Canada on how the expression right is limited. Without explicit hedging words (such as those that exist in s. 1 of the Canadian Charter), American courts have no choice but to balance the competing interests with the first amendment right itself. That balancing can be done either at the threshold of a case gaining admittance to the right, or by applying a proportionality test once a case has come within the right's sphere. The U.S. Supreme Court exerts control at both points, perhaps reflecting an ethos in American culture that idealizes free expression, and makes slaying this ideal on the chopping block of balancing hard to do.

Thus, the court has fashioned dispositive rules for certain expression cases. Comparatively valueless utterances such as defamation (33), "purely commercial advertising" (34), "fighting words" (35), or expression leading to a "clear and present danger" (36), do not cross the threshold of the constitutional guarantee. Also, laws which infringe expression purposefully, by a prior restraint on the expression's content, are "presumptively inconsistent" with the first amendment. (37) There is no balancing in cases of these kinds.

It is in the penumbra, outside these dispositive rules, that the court first gave balancing a contemptible birth. In the case of Konigsberg v. California, the Court divided narrowly (5:4) and bitterly on whether a lawyer, declining to answer questions about membership in the Communist Party, could be refused admission to the California state bar. (38) Harlan J., for the majority, curiously identified California's rule as content-neutral, deciding that in "the exercise of valid governmental powers a reconciliation must be effected, [which] ... perforce requires an appropriate weighing of the respective interests involved." (39) In the fearful era of the cold war, that balance naturally favoured the state over the suspected communist. But Black J., dissenting, thought this result (and the balancing test that made it possible) "penurious":

   The [majority of the] Court, by stating unequivocally that there
   are no 'absolutes' under the First Amendment, necessarily takes the
   position that even speech that is admittedly protected by the First
   Amendment is subject to the 'balancing test' and that therefore no
   kind of speech is to be protected if the Government can assert an
   interest of sufficient weight to induce this Court to uphold its
   abridgement. In my judgment, such a sweeping denial of the
   existence of any inalienable right to speak undermines the very
   foundation upon which the First Amendment, the Bill of Rights, and,
   indeed, our entire structure of government rest. The Founders of
   this Nation attempted to set up a limited government which left
   certain rights in the people-rights that could not be taken away
   without amendment of the basic charter of government. The
   majority's 'balancing test' tells us that this is not so. It tells
   us that no right to think, speak or publish exists in the people
   that cannot be taken away if the Government finds it sufficiently
   imperative or expedient to do so. Thus, the 'balancing test' turns
   our 'Government of the people, by the people and for the people'
   into a government over the people. (40)

What Black J. feared came true: balancing now dominates American freedom of expression cases. Balancing emerged haphazardly, from such opposition, and is now evident in the Court's diverse and inconsistent proportionality tests.

The prototypical American balancing test is that articulated by the Court in United States v. O'Brien. (41) Mr. O'Brien, a young man in the Vietnam era, chose to protest conscription by the provocative act of burning his draft card on the steps of the Boston courthouse. He was prosecuted under a federal statute which criminalized the willful destruction or mutilation of a draft card. In defence, O'Brien argued that he had not burned his card wantonly, but as an act of constitutionally protected political expression. The government, fearful of a nationwide conflagration of draft cards, warned that if O'Brien's defence succeeded, it would sap Congress of its constitutional power to raise an army. The extent to which the law operated as a limit on expression was merely incidental, the government argued, and justifiable in light of the government's compelling, constitutional interest.

Thus in O'Brien, the Court was faced with a content-neutral prohibition on expression, and a legitimate interest set against it. The Court balanced:

   This Court has held that when 'speech' and 'nonspeech' elements are
   combined in the same course of conduct, a sufficiently important
   governmental interest in regulating the nonspeech element can
   justify incidental limitations on First Amendment freedoms. To
   characterize the quality of the governmental interest which must
   appear, the Court has employed a variety of descriptive terms:
   compelling; substantial; subordinating; paramount; cogent; strong.
   Whatever imprecision inheres in these terms we think it clear that
   a government regulation is sufficiently justified if it is [i]
   within the constitutional power of the Government; [ii] if it
   furthers an important or substantial governmental interest; [iii]
   if the governmental interest is unrelated to the suppression of
   free expression; and [iv] if the incidental restriction on alleged
   First Amendment freedoms is no greater than is essential to the
   furtherance of that interest. (42)

The O'Brien test is in four parts, but so similar to the Canadian Oakes test that it can be described in Oakes' language. The first step requires that a limit on expression must be "prescribed by law"; the second step requires a "pressing and substantial" objective in limiting fundamental rights; and the fourth step is a proportionality test, equivalent to the "minimal impairment" test. (The third step is a vestige of the old dispositive rule, that laws purposefully infringing expression never survive the O'Brien test.)

Applying this test to the present facts, the Court held that the state had a "substantial interest in assuring the continuing availability of issued [draft cards]"; and by characterizing this interest tautologically with the criminal prohibition against destroying the cards, the Court upheld Mr. O'Brien's conviction. Characterizing the state's interest this way effectively reduced the O'Brien test to a nullity: if the state's interest is to protect draft cards, what less impairing way can exist but to forbid their destruction? Thus, in O'Brien the Court invented a brave new proportionality test, but cynically manipulated it on the first outing. (43)

Later American cases borrow from the O'Brien test, but show markedly greater sensitivity. In the case of Ward v. Rock Against Racism, measures adopted by New York City to control amplified sound from gigs at the Central Park bandshell were in question. (44) After persistent noise complaints and failed efforts at policing, the city mandated that users of the bandshell must employ the city's own amplification equipment and sound technician. Rock Against Racism argued that placing the mixing board in the city's hands removed artistic control over the acoustics and detracted from the emotive and expressive force of the music, which was "dedicated to the espousal and promotion of antiracist views". (45) The question was therefore whether this incidental limitation on expression was justifiable in light of the legitimate state objective of controlling noise.

The Court based its analysis on another case: Clark v. Community for Creative Non-Violence. (46) In that case, protesters camped out in a park within sight of the White House, in order to draw attention to the plight of the nation's homeless. The protesters were charged with violating Washington's anti-camping laws. In their defence, they argued that purely political expression was constitutionally inviolate. The Court ruled against them, holding even purely political expression could be subject to time, place and manner restrictions:

   Our cases make clear ... that even in a public forum the government
   may impose reasonable restrictions on the time, place, or manner of
   protected speech, provided the restrictions '[i] are justified
   without reference to the content of the regulated speech, [ii] that
   they are narrowly tailored to serve a significant governmental
   interest, and [iii] that they leave open ample alternative channels
   for communication of the information.' (47)

The time, place and manner cases like Rock Against Racism and Creative Non- Violence use two proportionality tests: the first to ensure restrictions are narrowly tailored to further a state interest of some gravity (i.e. the O'Brien test redux); the second, to ensure ostensible time, place and manner restrictions are not so onerous to foreclose "alternative channels for communication". (The latter is reminiscent of the neglected, final part of the Oakes test, which concerns proportionality between a measure, and its effects.)

Thus the Court in Rock Against Racism and Creative Non-Violence had to decide whether the impugned restrictions were "narrowly tailored" to the objectives of controlling noise and maintaining Washington city parks. But what should be the intensity of judicial review in "time, place and manner" cases? The Rock Against Racism majority wrote:

   ... a regulation of the time, place, or manner of protected speech
   must be narrowly tailored to serve the government's legitimate,
   content-neutral interests, but ... it need not be the least
   restrictive or least intrusive means of doing so. Rather, the
   requirement of narrow tailoring is satisfied 'so long as the ...
   regulation promotes a substantial government interest that would be
   achieved less efficiently absent the regulation.' (48)

This decision significantly dilutes the meaning of "the least restrictive or least intrusive means", and it is tantamount to judicial deference, as in Irwin Toy. In American legal language, the distinction is sometimes called "strict scrutiny" versus mere "reasonableness". (49) In plumping for reasonableness, the Rock Against Racism majority wrote:

   So long as the means chosen are not substantially broader than
   necessary to achieve the government's interest, however, the
   regulation will not be invalid simply because a court concludes
   that the government's interest could be adequately served by some
   less-speech-restrictive alternative. (50)

Why this deference? It would not take a very imaginative judge to strike down a law by proposing an improvement, no matter how trivial, signifying that the legislator's way was disproportionate and invalid. Seeking to limit such judicial shenanigans, the majority continued:

   The validity of time, place or manner regulations does not turn on
   a judge's agreement with the responsible decision maker concerning
   the most appropriate method for promoting significant government
   interests or the degree to which those interests should be
   promoted. (51)

Accordingly, New York City mixed the sound at its bandshell, and Washington's campers decamped. The Court was satisfied that these were reasonable solutions, which permitted alternative channels of musical or political expression.

Rock Against Racism, Creative Non-Violence, and O'Brien are examples of a "balancing revolution" in the American expression jurisprudence, which has largely replaced the old dispositive rules. Consider the test for commercial speech or advertising, which, although formerly unprotected by the first amendment, became subject to balancing in Central Hudson Gas v. Public Service Comission:

   At the outset, we must determine whether the expression is
   protected by the First Amendment. For commercial speech to come
   within that provision, it at least [i] must concern lawful activity
   and not be misleading. Next, we ask [ii] whether the asserted
   governmental interest is substantial. If both inquiries yield
   positive answers, we [iii] must determine whether the regulation
   directly advances the governmental interest asserted, and [iv]
   whether it is not more extensive than necessary to serve that
   interest. (52)

Rather than asking whether a measure is "no greater than is essential", "least restrictive", or "narrowly tailored" as O'Brien, Rock Against Racism and Creative Non-Violence do, the Central Hudson test asks whether it is "not more extensive than necessary". Central Hudson is also missing the second proportionality test of Creative Non-Violence, which pits measures against their effects-a major inconsistency even if that step of the test is hardly used in America.

Eventually, the Court tried to tidy up its plethora of balancing jurisprudence in a single, all-embracing balancing test. It did so in Board of Trustees of the State University of New York v. Fox, which is a commercial expression case:

   If the word 'necessary' is interpreted strictly, [it] would
   translate into the 'least-restrictive-means' test ... There are
   undoubtedly formulations in some of our cases that support this
   view--for example the statement in Central Hudson itself that 'if
   the governmental interest could be served as well by a more limited
   restriction on commercial speech, the excessive restrictions cannot
   survive' ... In San Francisco Arts & Athletics v. United States
   Olympic Committee, we said that the application of the Central
   Hudson test [for commercial expression] was 'substantially similar'
   to the application of the test for validity of time, place, and
   manner restrictions upon protected speech--which we have
   specifically held does not require least restrictive means ...

   Our jurisprudence has emphasized that 'commercial speech [enjoys] a
   limited measure of protection, commensurate with its subordinate
   position in the scale of First Amendment values' ,and is subject to
   'modes of regulation that might be impermissible in the realm of
   non-commercial expression' ...

   What our decisions require is a 'fit between the legislature's ends
   and the means chosen to accomplish those ends,' a fit that is not
   necessarily perfect, but reasonable; that represents not
   necessarily the single best disposition but one whose scope is 'in
   proportion to the interest served'; that employs not necessarily
   the least restrictive means but, as we have put it in the other
   contexts discussed above, a means narrowly tailored to achieve the
   desired objective. Within those bounds, we leave it to governmental
   decisionmakers to judge what manner of regulation may best be
   employed. (53)

Fox joins together the Court's sundry balancing tests, distilling from them a single, deferential proportionality inquiry. As with Irwin Toy in Canada, Fox calibrates the intensity of judicial review to the expression at question, and whether it accords with the teleology of free expression.

However, in another display of sloppiness, the Court failed to pronounce on the teleology of free expression, leading shortly to self-perplexity. (54)

In R.A.V. v. City of St. Paul, R.A.V., a minor, allegedly burned a cross in the garden of a black family, and was prosecuted under a city ordinance criminalizing hateful expression aimed at minorities. (55) R.A.V. pleaded that the ordinance was invalid because it placed a content-based restriction on expression. (56) Such a case ties the old dispositive rules in knots: was the law invalid because it was a prior restraint on the content of expression, or was the expression of burning a cross constitutionally unprotected because it is "fighting words"? The court swept aside those rules and attempted a balancing test, with curious results. As Scalia J. writes:

   St. Paul has not singled out an especially offensive mode of
   expression it has not for example, selected for prohibition only
   those fighting words that communicate ideas in a threatening (as
   opposed to a merely obnoxious) manner. Rather it has proscribed
   fighting words of whatever manner that communicate messages of
   racial, gender, or religious intolerance. Selectivity of this sort
   creates the possibility that the city is seeking to handicap the
   expression of particular ideas ...

   We do not doubt that [St. Paul's stated legislative objectives] ...
   are compelling, and that the ordinance can be said to promote them.
   But the 'danger of censorship' presented by a facially
   content-based statute requires that weapon to be employed only
   where it is 'necessary to serve the asserted compelling interest'.
   The existence of adequate content-neutral alternatives 'undercuts
   significantly' any defense of such a statute ... The dispositive
   question in this case, therefore, is whether content discrimination
   is reasonably necessary to achieve St. Paul's compelling interests;
   it plainly is not. An ordinance not limited to the favored [visible
   minorities] would have precisely the same beneficial effect. In
   fact the only interest distinctively served by the content
   limitation is that of displaying the city council's special
   hostility toward the particular biases thus singled out. (57)

Thus the Court's majority struck down the ordinance, for the unusual reason of underbreadth. The majority held that the state could lawfully prohibit hateful expression in all its forms, but could not prohibit it in only some of its forms. More to the point, the Court held that government could not only prohibit hateful expression aimed a historically disadvantaged minorities (which is content-based), but to be constitutional, it had to prohibit hateful expression aimed at all persons (which is content-neutral).

R.A.V. is an odd case. Instead of balancing the ordinance against the city's objective of "[ensuring] the basic human rights of members of groups that have historically been subjected to discrimination", (58) the Court balanced the ordinance against the "danger of censorship presented by a facially content-based statute". (59) That is, instead of asking if the ordinance minimally impaired the first amendment right to express oneself, the Court erred, and asked if it minimally impaired the first amendment law which doctrinally favours content-neutral prohibitions.

The Court would never have made this error if it kept its eyes properly on the teleology of the First Amendment. Cross-burning is hateful expression, and like commercial expression in Fox, it occupies a "subordinate position in the scale of First Amendment values", and merits a "limited measure of protection". (60) Accordingly, the Court should have relaxed its intensity of judicial review and given St. Paul's ordinance much deference. Instead, Scalia J. conjured up a clever alternative--a prohibition against hateful expression, no matter who it targets--that he imagined impaired rights less, and cited it as a reason to strike down St. Paul's law.

R.A.V. may be an aberration in American law; up to 2006, it has been cited but never really followed by the Court. Nevertheless, R.A.V. has thrown American expression jurisprudence into confusion. For example, R.A.V. says it is appropriate to balance laws that furnish a prior restraint on content, but O'Brien, Creative Non-Violence, and Rock Against Racism expressly reject balancing for such laws. Also, Central Hudson and Fox indicate that the intensity of judicial review should be calibrated based on the teleology of expression, but R.A.V. applies a very undeferential standard to what is worthless hate expression. These uncertainties represent the growing pains of new jurisprudence, as the U.S. transitions from dispositive rules based on categories of expression to the fine art of balancing. Which is better? Certainly balancing is the more intellectually satisfying--but before a howler of a case like R.A.V., it cannot be said that proportionality testing has led to the "ultimate rule of law", or advanced toward legal truth.

3. The European Court of Justice:

As noted earlier, the European Community is primarily an economic free trade area with sovereign and constitutional features, existing to foster the free movement of goods, services, workers, companies and capital. Both the legal institutions of the Community, and the commercial rights it seeks to promote, are outlined by the 1957 Treaty Establishing the European Community. (61) Early in the Community's history, the ECJ gave the Treaty quasi-constitutional force when it held that most of the Treaty's provisions and the Community's subordinate legislation have "direct effect"--which is to say that they have the force of law and bind member states even without being enacted in municipal law. (62) Direct effect, in the words of one commentator, is "drawn from a perception of the constitutional system of the EC, which ... continues to inspire the whole doctrine flowing from it". (63) Member states' legislative competence is limited by a duty of conformity with Community. (64)

The importance of proportionality in Community law cannot be emphasized enough. It is one of three "general principles" of Community law, (65) and although it is used by the ECJ in the balancing of competing interests, proportionality is actually in its weakest incarnation as a judicial tool. At its strongest, lack of proportionality can itself be a cause of action in judicial review. (66) Community legislation or administrative action may be challenged as disproportionate, even if it is substantively consistent with and legal under the Treaty. (67) Proportionality even towers over private law. For example, Article 119 of the Treaty stipulates that men and women are entitled to equal pay for equal work, whether in private or state employment, and if there is gender inequality, it must be demonstrably justified it must be proportionate. (68) Of course, nothing like this is possible in the common law paradigm of proportionality.

For the purposes of this study the focus will only be on the role of proportionality in public law, as it governs the free market of goods inside the Community.

Market integration is a core function of the Community. The Treaty abates two types of import trade barriers between member states. Explicit barriers, such as quotas or bans on certain goods, are known as quantitative restrictions. Implicit barriers, such as language laws for labeling goods, are known as "measures having an equivalent effect to quantitative restrictions" (MEQRs). (69) Both quantitative restrictions and MEQRs are prohibited by Article 30 of the Treaty:

Article 30. Quantitative restrictions on imports and all measures having equivalent effect shall, without prejudice to the following provisions, be prohibited between Member States.

A number of Treaty articles follow which qualify or derogate from Article 30. The most important of these is Article 36:

Article 36. The provisions of Arts. 30 to 34 shall not preclude prohibitions or restrictions on imports, exports or goods in transit justified on grounds of public morality, public policy or public security; the protection of health and life of humans, animals or plants; the protection of national treasures possessing artistic, historical or archaeological value; or the protection of industrial and commercial property. Such prohibitions or restrictions shall not, however, constitute a means of arbitrary discrimination or a disguised restriction on trade between Member States.

Therefore, if a member state wishes to impose a quantitative restriction, it must justify that restriction in accordance with the Article 36 derogations, or the quantitative restriction will be illegal. The law is slightly more generous for MEQRs, which may be justified in accordance with Article 36, or one of the additional derogations cited in the leading case of Cassis de Dijon:

   Obstacles to movement within the Community resulting from
   disparities between the national laws relating to the marketing [of
   goods] must be accepted [only] in so far as those provisions may be
   recognized as being necessary in order to satisfy mandatory
   requirements relating in particular to the effectiveness of fiscal
   supervision, the protection of public health, the fairness of
   commercial transactions and the defence of the consumer. (70)

Under Cassis de Dijon, member states may pass laws controlling the marketing of goods traded in the Community, so long as the laws are necessary to satisfy one of the mandatory requirements.

When a national law, a quantitative restriction or MEQR, conflicts with the right of free trade in Article 30, the resolution depends on balancing the two interests. A typical test for doing so is found in the ECJ's FEDESA case:

   The Court has consistently held that the principle of
   proportionality is one of the general principles of Community law.
   By virtue of that principle, the lawfulness of the prohibition of
   an economic activity is subject to [i]
   the condition that the prohibitory measures are appropriate
   and necessary in order to achieve the objectives legitimately
   pursued by the legislation in question; [ii] when there is a
   choice between several appropriate measures recourse must be
   had to the least onerous, and [iii] the disadvantages caused
   must not be disproportionate to the aims pursued. (71)

The FEDESA test presupposes the existence of valid legislative objectives, and resembles Canada's Oakes or America's Fox tests. The wording is a bit clumsy but the intent is unmistakable. (72) Context too matters, and just as the context of "fighting words" played havoc with America's proportionality test in R.A.V., context also matters in the Community, when the interests at stake are not human rights but economic rights.

In the German Beer case, the Commission challenged two German laws, alleging they were illegal MEQRs that blocked imports from the German beer market. (73) A "Purity Law" mandated that bottom-fermented beers be brewed from only malted barley, hops, yeast and water. A "Foodstuffs Law" prohibited food additives, except where those were shown to be harmless and necessary because they met an essential technological, nutritional or dietary need.

Germany argued that these laws were necessary to minimize the consumption of adulterated food, and to protect the health of its citizens as is outlined under Article 36. Adherence to Purity Law's beer recipe obviated the need for any additives or preservatives, and accordingly, it could not be said that additives were necessary under the Foodstuffs Law. While those laws had the effect of keeping foreign beers off the German market, this was desirable where beer was 26.7% of the average German male's caloric intake (!) and thus a staple food. (74)

The Commission, interestingly, contested none of these factual submissions, but cast a doubtful light on the proportionality of the German laws within the Community. It agreed that additives should only be used where necessary, but argued that the threshold of necessity was lower than mandated by the Foodstuffs Law. An additive is necessary where it "creates a real advantage for the consumer ... [for instance] to facilitate or improve the manufacture of a given product, make it keep better, or ... improve its quality or its presentation, or ... provide a wider variety of flavours." (75) The Commission argued that there must be a "presumption that products containing additives [used in another] Member State are harmless." (76)

In essence, where Germany stressed public health precaution for Germans, the Commission instead stressed commercial convenience for Europeans. The Court manipulated deftly:

   The Court has consistently held ... that 'in so far as there are
   uncertainties at the present state of scientific research it is for
   the Member States, in the absence of harmonization, to decide what
   degree of protection of the health and life of humans they intend
   to assure, having regard however to the requirements of the free
   movement of goods within the Community.'

   ... in such circumstances Community law does not preclude the
   adoption by the Member States of legislation whereby the use of
   additives is subjected to prior authorization ... Such legislation
   meets a genuine need of health policy, namely that of restricting
   the uncontrolled consumption of food additives.

   [However] ... It must be borne in mind that in its [past] judgments
   ... the Court inferred from the principle of proportionality
   underlying the last sentence of Article 36 of the Treaty that
   prohibitions on the marketing of products containing additives
   authorized in the Member State of production but prohibited in the
   Member State of importation must be restricted to what is actually
   necessary to secure the protection of public health. The Court also
   concluded that the use of a specific additive which is authorized
   in another Member State must be authorized in the case of a product
   imported from that Member State where, in view, on the one hand, of
   the findings of international scientific research, and in
   particular the work of the Community's Scientific Committee for
   Food, the Codex Alimentarius Committee of the Food and Agriculture
   Organization of the United Nations (FAO) and the World Health
   Organization, and, on the other hand, of the eating habits
   prevailing in the importing Member State, the additive in question
   does not present a risk to public health and meets a real need,
   especially a [technological] one.

   It must be observed that the German rules on additives applicable
   to beer result in the exclusion of all the additives authorized in
   the other Member States and not the exclusion of just some of them
   for which there is concrete justification by reason of the risks
   which they involve in view of the eating habits of the German
   population ...

   As regards the need, and in particular the technological need, for
   additives ... It must be emphasized that mere reference to the fact
   that beer can be manufactured without additives if it is made from
   only the raw materials prescribed in the Federal Republic of
   Germany does not suffice to preclude the possibility that some
   additives may meet a technological need ...

   Consequently, in so far as the German rules on additives in beer
   entail a general ban on additives, their application to beers
   imported from other Member States is contrary to the requirements
   of Community law as laid down in the case-law of the Court, since
   that prohibition is contrary to the principle of proportionality
   and is therefore not covered by the exception provided for in
   Article 36 of the EEC Treaty. (77)

This is curious logic, for it pivots 180 degrees before it is done. The Court begins by proclaiming deference--that in the face of scientific uncertainty, national authorities have a power to "decide what degree of protection of the health and life of humans they intend to assure". But that deference soon turns into an onerous burden of proof: national authorities must show that their measures to protect health have "concrete justification by reason of risks". The logic is deeply contradictory, because naturally there can be no concrete justifications where there is scientific uncertainty. In this case, science could not prove or disprove that certain additives were harmless, and yet, the Court imposed a presumption that an additive employed in one member state is fit to be sold in another member state. (78) Essentially, the Court disallowed precaution.

The striking fact about German Beer is that the Court did not dwell on whether Germany's purity law pursues a valid objective (precautionary health protection) by "the least onerous" means possible, so much as whether Germany's pursuit of that objective would wreak disproportionate adverse effects on the European common market. Germany's case failed, because precautionarily prohibiting additives would overly disrupt the Community's internal trade. Used in this way, proportionality metamorphoses from a tool for balancing competing interests in the pursuit of a legitimate state objective, into a tool to attack whether the state ought to have pursued the objective at all.

The tendency of the Court to find disproportionality in a measure's effects is further illustrated by the Danish Bottles case. (79) Denmark passed environmental laws prescribing a national recycling scheme for beer, soft drink, and mineral water bottles. The scheme comprised a deposit-and-return system, and a requirement that beverage makers market drinks in one of eighteen approved bottles. Limiting the variety of bottles made it possible for retailers to accept, sort and reuse any empties--not just the empties corresponding to beverages they sold. Evidence showed the scheme worked superbly: 99% of bottles were recovered, to be refilled up to 30 times.

The Commission attacked Denmark's scheme as an MEQR because it impeded the importation of beverages from other member states, and it asked Denmark to accommodate non-approved bottles in its scheme. Denmark granted foreign producers an exemption of 3000 hectoliters (300,000 liters) in non-approved bottles, where cash deposits were collected just as for Danish products. The Commission did not regard this as sufficient, and sued at the ECJ to set aside the cash deposit scheme, the approved-bottle requirement, and the 3000 hectoliter exemption.

The Court readily agreed with Denmark that environmental protection was a valid "mandatory requirement" (i.e. derogation) like those in Cassis de Dijon, and that the cash deposits on non-approved bottles were indispensable and proportionate measures to promote recycling. (80) However, it took a different view on the 3000 hectoliter exemption, and the approved bottle system:

   The Danish Government stated in the proceedings before the Court
   that the present deposit-and-return system would not work if the
   number of approved containers were to exceed 30 or so, since the
   retailers taking part in the system would not be prepared to accept
   too many types of bottles owing to the higher handling costs and
   the need for more storage space ...

   In these circumstances, a foreign producer who still wished to sell
   his products in Denmark would be obliged to manufacture or purchase
   containers of a type already approved, which would involve
   substantial additional costs for that producer and therefore make
   the importation of his products into Denmark very difficult.

   It is undoubtedly true that the existing system for returning
   approved containers ensures a maximum rate of re-use and therefore
   a very considerable degree of protection of the environment since
   empty containers can be returned to any retailer of beverages ...

   Nevertheless, the system for returning non-approved containers is
   capable of protecting the environment and, as far as imports are
   concerned, affects only limited quantities of beverages ... In
   those circumstances, a restriction of the quantity of products
   which may be marketed by importers is disproportionate to the
   objective pursued.

   It must therefore be held that by restricting ... the quantity of
   [beverages] which may be marketed by single producers in
   non-approved containers to 3000 hectolitres a year ... Denmark has
   failed ... to fulfil its obligations under Article 30 of the EEC
   Treaty. (81)

The Court never took issue with Demark's objective of environmental protection, but rather with the extent or the threshold to which Denmark sought to achieve its objective, finding that the effect would disproportionately interfere with the Community's trade in goods. Proportionality would require Denmark to accept something less than the "maximum rate of re-use" of bottles, and to compromise its "very considerable degree of protection of the environment". Advocate General Slynn made this very point in his preliminary opinion to the Court:

   I accept, as Denmark contends, that it achieves the highest
   standards of environmental protection in respect of the collection
   of containers ... I also accept that it may be difficult by other
   methods to achieve the same high standard. Yet it does not seem to
   me that Denmark must succeed in this application unless the
   Commission can show that the same standard can be achieved by other
   specified means. There has to be a balancing of interests between
   the free movement of goods and environmental protection, even if in
   achieving the balance the high standard of the protection sought
   has to be reduced. (82)

Militant though it is, the result in Danish Bottles remains a flawless invocation of the final step of the FEDESA test: "the disadvantages caused [by a measure] must not be disproportionate to the aims pursued". (83) The Danish Bottles and German Beer decisions stand out because of the seeming audacity of the ECJ. In Canada or America, it would be unheard of for courts to strike down legislation that protected the environment or public health too much. This is an important point discounted by those who argue that proportionality is the "ultimate rule of law". Used this way, proportionality testing actually seems more like a vehicle for the ultimate lack of deference.

It is possible to be cynical and interpret German Beer and Danish Bottles as transforming proportionality into a device that centralizes power in the Community, at the expense of sovereign member states. While German Beer is offensive because it does this duplicitously (while paying lip service to national jurisdiction), Danish Bottles is up front and honest about requiring Denmark to reduce its environmental protection to ease the Community's commercial integration. Though unfamiliar to Canadians and Americans, the differences ought not to be too surprising, as in some civil law jurisdictions, disproportionality can itself be a cause of action.

4. The Adjudicative Panels of the WTO:

The General Agreement on Tariffs and Trade (GATT), established in 1948 and incorporated into the World Trade Organization (WTO) in 2005, is the keystone of the multilateral world trade system. (84)

GATT began in a most unusual way--without a central institution of any kind. All GATT business, including dispute settlement, had to be managed through consensus decision making. Disputes between the GATT members were heard by three-member Panels, usually composed of members' delegates. Panels heard evidence and argument, rendering judgment much as courts do, yet their orders were not binding on parties until adopted by consensus of all members. As such, decision making in the GATT was at times more like arbitration than litigation, with a political flavor. (85)

The WTO's advent in 1995 provided the long-missing institution for GATT and its sister trade treaties. GATT's ad hoc dispute resolution was replaced with a less political Dispute Settlement Body (DSB), capable of making binding judgments, and operating under the rules of the Dispute Settlement Understanding (DSU). (86) Panel members are now more widely recruited and include trade lawyers and academics. (87) The WTO's advent imbued GATT dispute settlement with a gravity and legalism lacking under the old system. (88)

Article XI:1 of the GATT, similar to Article 30 of the Treaty, prohibits quantitative restrictions on imports and exports: (89)

Article XI:1. No prohibitions or restrictions other than duties, taxes or other charges, whether made effective through quotas, import or export licences or other measures, shall be instituted or maintained by any contracting party on the importation of any product of the territory of any other contracting party or on the exportation or sale for export of any product destined for the territory of any other contracting party.

The language of Article XI:1 is broad and captures both explicit quantitative restrictions and MEQRs, which are called "non-tariff barriers" in GATT jurisprudence. Foreign goods are also protected from discriminatory marketing rules by the doctrine of National Treatment, at Article III:4:

Article III:4. The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin in respect of all laws, regulations and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use ...

National Treatment prevents a party from imposing rules that prejudice imported goods, as compared to "like products" of domestic origin.

Similar to Community law, GATT law allows parties to derogate from their obligations where they legitimately protect health or the environment. The exceptions are found in Article XX: (90)

Article XX. Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:

(b) necessary to protect human, animal or plant life or health;

(g) relating to the conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption;

Article XX is approached as a two-tiered test. (91) First, the defending member pleads one of the relevant exceptions in the subparagraphs of Article XX, and a sort of balancing inquiry take place against that interest. Second, the analysis shifts to the prohibitions found in the lead paragraph (called the "chapeau") of Article XX, such that a measure that passes scrutiny under an Article XX exception could still violate GATT if it is a "disguised" or "arbitrary or unjustifiable" restriction on trade. The overall process has inherent requirements of proportionality: What is "necessary to protect ... life or health", or what is "relating to ... conservation" of natural resources, or what is "unjustifiable discrimination" always requires an evaluation of the legitimacy of competing interests and to balance them against the impugned measure.

Unfortunately the Article XX jurisprudence lacks an explicit test as found in Oakes, O'Brien or FEDESA. Previous cases have held that "disguised" restrictions on trade are essentially colourable restrictions, like MEQRs in Community law. (92) Similarly, "unjustifiable discrimination" means discrimination that is disproportionate, either because it is overbroad, or too burdensome on the free trade system. (93) Conversely, a "necessary" measure is one that achieves proportionality between its legitimate objectives and preservation of life or health. (94) This is very clearly stated in the U.S. Tariff Act case:

   [A] contracting party cannot justify a measure inconsistent with
   another GATT provision as 'necessary' ... if an alternative measure
   which it could reasonably be expected to employ and which is not
   inconsistent with other GATT provisions is available to it. By the
   same token, in cases where a measure consistent with other GATT
   provisions is not reasonably available, a contracting party is
   bound to use, among the measures reasonably available to it, that
   which entails the least degree of inconsistency with other GATT
   provisions ... [This] does not mean that a contracting party could
   be asked to change its substantive ... law or its desired level of
   enforcement of that law, provided that such law and such level of
   enforcement are the same for imported and domestically-produced
   products ... (95)

The requirement to use that measure "which entails the least degree of inconsistency with [the] GATT" is reminiscent of the "least impairing" or "no more extensive than necessary" standards of Canada's Oakes (96) or America's Central Hudson (97) tests. Similarly, the statement that members would not "be asked to change ... [their] desired level of enforcement" seems to rule out judgments such as the ECJ's in Danish Bottles, where the effects of a measure on the trade system were fatal.

The Thai Cigarettes case considered whether measures to restrict the trade in tobacco were "necessary" to protect human health. (98) Thailand's government discouraged smoking by imposing taxes, totally banning tobacco advertising, and maintaining a state-run tobacco monopoly whose regulations effectively banned imported cigarettes. The American government challenged the taxes and import ban. While the Thais conceded that the import ban imposed a quantitative restriction, they argued that it was "necessary to protect human ... life or health", as Article XX(b) allows. For assistance on this point, the GATT Panel invited the World Health Organization (WHO) to "present its conclusions on technical aspects of the case, such as the health effects of cigarette use and consumption, and on related issues for which the WHO was competent." (99)

The WHO's submissions were mixed, but militated toward caution. The WHO cited the usual health problems of smoking: cancers, cardiovascular disease, stillbirths, etc. But the WHO also upset one of the fundamental assumptions of the case: that Thai and American cigarettes were like products. The WHO submitted that Thai cigarettes were harsher than their American counterparts, because where the Thai product used crude tobacco leaves, American cigarettes achieved their smoothness by using blended tobaccos, adulterated with additives whose health effects were unknown. Because it was "enormously complex and expensive" to research the biological effect of these additives, the WHO explained there was "no scientific evidence that one type of cigarette was more harmful to health than another." But as to their sociological effect, the WHO concluded that American cigarettes were more dangerously addictive, because their smoothness "made smoking ... very easy for groups who might not otherwise smoke, such as women and adolescents", and perpetuated "the false illusion among many smokers that these brands were safer than ... native ones." (100) Since such perceptions drive tobacco use, the WHO concluded that American cigarettes were dissimilar and more dangerous than Thai cigarettes.

The WHO's case was based on real experience. In Latin America and Asia, the WHO found that "the opening of closed cigarette markets dominated by a state tobacco monopoly resulted in an increase in smoking". (101) The WHO also cautioned against placing reliance on advertising bans, because "multinational tobacco companies ... routinely circumvented [such] restrictions ... through indirect advertising and a variety of other techniques." It concluded:

   If the multinational tobacco companies entered the [Thai] market,
   the poorly-financed public health programmes would be unable to
   compete with the marketing budgets of these companies, as had been
   the case in other Asian countries whose markets had been opened. As
   a result, cigarette consumption and, in turn, death and disease
   attributable to smoking would increase. (102)

As the global expert agency on health, the WHO's opinion deserved considerable deference. However, the Panel dismissed the WHO's evidence almost entirely. It found Thailand's restrictions were not "necessary", applying the U.S. Tariff Act case standard:

   The Panel ... examined whether the Thai concerns about the quality
   of cigarettes consumed in Thailand could be met with measures
   consistent, or less inconsistent, with the General Agreement ... It
   noted that other countries had introduced strict,
   non-discriminatory labeling and ingredient disclosure regulations
   which allowed governments to control, and the public to be informed
   of, the content of cigarettes. [This] coupled with a ban on
   unhealthy substances, would be an alternative consistent with the
   General Agreement ...

   The Panel then considered whether Thai concerns about the quality
   of cigarettes consumed in Thailand could be met by measures
   reasonably available to it and consistent, or less inconsistent,
   with the General Agreement. The Panel first examined how Thailand
   might reduce the demand for cigarettes ... The Panel noted the view
   expressed by the WHO that the demand for cigarettes, in particular
   the initial demand for cigarettes by the young, was influenced by
   cigarette advertisements and that bans on advertisement could
   therefore curb such demand ...

   The Panel then examined how Thailand might restrict the supply of
   cigarettes ... [The] Panel could not accept the argument of
   Thailand that competition between imported and domestic cigarettes
   would necessarily lead to an increase in the total sales of
   cigarettes and that Thailand therefore had no option but to
   prohibit cigarette imports.

   In sum, the Panel considered that there were various measures
   consistent with the General Agreement which were reasonably
   available to Thailand to control the quality and quantity of
   cigarettes smoked and which, taken

   together, could achieve the health policy goals that the Thai
   government pursues ... The Panel found therefore that Thailand's
   practice ... [was] not "necessary" within the meaning of Article
   XX(b). (103)

In contrast to cases such as Canada's Irwin Toy, where scientific uncertainty is a reason to reduce the intensity of judicial review and accord deference, the Panel in Thai Cigarettes is far from deferential. The Panel even goes to creative lengths to conjure alternative ways that Thailand could have impaired trade less. Unfortunately, none of those alternatives are workable: "bans on advertisement", or "a ban on unhealthy substances", are options that the WHO's expert evidence rejected. Recall that the WHO's submission that advertising bans were "routinely circumvented", and that singling out individual unhealthy additives in cigarettes depended on scientific studies that were "enormously complex and expensive".

Thai Cigarettes epitomizes the absurdity of taking the minimal impairment proportionality test too literally--and too far. Courts can always find a way that the law could have impaired a competing interest a little less, which gives them limitless discretion to strike down laws.

GATT judgments after 1995--decided under the new, WTO dispute resolution system--are somewhat more analytical in their application of Article XX. In the Sea Turtles case, the WTO Appellate Body had to decide whether the U.S. could prohibit imported shrimp, caught by techniques that the U.S. believed killed an excess of endangered sea turtles. (104) Beginning in 1990, American fishing boats were required to use "turtle excluder devices" (TEDs) to reduce the turtle bycatch. To promote similar conservation efforts abroad, the U.S., between 1991 and 1996 passed Section 609 of the Endangered Species Act (105) and regulations, banning shrimp imports unless two conditions were met. First, countries had to be certified, either because their fishing environment posed no risk of turtle bycatch, or because they mandated TEDs similarly to the U.S. Second, each import into the U.S. had to bear a Shrimp Exporter's Declaration attesting that the shrimp came from a certified country. Together, these provisions sought to impose American-style conservation standards extraterritorially.

India, Malaysia, Pakistan and Thailand challenged the U.S. shrimp import ban as violating GATT Article XI:1. The U.S. countered that its measures were justified under Article XX(g), as "relating to the conservation of an exhaustible natural resource. In a murkily reasoned judgment the Panel disagreed, and the U.S. lost. (106) The Appellate Body improved upon their reasoning in the appeal by adopting clear language for proportionality testing, based on a two-tiered approach to Article XX: a "provisional justification" under the Article XX(g) exception first, followed by a chapeau analysis. (107)

The Appellate Body had little difficulty concluding that endangered sea turtles were "exhaustible natural resources" for the purposes of Article XX(g), (108) or that the U.S. measures were "relating to the conservation of [exhaustible natural resources]". (109) The Appellate Body also accepted that the extraterritorial obligations U.S. law imposed in order to reduce the turtle bycatch were "directly connected with the policy of conservation of sea turtles." (110) The Appellate Body continued:

   In its general design and structure ... Section 609 is not a
   simple, blanket prohibition of the importation of shrimp imposed
   without regard to the consequences (or lack thereof) of the mode of
   harvesting employed on the incidental capture and mortality of sea
   turtles. Focusing on the design of the measure here at stake, it
   appears to us that Section 609, cure implementing guidelines, is
   not disproportionately wide in its scope and reach in relation to
   the policy objective of protection and conservation of sea turtle
   species. The means are, in principle, reasonably related to the
   ends. The means and ends relationship between Section 609 and the
   legitimate policy of conserving an exhaustible, and, in fact,
   endangered species, is observably a close and real one ...
   In our view, therefore, Section 609 is a measure "relating to" the
   conservation of an exhaustible natural resource within the meaning
   of Article XX(g) of the GATT.

The Appellate Body used classic proportionality language: Conserving endangered species is a "legitimate policy"; Section 609 enjoys a "close and real" relationship to that policy, without being "disproportionately wide in its scope and reach". Under the Canadian Oakes or American Central Hudson test, this language would have sufficed. However, under GATT law, this was only enough to "provisionally justify" Section 609 under Article XX(g), and to advance to the second tier of inquiry: the Article XX chapeau. (111) The Appellate Body wrote:

   We consider that [the chapeau of Article XX] embodies the
   recognition on the part of WTO Members of the need to maintain a
   balance of rights and obligations between the right of a member to
   invoke one or another of the exceptions of Article XX ... and the
   substantive rights of the other Members under the GATT ...

   The task of interpreting and applying the chapeau is, hence,
   essentially the delicate one of locating and marking out a line of
   equilibrium between the right of a Member to invoke an exception
   under Article XX and the rights of other Members under varying
   substantive provisions (e.g. Article XI) of the GATT. (112)

The chapeau pits the right of a state to use an exception, such as under Article XX(g), against the right of other states to export goods freely under GATT, such as under Article XI--and this is where the U.S. law collapsed. The Appellate Body found that the Section 609 system violated the chapeau, because it required other states to adopt American-style conservation standards (e.g. TEDs) as a condition of certification. As it wrote:

   We scrutinize first whether Section 609 has been applied in a
   manner constituting "unjustifiable discrimination between countries
   where the same conditions prevail". Perhaps the most conspicuous
   flaw in this measure's application relates to its intended and
   actual coercive effect on the specific policy decisions made by
   foreign governments, Members of the WTO. Section 609, in its
   application, is, in effect, an economic embargo which requires all
   other exporting Members, if they wish to exercise their GATT
   rights, to adopt essentially the same policy ... as that applied
   to, and enforced on, United States domestic shrimp trawlers ...

   [It] is not acceptable, in international trade relations, for one
   WTO Member to use an economic embargo to require other members to
   adopt essentially the same comprehensive regulatory program, to
   achieve a certain policy goal, as that in force within that
   Member's territory, without taking into consideration different
   conditions which may occur in the territories of those other
   members ... (113)

The Appellate Body's reasoning is puzzling. The chapeau prohibits only one version of inequality "unjustifiable discrimination between countries where the same conditions prevail" (114)--but the Appellate Body uses it anyway to strike down Section 609, because it neglects to consider "different conditions" prevailing between Members. Logically that makes no sense, so one assumes that the real irritant for the Appellate Body was what it called "the most conspicuous flaw": that U.S. law coerced other countries into following its lead, as a condition precedent to exercising their trade rights.

Sea Turtles therefore presents an even more questionable analysis than Danish Beer. In both cases, a narrowly tailored measure with legitimate objectives was struck down for having disproportionate effects on free trade. But where the ECJ was up-front and honest about doing this, the Appellate Body is strikingly dishonest, and hides behind tendentious reasoning. In some ways these results flow from the avoidance of an explicit, checklist-like method of proportionality testing, of the Oakes, Central Hudson or FEDESA kind. The result is a foolish two-tiered exercise, which is difficult to follow and allows a provisionally justified measure that is narrowly tailored and proportionate to fail for unjustifiable discrimination. What value the second tier serves is hard to understand, since discrimination is itself a cause of action in GATT law--as that is what National Treatment, under Article III:4, forbids. It is paradoxical that Article XX is said to override the substantive trade requirements of GATT law, as the substantive requirement of National Treatment remains firmly embedded within the two-tier balancing exercise for Article XX.

The GATT jurisprudence is poorly designed. Not only is it outrageously undeferential, as Thai Cigarettes exemplifies, but it lacks an explicit or even coherent framework for proportionality testing and balancing, as Sea Turtles proves. Without proper tests, Article XX is a treacherous and arbitrary gauntlet to run, one seemingly dedicated to the ascendancy of free trade over all competing interests. In short the GATT is law nearing its most dishonest.

PART III--DISCUSSION:

1. Ubiquity, Consistency, and Explicitness in Judicial Review

While the foregoing examples do not paint a harmonious view of balancing and proportionality, every jurisdiction uses it, understanding that they would otherwise be unable to balance competing interests.

The American experience, most notably the commercial expression cases, illustrates why balancing competing interests is far preferable to the alternative of using dispositive rules. (115) A dispositive rule amounting to "commercial speech is not First Amendment speech", is not consistent with the holding in Fox--no amount of distinguishing past cases, or tergiversating about the rule, would make that judgment possible. Even those who feel commercial speech is unworthy of protection cannot be categorical about their opposition. Would Americans not feel provoked if their constitution failed to protect schools, churches or NGOs advertising a bake sale? Or prevented charities advertising for relief donations in the wake of humanitarian disasters?

These examples are admittedly the soft edge of commercial expression, but they are purely commercial all the same. A dispositive rule is a clumsy way to deny constitutional protection to marginal interests such as commercial expression, because it does so in a blanket fashion. Balancing and proportionality, however, allow nuance, and that is why their use should be ubiquitous in all cases involving competing interests.

But ubiquity is not enough. All sorts of jurisprudential train-wrecks occur where proportionality testing is not consistent, which is often the case when courts are too hasty and not explicit about the proportionality tests they are applying. A lack of consistency or clarity defeats law's most fundamental normative obligation legal certainty, knowing where the line between legality and illegality lies.

Of the courts surveyed, the Supreme Court of Canada is the most consistent about proportionality testing: the Oakes test is like a checklist and the Court reasons through each step explicitly. In my opinion, this is a good model. The Court may be limited in having only a single test, but it still has plenty of maneuvering room, as evident when the Court adjusts the intensity of judicial review based on social teleology of free expression. The farther away an utterance lies from that teleology (i.e. hate speech) the more intense the judicial review, and the less deference it is accorded. The solution is elegant, reasonable--and it works.

The United States Supreme Court, by contrast, is not so straightforward, juggling many inconsistent balancing and proportionality tests simultaneously. The jurisprudence is unclear as to whether balancing is done ubiquitously, and while they have been dormant for some time and seem motionless after R.A.V., the old dispositive rules may still be good law. There is a test for incidental infringements of expression (116); a test for "time, place and manner" restrictions (117); and yet another test for commercial expression. (118) The Court tells us in Fox that these tests are of a feather, but in fact they differ--sometimes in useful ways. Creative Non-Violence has a more specific inquiry (as opposed to a general "minimal impairment" inquiry) which asks whether a measure "leaves open ample alternative channels for communication of information". This is a very accurate litmus test for a "time, place and manner" restriction, because the lack of ample alternative channels suggests that the so-called "time, place and manner restriction" is more like a prior restraint on content, which is constitutionally much more dubious.

Therefore, the question remains, is American inconsistency, with its many specialized balancing and proportionality tests, better than Canada's greater consistency with the Oakes test? It depends. The American Court's multiple tests substitute a degree of particularization for judicial discretion. This has the advantage of bringing the discretion out into plain view, where it lives in the language of the test itself; but it has the disadvantage of littering the jurisprudence with a bevy of different balancing tests and dispositive rules, when a single test might suffice. The more tests and rules there are, the greater the risk of ending up with murky decisions, such as R.A.V. Therefore, while the American and Canadian approaches each have their pros and cons, both seem reasonable.

There is also the value of explicitness: When judgments balance competing interests but do not clearly articulate doing so, the law is difficult to understand and state with precision. The ECJ hardly dwells on proportionality testing at all: it often invokes the name of a proportionality test in a token way (perhaps FEDESA or Fromancais (119), then hurries on to judgment. The confusion this causes has inspired commentary on the uncertain meaning of "proportionality" in Community case law. (120) GATT law is even worse, where to this day Panels and the Appeal Body have yet to articulate explicit proportionality tests within the already murky two-tier balancing exercise. Thai Cigarettes and Sea Turtles, among other cases, demonstrate that proportionality reasoning is found in GATT case law--so why not state the applicable tests clearly?

To one trained in the common law tradition, which highly values precedent, non-explicit judgments seem unhelpful and dangerous. However, it is not entirely fair to expect EC and GATT law to reflect this common law preoccupation. ECJ and GATT judgments are consensus decisions; there are no separate dissenting or concurring reasons. Such a sparse judicial discourse would play havoc on the common law, yet works in other judicial systems that put less stock in precedent. Formally, neither Community nor GATT law have a rule of stare decisis, although there surely is a desire to promote internal consistency. (121)

2. Deference, or the subversion of law and sovereignty?

There remain a core set of concepts--legitimate objective, means related to the ends, minimal impairment--that appear in all the proportionality tests surveyed. From where does this commonality derive? It is not inherited, and it would be silly to trace a family tree and declare it the "descent of balancing". The better biological analogy is convergent evolution: a theory that maintains where environments are similar, so too may be the outcomes. 122 If the convergent evolution analogy holds for proportionality testing, the national courts should resemble each other, the trade courts should resemble each other, but there will be less in common between the two groups.

The Supreme Courts of Canada and the United States both give abundant deference to the lawmaker when legislation is shown to meet a pressing and substantial objective. Legislators decide how vigorously to go about fixing a problem, and the Courts do not question it. In the United States, Rock Against Racism decided that "the validity of time, place or manner regulations does not turn on a judge's agreement with ... [the] method for promoting significant government interests, or the degree to which those interests should be promoted. " (123) In Canada, Irwin Toy cautions that "if the legislature has made a reasonable assessment as to where the line [marking one set of claims from another] is most properly drawn ... it is not for the court to second guess". (124)

Succinctly put, the national courts do not question the threshold or level at which laws protect an interest. They accord this deference in two ways.

First, national courts dilute the literal meaning of "minimal impairment" proportionality. At face value, this is a draconian test indeed: It means of 100 possible measures to attain a legitimate objective, 99 may impair too much and be unconstitutional. A judge bent on striking down legislation could either demand evidence that none of the putative alternatives is less impairing, or could speculate in favor of an alternative of his or her own. As McLachlin J. (as she then was) of the Supreme Court of Canada put it:

   some deference must be paid to the legislators and the difficulties
   inherent in the process of drafting rules of general application. A
   limit prescribed by law should not be struck out merely because the
   court can conceive of an alternative which seems to it to be less
   restrictive, (125)

Or as Blackmun J. put it more colourfully:

   A judge would be unimaginative indeed if he could not come up with
   something a little less 'drastic' or a little less 'restrictive' in
   almost any situation, and thereby enable himself to vote to strike
   legislation down. (126)

There is a very good reason not to be so imaginative: When judges posit less impairing alternatives, the litigation stops being inter partes, and switches to a game of hypotheticals. The Canadian and American Courts recognize that fundamental rights are too important for such chicanery, and they stick to the record of evidence in proportionality testing, rather than encourage judicial speculation.

Second, national courts never strike down laws because of a perceived disproportionality between the law's legitimate objective and the law's undesirable effects. This meaning of proportionality is captured by the final step of the Oakes test: "there must be a proportionality between the effects of [a] measure...and the objective which has been identified as of 'sufficient importance'". Canadian courts have never relied solely on this part of the Oakes test to strike down laws (127)--and American judges have likewise refrained. (128) To do otherwise would transgress a recognized, if unwritten, boundary of judicial review in these countries.

Neither of these exercises of deference comes easily to the trade courts. Both the ECJ and the GATT Panels and Appellate Body apply the strict meaning of "minimal impairment", and occasionally strike down legislation because of its disproportionate collateral effects.

GATT law is replete with such instances. In Thai Cigarettes, it was a heroic act of judicial imagination which posited that alternative measures--banning harmful additives and banning cigarette advertisement--could stanch tobacco use while impairing trade less. The WHO gave expert evidence directly on point that these alternatives were not feasible, but that went unnoticed. If one assumes that the WHO is better situated than a panel of trade lawyers to assess the credibility of medical evidence, then the Panel's rejection of the WHO's submission implies the chilling thought that the Panel was untroubled to accept some extra smoking as a tolerable price for not hampering trade.

The tendency to curb legislation because of its disproportionate collateral effects also appears in the ECJ, although to a more muted degree. In Danish Bottles, the ECJ did not quibble with Denmark's objective of protecting the environment, but it took fright at the degree to which Denmark sought to protect the environment, because of the complications this posed to free trade. Much the same logic underlies German Beer, where the German desire to protect health was understood, but the ECJ struck down the precautionary prohibition on additives of uncertain safety because this inconvenienced free trade. Both the Danish and German approaches were narrowly tailored or minimally impairing. However, this was not the problem--it was simply that each court pursued its objective too aggressively.

The bottom line is that in the ECJ or GATT jurisprudence, health or environmental protection can override trade--as long as they don't override it too much. The Sea Turtles decision says it best: even a law that is "not disproportionately wide in its scope and reach in relation to ... [a] legitimate policy" may be struck down, if it breaches "a line of equilibrium between the right of a Member to invoke an exception under Article XX and the rights of other Members". (129)

It is extraordinary that the trade courts so readily enter this terrain, where the national courts absolutely fear to go. It makes the argument completely untenable that proportionality testing embodies, "The Ultimate Rule of Law". Professor Beatty takes an almost Panglossian view of proportionality, and argues that "Proportionality makes [legal] pragmatism the best it can possibly be." (130) But as we see, the trade courts have a strikingly different understanding of proportionality than the national courts--and logically there can't be two "bests". I agree with Professor Beatty that proportionality is found in many legal systems, and that it is special; but he exaggerates to say that it has an ultimate transcendence setting it apart from all other rules of law. It is an attractive, comforting thought, but one that is just wrong enough to be misleading and dangerous.

The aggressive use of proportionality testing, as the trade courts do it, is a cause for worry. When a national court strikes down a law for being more impairing of a competing interest than necessary, and thus disproportionate in its chosen means, the legislators can go back to the drawing board and come up with a more narrowly tailored law--and legislatures often do this. (131) But when a trade court strikes down a law for being too effective and thus disproportionate in its collateral effects, the legislators cannot go back to the drawing board. Instead the decision permanently diminishes the power of the legislators.

Return to Danish Bottles to consider this problem further. The ECJ understood that Denmark's bottle law succeeded in achieving a "maximum rate of re-use" (the evidence said 99%), and therefore "a very considerable degree of protection of the environment". But when the law was struck down for being narrowly tailored but overzealous, what kind of law should the Danish legislature next attempt? A modest and laxer law that would achieve 95% re-use? Laxer still and 60% re-use? Or maybe Denmark should discontinue reusing and adopt recycling instead? These scenarios are not at all hypothetical, and the Danish legislature had just this problem. Not content with the one victory, in 1999 the European Commission decided to sue again to force Denmark to accept recyclable, not just reuseable, containers. (132) By 2002 the prospect of defeat was so clear that Denmark settled, and repealed its bottle laws so that any container was allowed, including single-use disposable ones. (133) The ECJ's version of proportionality did not just slightly weaken Denmark's bottle law or marginally usurp the Danish legislature's jurisdiction over it; ultimately it destroyed both.

While this sort of outcome is not necessarily surprising on the European continent (as proportionality there is a cause of action) it poses a meaningful question in Canada and the rest of the common law world: By signing trade treaties, such as the GATT/WTO, are our legislatures and democracies knowingly submitting to a system where courts and not legislators decide on the degree to protect the environment and human health? Until now, legal theory held that legislators are best situated to integrate the views of different constituencies. 134 It is naive to ignore the empirical fact that GATT jurisprudence is almost without deference and disposed to rulings that promote free trade. Robert Housman calls it free trade uber alles. (135)

A similar caution extends to the WTO: Look no farther than the Appellate Body's curious excesses of proportionality testing to understand why environmentalists, for whom Article XX matters so much, are moved to fury and wish to abolish the WTO. There has been no shortage of commentary that GATT decisions appear trade-motivated and unyielding to competing interests. (136) To date, environmentalists have lost eight of nine cases under the GATT. (137) Public health cases, mostly under the SPS Agreement, are doing only slightly better. (138)

Unfortunately the WTO is hardly responsive to the analysts, who criticize its democratic credentials. The WTO's publicists dismiss the assertion that it weakens the rule of law, and actually tout it as a model for constitutionalism. As Ernst-Ulrich Petersmann has written:

   [The] 1994 WTO Agreement, and its mandatory worldwide dispute
   settlement system, are milestones on the long and winding road to
   worldwide economic freedom, consumer welfare and democratic peace.
   The Uruguay Round approach to reforming the old 'GATT h la carte
   system' through an integrated WTO legal and dispute settlement
   system could serve as a model also for a similar reforms of other
   multilateral agreements, including the UN Charter. Such a
   progressive 'constitutionalization' of discretionary foreign policy
   powers of national governments could strengthen democracies for the
   benefit of individual citizens, and of their common 'public
   interest' in the protection of equal liberties, rule of law,
   democracy and open markets across frontiers. (139)

The hypothesis that WTO jurisprudence would inspire "equal liberties, rule of law, [and] democracy" worldwide cannot be taken seriously. The reinfication of those constitutional values in America or Canada depended on a style of proportionality testing that is principled, nuanced and deferential--a way that WTO jurisprudence is brave, new and dangerous, and must adapt rather than undermine the equal liberties, rule of law, and democracy that have defined common law countries since the Age of Enlightenment.

3. How different jurisprudence philosophy affects deference

Where do the real differences between national and trade courts come from? How do they, with seemingly similar proportionality tests, appear to reach such different results?

There are two basic answers to these questions. The first answers is of philosophy: the national courts lean toward deontology, while the trade courts lean toward utilitarianism. The second answer is of institutions: the national courts belong to sovereign governments, while the trade courts belong to multinational organization.

It may seem odd to call on philosophy to explain proportionally. Paul Craig argues persuasively how philosophy can shape one's conceptions of proportionality, even if the concepts of proportionality are constant:

   Those with distinctive political philosophies may accept a general
   concept of proportionality, but their particular conception thereof
   could differ considerably. For example, assuming for the sake of
   argument that some notion of proportionality might be said to
   underlie discrimination cases, the content that should be given to
   this notion will differ depending on whether the commentator is a
   utilitarian, a Rawlsian liberal or a modern communitarian. Whether,
   for example, the provision of subsidized bus fares should be viewed
   as disproportionately favouring user interests versus ratepayer
   interests is a question which would be approached very differently
   by an advocate of early UK pluralism and an adherent of the more
   market-based pluralism ... (140)

It is this difference between conception and concept that makes the simple proportionality test so greatly malleable; philosophy is its starting point.

Deontology is rooted in Immanuel Kant's "categorical imperative". As the adjective suggests, the imperative has a single maxim--behave in such a way that you could wish your actions to be universal law. (141) Implied within that maxim is a duty of respect--to treat others as having ends of their own, and not as a means to someone else's ends. (142)

Plainly, the categorical imperative is reflected in the constitutions of democratic states. Human rights constitutions are themselves just collections of maxims, governing the relationship between the state and its citizens. Most such rights aim to ensure that the citizen is the end of governance itself, and not merely a means for the state's own ends. The famous words of Lincoln's Gettysburg Address "government of the people, by the people, for the people"--exude a democratic, Kantian morality. (143)

An important but easily confused fact about Kantian deontology is that while the imperative is called categorical, the emphasis it places on the individual is not. On the contrary, deontology's commandment is that one should behave so as to accommodate others. Perhaps to lessen this possible confusion, many prefer the alternative phrasing of deontology by John Rawls that better considers the rights of others: "Each person is to have an equal right to the most extensive basic liberty compatible with a similar liberty for others." (144)

Rawls' principle is a fine metaphor for balancing in national courts, demanding a "fit" between competing interests: one that gives each person's interest the fullest scope possible, subject to analogous interests of others. Where legislators weigh the competing interests and arrive at a measure that is minimally impairing, Rawls' principle is satisfied and the democratic process has made the "fit". Courts are loath to meddle, and will defer if the legislature's solution is rationally connected to its objective and not obviously overbroad, as in America's Rock Against Racism or Canada's Irwin Toy decision.

Deontology, however, is not an accurate metaphor for the trade courts. Rather, the weltanschauung of international trade emphasizes the value of allocating resources and factors of production, which is clear in the foundational trade theory. David Ricardo's principle of comparative advantage, which espouses that nations produce those goods for which they enjoy advantages, to trade with other states operating on the same principle, thereby enhancing the well-being of all compared to autarky. (145) Frequently this is expressed in a maritime analogy: states are like boats on a sea of commerce, and trade is like the tide that rises and lifts all the boats at once. It is the aggregate enhancement of wealth that is the impetus for freer trade. (146)

The philosophy of utilitarianism captures this wealth-maximizing drive. Jeremy Bentham, utilitarianism's extremist father, argued that correct decisions were those which maximized pleasure, the only criterion that matters. "Quantity of pleasure being equal," Bentham notoriously wrote, "pushpin is as good as poetry". (147) Maximizing pleasure, in Benthamite utilitarianism, is an apt metaphor for maximizing wealth in free trade.

However, the utilitarian ethic leaves little room when competing interests get in the way. Joseph Hume once retorted to Bentham:

   If it is really true that the will of the minority must bow before
   the will of the majority, and that the interest of the minority
   must be sacrificed to the interest of the majority, does it not
   appear legitimate that in a society composed of thirty individuals,
   twenty-nine should agree to roast and eat the thirtieth, if they
   find pleasure in so doing? (148)

Hume's point is eloquently made: The utilitarian ethic is inflexible, and if carelessly taken to its logical conclusion, it can lead to preposterous results. A preoccupation with utilitarianism is likely the reason why trade courts lack deference, and also why they look severe when they strike down laws which have disproportionate collateral effects.

While both philosophies hold themselves to be good for society as a whole, utilitarianism produces the more uncompromising results in the infrequent cases that push it to its logical limits. Utilitarianism accommodates competing fights poorly; Bentham believed this himself, and wrote, "I know of no natural rights," he wrote, "except what are created by general utility: and even in that sense it were much better the word were never heard of." (149) His contempt for the constitutionalization of human rights was primordial, and he wrote this to a leader in the French Revolution:

   I am sorry that you have undertaken to publish a Declaration of
   Rights. It is a metaphysical work--the ne plus ultra of
   metaphysics. It may have been a necessary evil,--but it is
   nevertheless an evil ... Let the articles be what they may, I will
   engage they must come under three heads--1. Unintelligible; 2.
   False; 3. A mixture of both ... You can never make a law against
   which it may not be averred, that by it you have abrogated the
   Declaration of Rights; and the argument will be unanswerable. (150)

One could say, a bit flippantly, that utilitarianism does not play well with others. Thus it should be no surprise when laws that derogate from wealth formation (e.g. by protecting the environment) come under the suspicion of trade courts with their utilitarian values, and are subsequently struck down, not for overbreadth but for a perceived disproportionate effect on the trade system.

Not everyone agrees with the distinction between deontology and utilitarianism that I argue here. Trade scholars argue that both underlie their field of study. Petersmann understands this fusion:

   A rights-based approach is firmly rooted not only in moral
   philosophy (such as Immanuel Kant's "categorical imperative") and
   in legal principles of justice (such as John Rawls' theory of
   justice). It is also vindicated by the economic-utilitarian,
   instrumental justification of liberties and actionable property
   rights as preconditions for the proper functioning of both economic
   markets as well as political markets, and for maximizing individual
   autonomy, human well-being, economic efficiency and social welfare
   in a free society. (151)

This has to be one of the more confused passages ever written about philosophy. Petersmann is correct that a "rights-based approach" could mean human rights in the deontological or Rawlsian sense, or could mean economic rights in the Benthamite or utilitarian sense. But commingling the two? Deontology does not have much to say about "maximizing ... economic efficiency", any more than utilitarianism has to do with "maximizing individual autonomy", and Petersmann is wrong to run these concepts together. Further, the only modern societies that have attained the summit of Petersmann's argument--"maximizing individual autonomy, human well-being, economic efficiency and social welfare in a free society"--are those societies where legal systems are essentially democratic and whose judges eschew utilitarian militancy, in favor of a more deferential balancing of interests that is typical of deontology.

4. Institutional location and the practice of balancing

The courts in this study are located in markedly different institutions. They may exist within a government that is sovereign and democratic (Canada and the USA), or exist outside of any government and be creatures of international law (the ECJ and GATT/WTO). The difference of institutional location seems to affect the deference the courts employ.

To appreciate why, consider that every court has a dilemma of torn allegiance. On the one hand, the court must serve the subjects of the law, those who are governed by it. On the other, the court must serve the authors of the law, those whose legislation they interpret. A court is less likely to strike down laws passed by a legislature that forms part of its own government, and more likely to strike down laws passed by legislators elsewhere.

In Canada, the system of government is traditionally thought of as giving Parliament omnipotent, sovereign, and self-correcting powers through representative democracy. This is the classical view of A.V. Dicey, and for a very long time it meant that Canada's courts could not correct Parliament. Dicey's model is of course less appropriate since the inception of the Charter in 1982, and the Supreme Court of Canada can now correct Parliament's errors and strike down laws that violate fundamental rights. Constitutional traditions, however, are slow to change in a small country like Canada, and it is perhaps understandable that in the century before 1982, the Court grew a real institutional closeness to Parliament which is remains evident in the deference that Parliament receives today in constitutional balancing cases.

The situation is not much different in the United States, where by long tradition government institutions form "checks and balances" against one another. Courts in America have always held the Congress and the Executive accountable. Americans do not talk about Dicey, but they also believe that Congress and the Executive are often self-correcting. Indeed, the overall structure and makeup of American and Canadian democratic institutions is similar enough (especially with respect to the role of their courts) that one would predict only modest differences between how courts in the two systems make use of deference--and this is exactly what is observed. (152)

However, there is marked difference in institutional location between the national courts and the trade courts, which explains in part why the trade courts are so much less deferential. Neither the European Community nor the World Trade Organization cohabits with a truly authoritative, democratically elected legislature.

The sibling institutions of the ECJ are the European Council, the Council of Ministers, the European Commission, and the European Parliament. One might imagine that there is a kind of Diceyan relationship between the European Parliament and the ECJ, but that is incorrect, because despite being directly elected by European citizens, the European Parliament is not terribly parliamentary. It cannot initiate legislation, and at most, it can veto or amend laws initiated by the European Commission. (153) The ECJ therefore has only a weak, incomplete proximity to a democratic legislative institution.

What the ECJ does have, however, is an important document in the Treaty of Rome. According to Pierre Pescatore, who is one of the most famous judges to have sat on the ECJ, the European institutions such as the ECJ themselves give life and transformation to the treaty, and have made it an ersatz constitution. Pescatore writes:

   En effet, le statut constitutif de toutes les organisations
   intemationales est represente, a l'origine, par une convention
   multilaterale; dans la suite, a partir de la mise en place des
   institutions, ce caractere contractuel s'estompe et c'est desormais
   le charactere institutionnel qui prime. La convention multilaterale
   se mue alors, pour ainsi dire, en constitution. (154)

When an international treaty becomes constitutionalized in this way, the ECJ acting as its interpreter--cannot help but assume a new, powerful, and more autonomous quality. Pescatore has said that, "the European Community [now the European Union] is nothing other than a form of corporate life of States". (155) The ECJ is not really part of a democracy, but a bureaucracy, and this is probably why it is not keen to defer to the laws of its democratic member states.

The GATT Panels are even more unplugged from democratic principles than the ECJ. GATT and the rest of the WTO treaties are born from diplomatic negotiations during circumstances of questionable transparency. The persons who sit on WTO dispute panels are chosen on an ad hoc basis by the Director-General's administrative staff, drawing from a shortlist comprised of trade lawyers, academics, bureaucrats and diplomats. (156) WTO panelists are basically employees or contractors of the organization, and do not have the security of tenure--and resultant judicial independence--that real judges do. The WTO is pure bureaucracy and diplomacy, and only by a very distant appeal to the fact that the majority of its members are democracies does it have any democratic credentials at all.

The rules of judging in a bureaucracy are not the same as in a democracy. Imagine for example a scenario where the Director-General of the WTO grows chary of appointing a panelist who, in past decisions, declined to uphold a far-reaching interpretation of the GATT, or who upset a powerful WTO member by his or her decision. Most panelists probably want to keep their job, and will therefore be under some pressure to conform to WTO secretariat thinking. That the WTO secretariat also helps panelists to draft their judgments is only another opportunity to exert influence. Deference to anything outside the small world of the WTO secretariat itself would be unexpected.

PART IV--CONCLUSION:

This study leaves two impressions: one reassuring, the other unsettling.

All legal systems must grapple with cases that pit one interest against another irreconcilably. The only feasible solution is to pick the one interest that trumps the other and to stipulate that the inferior interest should be infringed as little as possible in the process--the very essence of proportionality.

It is reassuring that tribunals in such diverse jurisdictions and settings have all converged on solving this problem through proportionality tests based on minimal impairment. The consistency is not accidental, but speaks to a kind of inherent rationality in judicial decision-making. This degree of consistency is not to be found in other areas of law (e.g. negligence law, even though blameworthy accidents are as old as civilization itself). At this superficial level, Beatty is correct, and proportionality does resemble "The Ultimate Rule of Law".

But whatever comfort one may draw from this fact, the great malleability of the proportionality principles is worrisome. Proportionality testing is both a shield and a sword--it can defend legislation or strike it down--depending on the intensity of judicial review. This is highly relevant when thinking of a legal system that is not one's own. A lawyer who is well schooled in the common law could find his or her knowledge of proportionality testing badly misleading, when trying to conceive how a trade law case involving proportionality testing may be decided. We see this most clearly in the willingness of judges in EC or GATT law to be inventive about positing measures that are less impairing.

It is necessary that legal observers keep their eyes open to these differences, for two reasons.

First, there is nothing to say that the common law's more deferential view on proportionality testing is always superior. There may be salutary effects in expanding administrative law doctrine so that disproportionate laws are actionable as such, as they are in the civil law systems of continental Europe. (157) There may also be merit in requiring proportionality of the effects as an aspect of constitutionality. In Canada, the leeway to scrutinize laws based on the proportionality of their effects already exists in the last step of the Oakes test. In the United States, the law of balancing does not even mention proportionality of the effects. It would take an extraordinary case and set of facts to tempt common law judges across this Rubicon. However, since judicial review has been more aggressive in the past (most particularly in the discredited Lochner era) it is not unthinkable that it could be so again, and proportionality testing is one way that it could happen.

Second, democratic legislators must be realistic about the almost total lack of deference at international trade courts, if they wish their legislation to survive such challenges. Those who draft laws tend to observe their national jurisprudence and do not often pay attention to trade law, much less the subtleties by which trade courts rule. Legislatures are not prepared for the budgetary consequences of passing laws that will pass muster with the trade courts. For example, the WTO Appeal Body has struck down an American law, largely because the U.S. government had found an alternative that impaired trade less, but which Congress dismissed as too costly. (158) As it wrote:

   The fact that the United States Congress might have intervened, as
   it did later intervene, in the process by denying funding, is
   beside the point: the United States, or course, carries
   responsibility for actions of both the executive and legislative
   departments of government. (159)

A decision of this kind means that a country wishing to meet its GATT obligations may have to rank those obligations among its highest fiscal priorities. Surely that is either a source of frustration or injustice waiting to happen, when the WTO includes very rich and very poor countries. Switzerland might be able to spend a king's ransom making its laws GATT-compliant--but how about Swaziland?

The answer to that, of course, hangs in a wobbly balance.

(1) See National Socialist Party of America v. Skokie, 432 US 43 (1977).

(2) General Agreement on Tariffs and Trade (1947), as amended online: WTO [legal texts--Marrakesh Agreement <http://www.wto.org/english/docs_e/legal_e/gatt47_01_e.htm> for further discussion see Part II--Section 4, The Adjudicative Panels of the WTO, below.

(3) See generally Plessy v. Ferguson, 163 US 537 (1896); Brown v. Board of Education, 347 US 483 (1954).

(4) Adapted, with modifications by the author, from Paul P. Craig, Administrative Law, 3rd ed. (London: Sweet & Maxwell, 1994) at 414-15.

(5) See Joel Bakan, Just Words: Constitutional Rights and Social Wrongs, (Toronto: University of Toronto Press, 1997), particularly his discussion of the Jones case in c. 2. The "truth" and "trust" models herein are borrowed from Professor Bakan.

(6) See C. Fried, "Two Concepts of Interests: Some Reflections on the Supreme Court's Balancing Test" (1963) 76 Harv. L. Rev. 755; R. Nagel, "Legislative Purpose, Rationality and Equal Protection" (1972) 82 Yale L.J. 123.

(7) See Bakan, supra note 5.

(8) See especially D.M. Beatty, The Ultimate Rule of Law (New York: Oxford University Press, 2004).

(9) Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act, 1982 (U.K.), 1982 c.11 [Charter].

(10) Nevertheless "prescribed by law" is important, and means that a limitation on rights implemented other than through the substantive or common law can never be justified. Even before the Charter, state action limiting rights other than by prescription of law was ultra vires for lack of jurisdiction. See Roncarelli v. Duplessis, [1959] S.C.R. 121; Short v. Poole Corporation, [1926] Ch. 66.

(11) See Irwin Toy Ltd. v. Quebec, [1989] 1 S.C.R. 927 at para. 42 [Irwin Toy]:

   We cannot ... exclude human activity from the scope of guaranteed
   free expression on the basis of the content or meaning being
   conveyed. If the activity conveys or attempts to convey a meaning,
   it has expressive content and prima facie falls within the scope of
   the guarantee ... The content of expression can be conveyed through
   an infinite variety of forms of expression: for example, the
   written or spoken word, the arts, and even physical gestures or
   acts ... [however] certainly violence as a form of expression
   receives no such protection.

(12) Ibid. at paras. 48-52; see also Ramsden v. Peterborough (City), [1993] 2 S.C.R. 1084.

(13) [1986] 1 S.C.R. 103 [Oakes].

(14) Ibid._at paras. 69-70 [emphasis added, footnotes omitted].

(15) P. Hogg, Constitutional Law of Canada, 3rd ed. (Carswell, Scarborough, 1992) at 866.

(16) Even the Court decries, "the view, unfortunately still held by some commentators, that the proportionality requirements established in Oakes are synonymous with, or have even superceded, the requirements set forth in s. 1" in RJR-MacDonald Inc. v. A-G Canada (1995), 127 D.L.R. (4th) 1 at para. 62 [RJR-MacDonald]. Yet habitual resort to s. 1 carries on; see La Forest J., dissenting at para. 46.

(17) There are cases where the Court has found a measure fails the second proportionality inquiry, but this is always obiter dicta, because the measure failed on the earlier steps of the Oakes test too. See e.g. Vriend v. Alberta, [1998] 1 S.C.R. 493; and M. v. H., [1999] 2 S.C.R. 3.

(18) The likelihood that a rights-impairing measure is disproportionate follows on how one first characterizes the objective for which the measure is taken. If the objective is characterized narrowly, so that it is tautologous with the measures to achieve the objective, then the "fit" between the objective and the measures is perfect, and always minimally impairing. Justice O'Connor of the United States Supreme Court identifies this subterfuge nicely, "the [respondent] has taken the effect of the statute and posited that effect as the State's interest. If accepted, this sort of circular defense can sidestep judicial review of almost any statute, because it makes all statutes look narrowly tailored." Simon & Schuster, Inc. v. Members of New York State Crime Victims Board, 502 US 105 at 120 (1992) [Simon & Schuster].

(19) I paraphrase Dickson C.J.C.'s statement in Irwin Toy, supra note 11 at 976.

(20) (1990), 61 C.C.C. (3d) 1 [Keegstra].

(21) Ibid. at 51.

(22) The objective is stated ibid. at 38.

(23) Ibid. at 65.

(24) See RJR MacDonald, supra note 16. In this case, a narrow majority struck down a ban on tobacco advertising as being overbroad, for the reason that it prohibited all forms of advertising, including those whose abolition would not necessarily stanch smoking.

(25) Irwin Toy, supra note 11 at para. 80.

(26) Ibid.

(27) Ibid. at para. 75.

(28) Ibid. at paras. 80-81.

(29) Ibid. at para. 80.

(30) See Bakan, supra note 5.

(31) See the reasons of Me Lachlin J. (majority) and La Forest J. (minority) in RJR-MacDonald, supra note 16, wherein the 5:4 split of the Court arose from a disagreement as to how much deference should be applied in the proportionality test. The case concerned a total ban on tobacco advertising. Both the majority and dissent agreed that such a ban was aimed at protecting vulnerable persons, and was conceived in a realm of some scientific uncertainty, but where the minority would have deferred to Parliament's choice of a total ban, the majority struck down the total ban as disproportionate, because the state failed to demonstrate the total ban was more effective than a partial ban on only certain forms of advertising. The decision is one of the most controversial in Canadian constitutional law. See M. Jackman, "Protecting Rights and Promoting Democracy: Judicial Review Under Section 1 of the Charter" (1996) 34 Osgoode Hall L.J. 661.

(32) Fiske v. Kansas, 274 US 380 (1927).

(33) Beauharnais v. Illinois, 343 US 250 (1952).

(34) Valentine v. Chrestensen, 316 US 52 (1942).

(35) Chaplinsky v. New Hampshire, 315 US 568 at 571 (1942), where the Court ruled that the appellations "'God damned racketeer" and "damned Fascist" did, in the circumstances, tempt a breach of the peace:

   There are certain well-defined and narrowly limited classes of
   speech, the prevention and punishment of which have never been
   thought to raise any constitutional problem. These include the lewd
   and obscene, the profane, the libelous, and the insulting or
   'fighting' words ... It has been well observed that such utterances
   are no essential part of any exposition of ideas, and are of such
   slight social value as a step to the truth that any benefit that
   may be derived from them is clearly outweighed by the social
   interest in order and morality.

(36) Schenck v. U.S., 249 US 47 (1919).

(37) "[A]bove all else the First Amendment means that government has no power to restrict expression because of its message, its ideas, its subject matter, or its content." in Police Dept. of Chicago v. Mosley, 408 US 92 at 95 (1927), further citing a half-dozen authorities for this proposition.

(38) Konigsberg v. State Bar of California, 366 US 36 (1961).

(39) Ibid. at 50.

(40) Ibid. at 67-68.

(41) United States v. O'Brien, 391 US 367 (1968) [O'Brien].

(42) Ibid. at 376-77 [numbering added by author, footnotes omitted].

(43) The Court would later disapprove of this subterfuge. See O'Connor J's warning in Simon & Schuster. supra note 18 at 121.

(44) Ward et al. v. Rock Against Racism, 491 US 781 (1989) [Rock Against Racism].

(45) Ibid. at 784.

(46) Clark v. Community for Creative Non-Violence, 468 US 288 (1984) [Creative Non-Violence].

(47) Rock Against Racism, supra note 44 at 791; citing Creative Non-Violence, ibid. at 293 [numbering added by author, footnotes omitted].

(48) Rock Against Racism, ibid. at 798-99; citing United States v. Albertini, 472 US 675 (1985) at 689 [Albertini] [footnotes omitted].

(49) Fair Political Practices Com. v. Superior Court, 25 Cal.3d 33 at 47 (1979).

(50) Rock Against Racism, supra note 44 at 800.

(51) Rock Against Racism, supra note 44 at 800, citing Albertini, supra note 48 [punctuation omitted by author].

(52) Central Hudson Gas & Elec. Corp. v. Public Service Commission of New York, 447 US 557 at 566 (1980) [Central Hudson] [numbering added by author].

(53) Board of Trustees of the State University of New York v. Fox, 492 US 469 at 476-80 (1989) [Fox] [footnotes omitted].

(54) Actually, almost a century ago Brandeis J. hinted at the teleology of free expression, but it was forgotten in modern cases. See Whitney v. California, 274 US 357 at 375-6 (1927):

   Those who won our independence believed ... that freedom is to
   think as you will and to speak as you think are means indispensable
   to the discovery and spread of political truth ... But they knew
   that order cannot be secured merely through fear of punishment for
   its infraction; that it is hazardous to discourage thought, hope
   and imagination ... Believing in the power of reason as applied
   through public discussion ...[and] recognizing the occasional
   tyrannies of governing majorities, they amended the Constitution so
   that free speech and assembly should be guaranteed.

(55) R.A.V. Petitioner v. City of St. Paul, Minnesota., 505 US 377 (1992) [R.A.V.].

(56) The St. Paul Bias-Motivated Crime Ordinance, [section] 292.02 (1990), reads:

   Whoever places on public or private property, a symbol, object,
   appellation, characterization or graffiti, including, but not
   limited to a burning cross or Nazi swastika, which one knows or has
   reasonable grounds to know arouses anger, alarm or resentment in
   others on the basis of race, color, creed, religion or gender
   commits disorderly conduct and shall be guilty of a misdemeanor.

(57) R.A.V., supra note 55 at 393-96.

(58) Ibid. at 395.

(59) Ibid. at 393.

(60) The minority did consider this, noting that the Court had upheld categories of content based proscriptions where "their expressive content is worthless or of de minimis value to society ... his categorical approach has provided a principled and narrowly focused means for distinguishing between expression that the government may regulate and that which it may regulate on the basis of content only upon showing a compelling need", ibid. at 400.

(61) Treaty Establishing the European Community, 25 March 1957, online: Celex Test <http://europa.eu.int/eur-lex/en/treaties/dat/EC_consol.html>, reproduced in Bernard Rudden & Derrick Wyatt, Basic Community Laws. 5th. Ed. (Oxford: Oxford University Press, 1994). The Treaty is colloquially known as the Treaty of Rome.

(62) For the direct effects of the Treaty, see NV Alegemene Transporten--en Expeditie Onderneming van Gend & Loos v. Netherlands Inland Revenue Administration, C-26/62, [1963] E.C.R. I-1, [1963] C.M.L.R. 105; for the direct effects of the Community directives, see Yvonne van Duyn v. Home Office, C-41/74, [1974] E.C.R. I-1337.

(63) P. Pescatore, "The Doctrine of Direct Effect: An Infant Disease of Community Law" (1983) 8 E.L.Rev. 155 at 158.

(64) See Paul P. Craig & Grainne de Burca, EC Law: Text, Cases and Materials (Oxford: Oxford University Press, 1995) c. 4-6.

(65) The other two are the principles of "legal certainty" and "legitimate expectations".

(66) The common law is exceptional for not having a free-standing cause of action in proportionality, but using it only (taking the case of Canada) as a jurisprudential tool in the law of the Charter, as well as aboriginal-constitutional law and criminal sentencing. The legal systems of most European nations recognize disproportionality as a cause of action, and some would argue a pale imitation exists in English law. See J. Schwarze, European Administrative Law (Sweet & Maxwell, London, 1992) c. 5; N. Emiliou, The Principle of Proportionality in European Law (London: Kluwer, 1996); and S. Boyron, "Proportionality in English Administrative Law: A Faulty Translation" (1992) O.J.L.S. 237.

(67) See R. v. Intervention Board far Agricultural Produce Ex p. ED&F (Sugar) Ltd, C-181/84, [1985] E.C.R. I-2889; R. v. Ministry of Agriculture, Fisheries and Food Ex p. FEDESA, C-331/88, [1990] E.C.R. I-4023, [1991] 1 C.M.LR. 507 [FEDESA].

(68) See Bilka-Kaufhaus GmbH v. Weber von Hartz, C-170/84, [1986] E.C.R. I-1607, [1986] 2 C.M.L.R. 701.

(69) Procureur du Roi v. Dassonville, C-8/74, [1974] E.C.R. I-837, [1974] 2 C.M.L.R. 436 at para. 5 [cited to E.C.R.].

(70) Rewe-Zentral AG v. Bundesmonopolverwaltung fur Branntwein ("Cassis de Dijon "), C-120/98, [1979] E.C.R. I-649, [1979] 3 C.M.L.R. 494 at para. 8 [cited to E.C.R.]. The Cassis list of mandatory requirements is not exhaustive and has been added to in subsequent cases.

(71) FEDESA, supra note 67 at para. 13 [numbering added by author].

(72) See Grainne de Burca, "The Principle of Proportionality and its Application in EC Law" [1993] Y.B. Eur. L. 105 at 113 [de Burca, "Proportionality"], for a rewording of the E.C.J.'s test while staying true to its meaning:

   First, the articulation of the State's interest, i.e. was the
   measure a useful, suitable or effective means of achieving a
   legitimate aim or objective? Secondly, the articulation of the
   affected interest, i.e. was there a means of achieving that aim
   which would be less restrictive of the applicant's interest.
   Thirdly, even if there was no less restrictive means of achieving a
   legitimate public aim, does the measure have an excessive or
   disproportionate effect on the applicant's interest?

(73) Commission v. Germany, C-178/84, [1987] E.C.R. I-1227, [1988] 1 C.M.L.R. 780 [German Beer]. See also the case comment of H. Clark, "The Free Movement of Goods and Regulation for Public Health and Consumer Protection in the EEC: The West German 'Beer Purity' Case" (1988) 28 Va. J. Int'l L. 753.

(74) See the opinion of Advocate General Slynn, German Beer, ibid. at I-1253 [cited to E.C.R.].

(75) Ibid. at 1239.

(76) Ibid. at 1237.

(77) Ibid. at paras 41-53.

(78) The Commission led in evidence "Acceptable Daily Intakes" (ADIs) for some, but not all, of the additives in beer; but these ADIs, as the Commission's own witness pointed out, "do not guarantee absolute safety". Thus, for none of the additives in question was there conclusive evidence of safety or harmfulness. See the preliminary opinion of Advocate General Slynn, ibid. at 1258.

(79) C-302/86, [1988] E.C.R. I-4607, [1989] 1 C.M.L.R. 619. [Danish Bottles cited to E.C.R.]

(80) Ibid. at paras. 9, 13.

(81) Ibid. at paras 15-22.

(82) Ibid. at I-4625-26.

(83) FEDESA, supra note 67.

(84) GATT, supra note 2.

(85) See J. Jackson, The World Trading System: Law and Policy of International Economic Relations (Cambridge: MIT Press, 1989) c. 2; R. Hudec, "The Judicialization of GATT Dispute Settlement" in M. Hard & D. Steger, In Whose Interest? Due Process and Transparency in International Trade (Ottawa: Centre for Trade Policy and Law, 1992) 14.

(86) See World Trade Organization, "Understanding on Rules and Procedures Governing the Settlement of Disputes", online: WTO <www.wto.org/English/docs_e/legal_e/28-dsu.pdf> [WTO, "Understanding Settlement"].

(87) Ibid. at Article 8.

(88) Confusion can arise as to the usage of "WTO" and "GATT", and they are sometimes used interchangably. I use "WTO" to denote the Geneva-based bodies that administer the GATT treaty, including the dispute settlement Panels. Reference to the "GATT", in contrast, is meant to refer to the treaty itself.

(89) It is true that Article XI:1 specifically exempts duties, taxes and other charges where Article 30 does not, but for the present analysis, nothing turns on this difference.

(90) Other derogations can be found at Articles XI, XII, XIV, and XXI.

(91) WTO, United States-Standards for Reformulated and Conventional Gasoline--Report of the Panel, WTO Doc. WT/DS2/AB/R (1996), online: 2ABR.wpf <http://docsonline.wto.org/DDFDocuments/ t/WT/DS/2ABR.WPF> [U.S. Gasoline].

(92) GATT, United States--Prohibition of Imports of Tuna and Tuna Products from Canada--Report of the Panel, GATT Doc. L/5198, 29s B.I.S.D. (1982) 91 at para. 3.9, online: WTO <http://www.wto.org/gatt_docs/English/SULPDF/90990141.pdf>.

(93) This is discussed in the analysis of the Sea Turtles appeal, infra note 106.

(94) For a different view that proportionality has nothing to do with Article XX, see E. Petersmann, The GATT/WTO Dispute Settlement System (London: Kluwer, 1997) at 118 [Petersmann, GATT/WTO Dispute Settlement System].

(95) GATT, United States--Section 337 of the Tariff Act of 1930, GATT Doc. L/6439, 36s BISD (1989) 345 at paras. 5.25-5.3, online: 91390261.pdf <http://www.wto.org/gatt_docs/English/SULPDF /91390261.pdf>. Note that this case concerned the Article XX(d) exemption, which also uses the word "necessary".

(96) Oakes, supra note 13.

(97) Central Hudson, supra note 52.

(98) GATT, Thailand--Restrictions on Importation of and Internal Taxes on Cigarettes--Report of the Panel GATT Doc. DS10/R, 37s B.I.S.D. (1990) 200, online: 91520143.pdf <http://www.wto.org/gatt_docs/English/SULPDF/91520143.pdf> [Thai Cigarettes].

(99) Ibid. at para. 50.

(100) Ibid. at paras. 52-53.

(101) Ibid. at para. 55.

(102) Ibid. at para. 52.

(103) Ibid. at paras. 75-81 [emphasis added].

(104) Sea Turtles, infra note 106.

(105) Endangered Species Act, 16 U.S.C. [section] 609 (1991).

(106) WTO, United States--Import Prohibition of Certain Shrimp and Shrimp Products--Report of the Panel, WTO Doc. WT/DS58/R (1998), online: 58R00.wpf <http://docsonline.wto.org/DDFDocuments/t/WT/DS/58R00.WPF> [Sea Turtles].

(107) Ibid. at paras. 118-119, 125. The "provisional justification" rule comes from the appeal decision in U.S. Gasoline, supra note 91.

(108) Ibid. at para. 134.

(109) Ibid. at para. 138.

(110) Ibid. at paras. 138-140.

(111) Ibid. at para. 147.

(112) Ibid. at paras. 156, 159.

(113) Ibid. at paras. 161,164 [italics in original].

(114) Ibid. at para. 161.

(115) For a discussion of this salutary effect of balancing in contexts other than commercial speech, see J. Ely, "Flag Desecration: A Case Study in the Roles of Categorization and Balancing in First Amendment Analysis" (1975) 88 Harv. L. Rev. 1482.

(116) O'Brien, supra note 41.

(117) Creative Non-Violence, supra note 46.

(118) Central Hudson, supra note 52.

(119) Fromancais SA v. FORMA, C-66/82, [1983] E.C.R. 1-395, [1983] 3 C.M.L.R. 453.

(120) See de Burca, "Proportionality", supra note 64. The conclusion of that paper for the diversity of meanings of "proportionality" one may glean from the ECJ's decisions.

(121) On the ECJ, see A. Arnull, "Owning up to Fallibility: Precedent and the Court of Justice" (1993) 30 CML Rev. 247. On the GATT, see Jackson, supra note 85 at 89-90.

(122) See D. Futuyma, Evolutionary Biology (Sunderland, MA: Sinauer, 1998) at 110-111.

(123) Rock Against Racism, supra note 44 at 800.

(124) Irwin Toy, supra note I I at para. 80.

(125) Committee for the Commonwealth of Canada v. Canada, [1991] I .S.C.R. 139 at para. 273.

(126) Illinois Elections Bd. v. Socialist Workers Party, 440 US 173 at 188-89 (1979).

(127) See discussion above, note 17. Constitutional law scholars have accordingly expressed the view that this part of the Oakes test is redundant: see Hogg, supra note 15 at 883.

(128) This was decided in Rock Against Racism, supra note 44.

(129) Sea Turtles, supra note 106.

(130) Beatty, supra note 8 at 187.

(131) An example of this is the Supreme Court of Canada's decision in RJR- Macdonald, supra note 16, striking down sections of the Tobacco Products Control Act. Parliament responded by re-legislating provisions similar to those that had been declared unconstitutional, taking its means more carefully into account.

(132) The European Commission has announced that it is will again litigate, this time to compel Denmark to accept recyclable containers, and end its reliance on re-useable ones: see "EU Takes Denmark to Court Over Canned Drink Ban", Agence France Presse (21 April 1999).

(133) See "Denmark repeals ban on canned beer and soft drinks" Miljostyrelsen (15 February 2002), online: Miljostyrelsen <http://glwww.mst.dk/news/09030000.htm>.

(134) See J.H. Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge: Harvard University Press, 1980).

(135) R. Housman, "A Kantian Approach to Trade and the Environment" (1992) 49 Wash. & Lee L.Rev. 1373.

(136) See M. Shenk, "US Gasoline--Case Comment" (1996) 90 AJIL 669 at 672-74; J. Waincymer, "Reformulated Gasoline under Reformulated WTO Dispute Settlement Procedures: Pulling Pandora out of a Chapeau?" (1996) 18 Mich. J. Int'l L. 141 at 154-76.