Legal Bilingualismt
Roderick A. Macdonald*
Legal bilingualism is an ideal to which the Cana-
dian legal order aspires. In practice this aspiration has
led to the production of legislation and federal judicial
decisions in both English and French. This essay con-
siders the semantic and epistemological foundations of
legal bilingualism. It argues that postmodem critiques
of legal indeterminacy rest on the same impoverished
view of human communicative symbolisms that sus-
tains the claim that legal bilingualism can be realized
by simply translating legal texts. Once the deferential
and presentational capacities of human symbolisms are
recognized alongside
their rational and discursive
properties, the relationship between language and legal
knowledge reveals its complexity. This essay explores
the theory and practices of legal bilingualism and re-
views the institutional requirements of a bilingual legal
order. Legal bilingualism is contrasted with legal dual-
ism in the conclusion to the essay, where the semantic
and epistemological lessons of legal bilingualism are
then read back into the interpretation of unilingual legal
orders.
les
critiques
post-modemes
Le bilinguisme juridique est un id al auquel as-
pire l’ordre juridique canadien. En pratique, cette aspi-
ration a men6 A Ia production de textes lgislatifs et de
decisions judiciaires et en frangais et en anglais. Le
present article examine les fondations semantiques et
6pistemologiques du bilinguisme juridique. L’auteur
soutient que
de
l’inddterminisme juridique reposent sur la vision ap-
pauvrie des symbolismes communicatifs humains qui
sert de base aux revendications de ceux qui croient
qu’il est possible de traduire un pur metalangage 16gal
en un langage naturel. Une fois les capacites defdren-
tielles et presentationnelles des symbolismes humains
reconnues parall~lement A leurs proprietes rationnelles
et discursives, le lien entre la langue et le savoirjuridi-
que rev le sa complexite. Cet article explore la theorie
et les pratiques du bilinguisme juridique et passe en re-
vue les impdratifs institutionnels indispensables A un
ordre juridique bilingue. L’auteur met en 6vidence la
distinction entre le bilinguisme juridique et le dualisme
juridique. La conclusion intgre les lemons semantiques
et 6pistdmologiques du bilinguisme juridique pour en-
suite les appliquer 4 des ordres juridiques unilingues.
‘This essay originated in a project entitled “Canada’s Bilingual and Bicultural Legal System”, pre-
pared while I held a Canadian Law Information Council Summer Research Fellowship in 1981. At
that time I benefitted from the research assistance of Mindy Paskell-Mede. A revision of the C.L.I.C.
study, prepared by Stephen Toope, was delivered to the Canadian Section of the International Asso-
ciation for Legal and Social Philosophy at Vancouver in May 1983. In 1988, Richard Janda reshaped
the essay for delivery to legal-theory workshops at McGill and the University of Toronto. In 1990, the
text was edited once more by David Lametti and was presented at faculty seminars at the Australian
National University and Murdoch University. Desmond Manderson helped me rework the essay in a
French-language version for presentation to a workshop at the University of Montreal in 1994. The
essay was again substantially revised for presentation at a faculty seminar at U.B.C. in January 1996
and at “The Blasted Pine” conference. To my various research assistants, to those participants in the
seminars at which earlier versions of this text were presented, to the designated commentator at “The
Blasted Pine” conference, Mary Ellen Turpel, and to my colleagues and former colleagues G. Blaine
Baker, Michael Bridge, Daniel Jutras, Nicholas Kasirer, Dennis Klinck, Shauna Van Praagh and Jer-
emy Webber – each of whom read and commented upon the essay (some in several of its forms) –
I
am most grateful. None, however, should be held responsible for defects in the present essay.
. RR. Scott Professor of Constitutional and Public Law, Faculty of Law, McGill University; Fellow,
Law and the Determinants of Social Order Programme, Canadian Institute for Advanced Research.
McGill Law Journal 1997
Revue de droit de McGill
To be cited as: (1997) 42 McGill L.J. 119
Mode de rdfdrence : (1997) 42 R.D. McGill 119
120
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 42
Prologue
Introduction
I. The Semantics of Legal Bilingualism
II. The Epistemology of Legal Bilingualism
Ill. Theorizing about Legal Bilingualism in Canada
IV. Practicing Legal Bilingualism in Canada
Conclusion
19971
R.A. MACDONALD – LEGAL BILINGUALISM
Prologue
11.1 Now the whole earth had one language and few words … 4. Then they
said, “Come, let us build ourselves a city, and a tower with its top in the heav-
ens, and let us make a name for ourselves, lest we be scattered abroad upon the
face of the whole earth.” 5. And the LORD came down to see the city and the
tower, which the sons of men had built. 6. And the LORD said, “Behold, they
are one people, and they have all one language; and this is only the beginning
of what they will do; and nothing that they propose to do will now be impossi-
ble for them. 7. Come, let us go down, and there confuse their language, that
they may not understand one another’s speech.” 8. So the LORD scattered
them abroad from there over the face of all the earth, and they left off building
the city. 9. Therefore its name was called Babel, because there the LORD con-
fused the language of all the earth; and from there the LORD scattered them
abroad over the face of all the earth.
4. Chap. XI, 1-9: Causes de la Dispersion des Peuples et de la Confusion
des Langues; La Tour de Babel.
Toute la terre avait une seule langue et les m~mes mots … Ils dirent encore:
“Allons, batissons-nous une ville et une tour dont le sommet soit dans le ciel, et
faisons-nous un monument, de peur que nous ne soyons dispers6s sur Ia face
de toute la terre.” Mais Yahweh descendit pour voir la ville et ]a tour que batis-
saient ls fils des hommes. Et Yahweh dit: “Voici, ils sont un seul peuple et ils
ont pour eux tous une mame langue; et cet ouvrage est le commencement de
leurs entreprises; maintenant rien ne les empchera d’accomplir leurs projets.
Allons, descendons, et IM mime confondons leur langage, de sorte qu’ils
n’entendent plus le langage les uns des autres.” C’est ainsi que Yahweh les
dispersa de IA sur la face de toute la terre, et ils cess rent de bAtir Ia ville. C’est
pourquoi on lui donna le nom de Babel, car c’est IM que Yahweh confondit le
langage de toute la terre, et c’est de li que Yahweh les a disperses sur la face de
toute la terre.’
‘The two quotations are from: (i) The Holy Bible (Revised Standard Version, 1952) Genesis 11:1,
4-9; and (ii) La Sainte Bible (La Socidtd de Saint-Jean l’6vangeliste, 1968) la Gen~se; I’ partie; V”
section, sous-section 4; 11″- chapitre: 1, 4-9. Apart from linguistic differences relating to vocabulary
other semiotic divergences
and syntax –
between these texts are immediately apparent: the use of numbered verses in English; the use of
capitals in English and bold text in French; the use of an introductory title in French. Each of these is
a significant element through which the reader apprehends the text and ascribes meaning to it. But we
should not conclude that any of these devices are specific to either English or French.
the traditional referents for the derivation of meaning –
All communication is translation, even where the “same” language is being deployed. In order to
give a preliminary indication of this subthesis of the present essay it is helpful to compare the render-
ing of the story of the Tower of Babel given in the King James Version of the Bible, which reads as
follows:
11.1 And the whole earth was of one language, and of one speech … 4. And they said,
Go to, let us build us a city and a tower, whose top may reach unto heaven; and let us
make us a name, lest we be scattered abroad upon the face of the whole earth. 5. And
122
MCGILL LAW JOURNAL/REVUE DEDROITDE MCGILL
[Vol. 42
Introduction
1. On its traditional theological reading, the story of the Tower of Babel teaches that
the multiplicity of human languages is evidence of our fall from grace. If only we had
not been punished by God, we could have built to the heavens. One language would
give us all a shared knowledge and knowledge, after all, is power.’
But the story leaves us uncertain as to whether the multiplicity of languages after
Babel is a genuine barrier to shared knowledge or whether it, in fact, best reflects the
nature of the knowledge that is accessible to us through language. Is the story of Ba-
bel about limits on the capacity of human beings to communicate with each other, or
is it also about limits on the capacity of human beings to deploy language in express-
ing what they know?
the LORD came down to see the city and the tower, which the children of men builded.
6. And the LORD said, Behold, the people is one, and they have all one language; and
this they begin to do: and now nothing will be restrained from them, which they have
imagined to do. 7. Go to, let us go down, and there confound their language, that they
may not understand one another’s speech. 8. So the LORD scattered them abroad from
thence upon the face of all the earth: and they left off to build the city. 9. Therefore is
the name of it called Babel; because the LORD did there confound the language of all
the earth: and from thence did the LORD scatter them abroad upon the face of all the
earth.
It is to be noted that, while both are in English, the Revised Standard Version differs from the
King James Version not only in its grammar, syntax and vocabulary, but also in its translation and in-
terpretation of Hebrew, Greek and Latin texts. These discrepancies are especially evident in verse 1,
where the Revised Standard Version suggests, in using the expression “few words” rather than the
expression “one speech” (found in the King James Version), an interpretation of the story that raises
not only questions of translation (semantics), but more fundamental questions of interpretation
(epistemology). Moreover, the French-language version presents itself as a combination of the two
English-language translations: between “few words” and “one speech” lies “les m~mes mots”. Here a
distinction is being drawn between grammar and syntax on the one hand, and vocabulary on the other.
2The classical reading of the Tower of Babel is presented in N. Frye, The Great Code: The Bible
and Literature, 1st ed. (New York: Harcourt Brace Jovanovich, 1982) at 158,230. See also G. Steiner,
After Babel: Aspects of Language and Translation, 2d ed. (Oxford: Oxford University Press, 1992)
passim, on the theology of Babel.
Socio-linguists have another reading of the story of Babel, formulated as the “Law of Babel”.
See J.A. Laponce, “The Language System and the Language Policies of Canada, Functionalities and
Dysfunctionalities” in K. Kulcsar & D. Szabo, eds., Dual Images: Multiculturalisin on Two Sides of
the Atlantic (Budapest: Royal Society of Canada and Hungarian Academy of Sciences, 1996) 98 at
98-99:
[T]he number of distinct languages used for communication in a closed system is re-
lated to the density of communication within that system. Concentrate all the individu-
als … in a single and dense communication network, such as a single big city, and they
will, over time, develop a common language … On the contrary, if you scatter people
all over the map, if you separate them by seas, jungles, rivers and mountains (and, for
good measure you may also wish to separate them by hatred and fear), then, says the
Law of Babel, men will develop a multiplicty of languages and will cease to under-
stand one another.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
2. Differences between languages seem to make it more difficult to converse with
someone who does not speak one’s own tongue than with someone who does. Yet,
these differences do not render us mute when faced with people who speak a lan-
guage foreign to us. The fact that we can communicate despite differences in lan-
guage points to the possibility of a shared human knowledge beyond language? In-
deed, the history of intercultural contact, whether in colonial settings or otherwise,
confirms the capacity of human beings to communicate and to symbolize prior to
learning a “natural” language.’
Conversely, the existence of mutually incomprehensible tongues suggests a fun-
damental individuating impulse of language itself.’ For example, some thirty different
varieties of English are now spoken around the world.’ Each human language, and
each dialect within each human language, is a distinctive symbol system –
as diver-
gent and as productive of insight as music and painting. Each human language and
each dialect within each human language permits each speaker or listener, writer or
reader, to fashion a different communicative timbre. Grammatical differences between
active and passive voices or among indicative, subjunctive and imperative moods
within a language, permit us, and at the same time oblige us, to fashion what we wish
to say in particular ways. This is even more so the case with grammatical and syntac-
tical differences as among human languages.
But the lesson is deeper. Ultimately, all communication is translation; ultimately,
all listeners and readers are imperfectly recoding speech or text into their own per-
sonal language. Learned conventions and generalized, recurrent contexts for the de-
ployment of natural languages desensitize us to the complexity of communication
through language, especially among those who “share” the same natural language.’
But confronted with Babel – with the confusion of tongues – we cannot escape this
‘On various understandings of the human capacity to acquire language, and to acquire a second or
third language, see S. Pinker, The Language Instinct: How the Mind Creates Language (New York:
William Morrow, 1994).
‘In this essay, unless the context clearly indicates otherwise, I use the word “language” to refer to
what linguists call “natural languages”. Conventionally, natural languages are said to be human lan-
guages like English or Japanese, as opposed to a computer language, musical notation, formulas in
logic, and so on. See generally H. van Riemsdijk & E. Williams, Intmduction to the Theory of
Granmnar (Cambridge: M.I.T. Press, 1986).
‘This assertion is not, however, uncontroversial. Traditional linguistic analysis had it, to paraphrase
the formula popularized by Cailleux, that “for every language that dies, two are born” (see A.
Cailleux, “L’evolution quantitative du langage” (1953) Bulletin socidt6 prdhistorique frangaise 505).
Yet, today, the trend seems to be reversing itself. Given increasingly globalized communications
linguists predict that some 90 percent of the world’s 7000 languages
structures –
are likely to disappear in the not too distant future (see S. Moisan, “La disparition menace 90% des
langues de la plan&e” Le Devoir [Montral] (10 September 1992) 11, cited in Laponce, supra note 2,
98 at 99).
‘ See D. Crystal, The Cambridge Encyclopaedia of the English Language (Cambridge: Cambridge
the global village –
University Press, 1994).
7 See, however, J.-C. Gtmar, “La traduction juridique: art ou technique d’interprtation?” (1987) 18
R.G.D. 495, for a sensitive presentation by a linguist of the point that there is little legal activity that
does not involve the same intellectual operations that a “legal translator” routinely deploys.
MCGILL LAW JOURNAL /REVUE DE DROITDE MCGILL
[Vol. 42
engagement with the capacities and limitations of human language. More than this,
were it not for Babel we might miss the richness of our expressive resources beyond
language. Hence, the paradox of Babel: we might actually be richer rather than poorer
for having many languages.8
3. Neither mainstream nor critical legal scholarship today seems especially sensitive
to this paradox. Mainstream jurists who ascribe to legal texts the capacity for the ex-
act reflection of human normative intention presume also the possibility of exact
translation of legal texts. At its limits, this presumption reflects a longing for one
shared legal discourse, a longing that finds its way into many considerations of legal
translation. Some time ago, R. Michael Beaupr6 wrote approvingly of efforts to iden-
tify a “legal metalanguage”:
It would not be one artificially concocted and imposed, but one that is, in a
sense, waiting to be discovered through joint research by linguists, translators,
comparativists (in addition, presumably, to philosophers and psychologists)
who would seek to unlock a legal prototype which, it may be argued, is already
part of our “collective unconscious”
These contemporary legal Kabbalists, who hold fast to theories of linguistic
monogenesis and universal grammar, and for whom redemption lies in the rediscov-
ery of the “translucent immediacy of that primal lost speech shared by God and
Adam”,’ assume that law would be perfected by being reduced to one language
“esperanto juridiko” – will, it is thought,
form.” A “positive” legal metalanguage –
save legal discourse from anarchy.’2 The re-establishment of a transnational us com-
mune, to be expressed in a scientifically constructed language that is rationally com-
‘Put differently, it appears that bi- or multi-lingual persons have insights about the connections
between knowledge and language and between a symbol and the symbolized that escape unilinguals.
See K. Hakuta, Mirror of Language (New York: Basic Books, 1986), especially c. 9, “Reflections on
Bilingualism”.
9 R.M. Beaupr6, “La traduction juridique: Introduction” (1987) 28 C. de D. 735 at 745. Beaupr6
cites the “tentative progress toward standardizing ‘common law French’ as well as efforts towards the
discovery of ‘neutral’ common law – civil law terminology” as evidence that a legal metalanguage
might be possible. For the suggestion that computerization may assist in generating a neutral, positive
legal language that lends itself to direct translation, see D. Bourcier & E. Andreewsky, ‘Traduction et
polys6mie: un exemple de traitement automatique en informatique juridique” in J.-C. Gdmar, ed.,
Langage du droit et traduction (Quebec: ,diteur officiel du Quebec, 1982) 233. See also R.M. Beau-
pr6, Interpreting Bilingual Legislation, 2d ed. (Toronto: Carswell, 1986); R.M. Beaupr6, “Litigating
the Meaning of Bilingual Legislation” (1988) 9 Advocates’ Q. 327 [hereinafter “Litigating the
Meaning”].
‘0 Steiner, supra note 2 at 474.
“As an antidote to uncritical assent to this proposition, one might reflect upon the apocalyptic
prophesy of “pure speech” in Zephaniah 3:9, or the “gift of tongues” in Acts 2:4-12, as Biblical alle-
gory for the preconditions for, and consequences of, universal discourse.
” Here the hope is to discover the legal equivalent of pure mathematical symbols: a vocabulary (1,
2, 3, pi, etc.); a grammar (+, -, x, =, etc.); and a syntax (rules for reading and prioritizing mathematical
operations). On the emergence of mathematical symbolisms, see G. Ifrah, L’Histoire universelle des
chiffres: L’intelligence des hommes racontde par les nombres et le calcul, vol. 1 (Paris: Robert Laf-
font, 1994) passim, but especially at 1-20.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
plete and that exhausts the imaginable legal universe, motivates legal Kabbalists. Both
nostalgic Kabbalists who seek to reinvent the hegemony of Roman law and revolu-
tionary Kabbalists who promote expert-systems as the universalizing legal discourse
share the ambition.’3 Like jurimetricians, empiricists and social scientists before them,
today’s revolutionary legal Kabbalists believe that the project of law may be saved if
recentred on the description of things, not norms.”
4. Contrast this belief with that of contemporary critical legal scholars. Latter-day
American Legal Realists, who claim that the inherent “arbitrariness, incoherence and
indeterminacy” of language fatally infects legal discourse,’5 at first seem far removed
in their diagnosis from the Kabbalists. Yet their preoccupations are not so different.
While they see that formal statements of law posing as objective norms are readily
dissolved, they too assume that language and meaning are one, and therefore deny
that law can have meaning apart from its use by the powerful to coerce:
Knowledge and social power are inseparable … [T]here is no way to achieve
closure with respect to the meaning of expressions or events. The distribution
of meaning depends on socially created and contingent representational con-
ventions. Each attempt to fix meaning is belied by the dependence of meaning
on language. Meaning is dependent on artificial and differential signification
practices.
,3 For a detailed critique of nostalgic Kabbalism, especially in its neo-Romanist versions (e.g., H.
Berman & C. Reid, “Roman Law in Europe and the ius commune” in Scintillae Juris: Studi in
Memoria di Gino Gorla, I (Milan: Guiffre, 1994) 989), see P. Legrand Jr., “Antiqui Juris Civilis
Fabulas” (1995) 45 U.T.LJ. 311. A similar critique of revolutionary Kabbalism –
that promoted by
devotees of “expert systems” (e.g., J. Goulet, La machine a faire le droit (Sillery: Presses de
an Assess-
l’Universit6 du Qu6bec, 1987)-
ment of its Potential for Artificial Intelligence Applications in Law” (1991) 2 J.L. & Info. Sci. 137,
and in “The Open Texture of Language: Handling Semantic Analysis in Decision Support Systems”
(1993) 4 J.L. & Info. Sci. 330. See also A. Wolfe, “Algorithmic Justice” (1990) Cardozo L. Rev.
1409; S. Utz, “Rules, Principles, Algorithms and the Description of Legal Systems” (1992) 5 Ratio
Juris 23.
is developed by R.N. Moles in “Logic Programming –
1See EH. Shuck, “Why Don’t Law Professors Do More Empirical Research?” (1989) 39 J. Legal
Educ. 323, who believes that empirical research can give rise to “positive knowledge” and thus a firm
foundation to replace the indeterminacy of its existing “language-bound rules”. Richard Posner makes
a similar plea in Overcoming Law (Cambridge: Harvard University Press, 1995). Compare, however,
D.M. Trubek & J. Esser, “‘Critical Empiricism’ in American Legal Studies: Paradox, Program or
Pandora’s Box?” (1989) 14 L. & Soc. Inquiry 3; RJ. Coombe, “Room for Manoeuver. Toward a
Theory of Practice in Critical Legal Studies” (1989) 14 L. & Soc. Inquiry 69; R.J. Coombe, “‘Same
As it Ever Was’: Rethinking the Politics of Legal Interpretation” (1989) 34 McGill L.J. 603; see also
C. Geertz, Local Knowledge: Further Essays in Interpretive Anthropology (New York: Basic Books,
1983) at 167.
” The periodical and monographic literature on these so-called “indeterninacy truths” is over-
whelming. For a recent and representative Canadian expression, see A.C. Hutchinson, Waiting for
Coraf. A Critique of Law and Rights (Toronto: University of Toronto Press, 1995).
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 42
[Llegal thought is a representational discourse which purports to re-present
social relations in a neutral manner. But like language generally, legal discourse
can never escape its own textuality.”‘
the conception of the legal enterprise as a mask for oppression –
Here one confronts the obverse to Kabbalism. Adherents to the “contingency
thesis” –
also ask
of law what Ayer and other logical positivists earlier asked of language. But far from
lamenting the law’s failure to live up to its promise to be a universal discourse –
as if
it made sense in the first place to assume that law should ever have been a linguistic
edifice logically erected upon protocol sentences”‘ –
scholars affiliated with the Criti-
cal Legal Studies movement, and their fellow travellers among less sophisticated legal
postmodernists,” seize on the failure of the Kabbalists’ claim about language as proof
that legal normativity itself is impossible.’9 Knowledge and social power being insepa-
rable in their view, law and presumably every other human symbolism derive from
and reinforce existing distributions of power; none has any critical purchase on power
20
itself.
5. Between the illusion of homogeneous, neutral, determinate legal language and the
illusion of contingent, arbitrary, indeterminate legal language lies the insight of legal
bilingualism. Legal bilingualism (or more radically, legal multilingualism) takes as
given that the complete normative content of law cannot be expressed by a particular
set of words in one or any number of languages; but it also takes as given that lan-
guage is a privileged communicative symbolism for apprehending law’s normativity.
All law, given this insight, is multilingual.
“G. Peller, “The Metaphysics of American Law” (1985) 73 Calif. L. Rev. 1152 at 1170, 1182
×..
‘7 Like much modem legal scholarship, the Critical Legal Studies project has learned only half the
lesson of Kelsen’s and Camap’s failed program. That Camap did not succeed does not mean that all
language is metaphysical nonsense, only that a universal grammar which exhausts human knowledge
is impossible. See R. Camap, “The Elimination of Metaphysics Through Logical Analysis of Lan-
guage” in A.J. Ayer, ed., Logical Positivism (Glencoe, Ill.: Free Press, 1959) 60.
” A helpful introduction to legal postmodemism, which argues the contingency thesis through a
critique of traditional legal theory, is presented in C. Douzinas & R. Warrington, Postmodem Juris-
prudence: The Law of Text in the Texts of Law (New York: Routledge, 1991). See also C.J.G. Samp-
ford, The Disorder of Law: A Critique of Legal Theory (Oxford: Blackwell, 1989).
‘” For an alternative conception of the claims of legal discourse, see 0. Barfield, “Poetic Diction and
Legal Fiction” in C.S. Lewis et al., eds., Essays Presented to Charles Williams (Oxford: Oxford Uni-
versity Press, 1947) 106; C.D. Stone, “From a Language Perspective” (1981) 90 Yale L.J. 1149.
” A powerful statement of this perspective may be found in A. Hunt, Explorations in Law and So-
ciety: Toward a Constitutive Theory of Law (New York: Routledge, 1993). The fundamental incoher-
ence of an argument that social power shapes human symbolizing but is itself not a form of human
symbolizing subject to its own critique is carefully demonstrated in H. Putnam, Pragmatism: An
Open Question (Cambridge: Blackwell, 1995). See further A. Honneth, The Critique of Power: Re-
flective Stages in a Critical Social Theory, trans. K. Baynes (Cambridge: M.I.T. Press, 1991) who ar-
gues that postmodemism, properly understood, provides a vantage point of resistance to power, be-
cause it denies “authorial authority” and thus democratizes interpretation.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
Especially during the past three decades, the idea of legal bilingualism has re-
ceived much attention in Canada.’ Federal-government policy promoting bilingual-
ism in general, including minority-language educational rights, is a central component
of the contemporary Canadian legal order.’ Statutes, regulations, judicial decisions
and government documents are now being translated from English to French and to a
lesser degree from French to English. Today, most new federal legislation is actually
being drafted in two original language versions. Bilingual courts and administrative
agencies have been established. Defendants in criminal cases may insist on being
tried in the language of their choice. A comprehensive body of published legal doc-
trine in French and English is emerging. French-language common law legal educa-
tion is a reality, and English-language civil law education has a distinguished history.
Finally, the constitutional guarantees of legal bilingualism set out in section 133 of the
Constitution Act, 1867′ have been elaborated and extended by the Official Languages
Acr’ and by the Canadian Charter of Rights and Freedoms.’ Many now apply to
certain provinces as well as to the federal government.’
6. Together, these features of Canadian law have contributed to the generation of an
official legal symbol system that, at least in its theory and in its formal trappings, op-
2 Of course, the possibility and problems of legal bilingualism have an impeccable pedigree in
Canada dating back to the late eighteenth century. For a review of the context of the codification of
the private law of Lower Canada, see J.E.C. Brierley, “Quebec’s Civil Law Codification” (1968) 14
McGill L. 521. See also Royal Commission on Bilingualism and Biculturalism, Final Report, vol. 1
(Ottawa: Queen’s Printer, 1967) at 52-55, and Royal Commission on Bilingualism and Biculturalism,
The Law of Languages in Canada (Study No. 10) by C.-A. Sheppard (Ottawa: Information Canada,
197 1). These problems are not, however, to be confounded with problems of legal bijuralism –
the
need for federal statutes to speak equally to both civil-law and common-law systems. On this collat-
eral issue, see J.E.C. Brierley, “Legal Bijuralism in Canada” in H.P. Glenn, ed., Contemporary Law:
Droit contemporain (Montreal: Yvon Blais, 1991) 22 at 41 [hereinafter “Legal Bijuralism”].
2 The literature on these themes over the past quarter-century is vast. See generally, and for a repre-
sentative sampling, M. Bastarache, ed., Les droits linguistiques au Canada (Montreal: Yvon Blais,
1986); M. Bastarache, ed., Language Rights in Canada (Montreal: Yvon Blais, 1987); A. Martel, Les
droits scolaires des minorites de langue officielle au Canada: de I’instruction a la gestion/Official
Language Minority Education Rights in Canada: From Instruction to Management (Ottawa: Office
of the Commissioner of Official Languages, 1991); D. Schneiderman, ed., Language and the State:
The Law and Politics of Identity (Cowansville, Que.: Yvon Blais, 1991).
23Constitution Act, 1867 (U.K.), 30 & 31 Vict., c. 3.
24 Offlcial Languages Act, R.S.C. 1985, c. 31 (4th Supp.).
to the Canada Act 1982 (U.K.), 1982, c. 11 [hereinafter Charter].
Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B
26 On minority-language guarantees under the Canadian constitution, see generally P.W. Hogg,
Constitutional Law of Canada, 3d ed. (Toronto: Carswell, 1992) at 1197-226 and A. Tremblay, “Les
droits linguistiques” in G.-A. Beaudoin & E. Mendes, eds., The Canadian Charter of Rights and
Freedoms, 3d ed. (Toronto: Carswell, 1996) 15-1. On the constitution itself as a bilingual document,
see R.M. Beaupr6, “Vers l’interprtation d’une constitution bilingue” (1984) 25 C. de D. 939; J.R
McEvoy, “The Charter as a Bilingual Instrument” (1986) 64 Can. Bar Rev. 155. For a more general
discussion of the theory of minority-language guarantees and their application in various countries,
see further P. Pupier & J. Woehrling, eds., Langue et droit/Language and Law (Montreal: Interna-
tional Institute of Comparative Linguistic Law, 1989).
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 42
erates in “two official languages”. Substantively, however, there may be a different
story. Constitutionalized bilingualism seems to be leading scholars and commentators
to adopt an inapt and inept view of what legal bilingualism actually requires. Briefly,
the way official legal bilingualism has been understood and practiced to date in Can-
ada implies that language precedes law and that language fully captures law.” On such
a conception of the enterprise, a bilingual legal order is composed simply of two sepa-
rate and equally authoritative collections of literary sources, each of which can be
generated and deployed in isolation from the other.’
To understand legal bilingualism as nothing more than textual duality is to ignore
the paradox of Babel. If one is to have a truly bilingual legal culture, one cannot be
content merely with producing legal artifacts in two languages. To presume that lan-
guage is merely a cipher of experience, and to act as if English and French stand more
or less in the same relation as semaphore and Morse code, or as Roman numerals and
Arabic numbers,’ is to retreat to the questionable assumptions that law can be fully
rendered by language and that language can be a universal discourse.”
7. Legal bilingualism presupposes finding a method for reading and interpreting
these legal materials that recognizes their equal authority (to the extent any text may
27 It is not difficult, however, to prove these implications to be counterfactual. To take one example,
if the federal legal order is bilingual (extrapolating from section 133 of the Constitution Act, 1867,
supra note 23), and if the federal legal order carries forth pre-confederation law in both statutory and
common-law form (extrapolating from section 129 of the Constitution Act, 1867, ibid.), then a
French-language legal norm applicable in Ontario must exist prior to the translation of the pre-
confederation statute by which it found initial textual expression, and prior to the development of a
French-language common-law rule that gives it linguistic form. On these features of section 129 for
the character of Canadian constitutionalism, see R.A. Macdonald, “The Constitutional Position of the
Civil Code of Lower Canada and the Civil Code of Quebec as an Expression of Federal Suppletive
Law” (Study prepared for the Department of Justice, March 1996) [unpublished]. See also R.A. Mac-
donald, “Encoding Canadian Civil Law” in J.E.C. Brierley et al., eds., Mdlanges Paul-Andrd Crd-
peau (Cowansville, Que.: Yvon Blais, 1997) [forthcoming].
28This theme is carefully developed in N. Kasirer, ‘The Annotated Criminal Code en version qud-
bdcoise: Signs of Territoriality in Canadian Criminal Law” (1990) 13 Dalhousie L.J. 520, where the
inadequacy of such a view is exposed.
course, despite their greater propensity to be so treated, even Morse code and semaphore are
not neutral ciphers; nor indeed are Roman numerals and Arabic characters. In the hands of skilled op-
erators, the subtext of what can be conveyed by each of the first pair differs in the way the letters are
paced and spaced; similarly, the computations that can be undertaken using Arabic characters are in-
finitely more complex than those manageable in Roman numeric characters, with the consequence
that even though both are ostensibly ciphers, one permits (encourages) intellectual operations that the
other does not – with significant practical consequences not only for mathematics, but for architec-
ture, engineering and science generally (see Ifrah, supra note 12, c. 24-27).
290f
0 The debate between those who believe the deaf should be taught a sign-language system deriva-
tive from lip reading (the “oralise’ tradition) and those who favour indigenous sign systems that ex-
ploit the cultural and gesticulatory media of communities of the deaf, is a debate between
“universalists” and those who see language as communicative symbolisms grounded in more local
practice. For discussion, see P Siple, ed., Understanding Language Through Sign Language Research
(New York: Academic Press, 1978); O.W. Sacks, Seeing Voices: A Journey Into the World of the Deaf
(Berkeley: University of California Press, 1989).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
be an authoritative expression of normativity) and that, in Canada, necessarily draws
on both English- and French-language versions. Without such a methodology, the
promise of legal bilingualism risks being transformed into a practice of defacto legal
dualism, that is, the pretence that Canadian law can be completely understood by re-
ferring to only one of the two official texts.’
This distinction between legal bilingualism and legal dualism suggests the need
for further inquiry into the theoretical foundations of legal bilingualism. It requires
asking what legal bilingualism implies about the relation between law and its various
forms of expression in language – here called the “semantic question” –
and what it
tells us about the nature of legal knowledge accessible through language –
here
called the “epistemological question”.
8. The burden of this essay is both practical and theoretical. In the realm of practice,
the essay illustrates why certain officially accepted contours and institutions of what
is deemed to be legal bilingualism in Canada are inadequate to satisfy the ambitions
of a bilingual legal order. Theoretically, the essay claims that the answers it offers to
the semantic and epistemological questions posed above are applicable to all text-
reliant legal orders –
bilingual or not. More than this, the essay suggests that these
questions inhere in any attempt to express legal normativity – whatever the chosen
vehicle of communication. 2 That is, while many problems of law in modem society
are intimately connected with the capacities of language as a mode of communication,
from a broader perspective they are at the same time connected to the intellectual and
moral capacities of human beings to symbolize.”
” The temptations of dualism are everywhere. See R.A. Macdonald, “Bilingualism or Dualism?”
(1988) 25 Lang. & Soc’y 40, and “Bilinguisme ou Dualisme?” (1988) 25 Langue et Socidt 40. These
two short texts are the English-language original and an “official” French-language translation of a
speech pointing to the perils of dualism and arguing that bilingual legal texts must necessarily be cre-
ated in two languages given their symbolic functions. How deep the irony that a speech making such
an argument should have been simply “literally” translated (without the author’s knowledge or ap-
proval) in a publication reflecting the then present practice of legal bilingualism in Canada.
2 See N. Kasirer, “Larger than Life” (1995) 10 Can. J. L. & Soc’y 185, for an evocation of the ca-
pacity of representational art to give access to legal normativity. This suggests an important caveat.
While I agree with Kasirer about the normativity of, among other symbols, art, I do not argue explic-
itly in this essay that the analysis of legal normativity being advanced applies to legal orders that have
no language component (e.g., a hypothetical legal order that is both practice-based and that records
and passes on its normative content by means of non-language symbolisms). Nor do I claim that the
analysis necessarily applies to legal orders that, while language-symbolized, have no textuality (e.g., a
hypothetical legal order that is practice-based but that records and passes on its normative content en-
tirely by oral rather than written means). The analytical claim is meant to extend only to text-based
legal orders – whether unilingual, bilingual or multilingual. On the importance of these distinctions,
see W.J. Ong, Orality and Literacy: The Technologizing of the Word (London: Methuen, 1982).
33I acknowledge that I am only scratching the surface of an enormously difficult problem that has
preoccupied the world’s greatest philosophers. For present purposes, I need only acknowledge that
although I came to most of my conclusions independently, I have since discovered a close common-
ality between the views I present here and those of Donald Davidson. See, in particular, D. Davidson,
Essays on Actions and Events (Oxford: Clarendon Press, 1980); D. Davidson, Inquiries into Truth
and Interpretation (Oxford: Clarendon Press, 1984).
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 42
Briefly, and with a particular reference to Canadian experience, the question dis-
cussed in this essay is this: Given legal bilingualism, how is it possible?’
I. The Semantics of Legal Bilingualism
9. What might the possibility of legal bilingualism tell us about the relationship
between language and law? A first step in answering this question is, obviously, to
identify different ways in which language is routinely deployed to express legal nor-
mativity. In the mid-1990s it is no longer necessary to argue at length for several basic
semantic propositions.” I take for granted the appropriation of Wittgenstein and Mer-
leau-Ponty by progressive legal scholars: who can fail to acknowledge the indetermi-
nacy and contingency of vocabulary, grammar and syntax?” Despite the assertions by
critical scholars of the novelty of their insight, the point is practically trivial. Few dis-
” Of course, even to pose this question in a law-review article raises a paradox. Like any legal es-
say, this one attempts to reduce to more or less discursive writing its answers to the problem posed. It
is not, that is, either a poem or a short story. This attempt should not be understood as a concession to
the view that law is or ought to be a neutral linguistic discourse. We are always seeking to capture in
discursive language that which is, fundamentally, metaphorical. Though a poem, play or novel might
better serve my purposes, even these language forms carry their own conventions of discursiveness.
The fact of this being an essay raises a further paradox. Language, however deployed, is only
one communicative resource open to human beings. In any essay, we are driven to capture in words
that which is not merely linguistic: would my painting, concerto, or dance (rather than my speech)
impoverish or enrich my meaning? Ineluctably, the answer would depend on my own ability to de-
ploy these symbolisms powerfully, and on the capacity of my audience to engage with them. The
conventions of legal academia being typically limited to the apprehension and analysis of text, a law-
review article in the form of a multimedia compact disc risks confounding rather than illuminating its
audience.
Finally I had considered producing the text of this essay in French or in alternating paragraphs of
English and French so as to reinforce this point about discursivity, but were I to have done so, I would
have run up against the paradox of Babel. A text in a foreign language will initially counter the
shackles that discursiveness places on readers. But absent a bilingual audience it risks being under-
stood simply as “pidgin” and not as the intended metaphor. See the Eighth Study in P. Ricoeur, The
Rule of Metaphor: Multi-disciplinary Studies of the Creation of Meaning in Language, trans. R.
Czemy (Toronto: University of Toronto Press, 1977) 257.
35It bears notice, however, that when a first version of this essay was presented in 1981, practically
every semantic point was rejected by the legal audience to which it was addressed. Despite its funda-
mentally orthodox philosophical premises, the text was dismissed as yet another “Fish-story” (sic),
the reference being to S. Fish, Is There A Text in this Class: The Authority of Interpretive Communities
(Cambridge: Harvard University Press, 1980), which had just been published. It is a measure of how
much more sophisticated legal scholarship has become that in the short space of 15 years, the semi-
otic assumptions of a “radical text” are now taken as relatively mainstream, even by those who tend
toward postmodern and critical legal studies positions. Compare D. Klinck, The Word of the Law:
Approaches to Legal Discourse (Ottawa: Carleton University Press, 1992) c. 1-4, 10, 11; P Goodrich,
Languages of Law (London: Weidenfeld & Nicholson, 1990).
1 The central issue is to understand how contingency arises and what resources we have, both
within and outside language, for dealing with it. For an exchange on this point, see B. Langille,
“Revolution Without Foundation: The Grammar of Scepticism and Law” (1988) 33 McGill L.J. 451;
A. Hutchinson, “That’s Just The Way It Is: Langille on Law” (1989) 34 McGill L.J 145. See also B.
Langille, “The Jurisprudence of Despair, Again” (1989) 23 U.B.C. L. Rev. 549.
19971
R.A. MACDONALD – LEGAL BILINGUALISM
pute that language is the translation of ideas and representations into a particular sym-
bolic form, and few dispute that this particular symbolic form only gives the appear-
ance of presenting precise meanings that are broadly shared within a particular com-
munity.
Yet we ought not to be transfixed by language or even by examples drawn from
the deployment of language. However special language may be as a mode of human
communication, its indeterminate and contingent elements are little different than
those inherent in all communicative symbolisms.” The problem of human communi-
cation – of translating one’s thoughts into a communicative symbol, of then deploy-
ing that symbol, and of attempting to respond to another’s translation and reexpres-
is universal. It follows that the so-called indeterminate
sion within that symbolism –
and contingent properties of language can be identified, highlighted and analyzed by
means of semiotic procedures common to human symbolisms generally?
10. Law, as a species of human endeavour, necessarily engages its practitioners in a
complex exercise of formulating and deploying symbols.” These symbols operate at
an instrumental level where they may be used as vehicles for expressing and com-
municating the meaning of everyday experience. But they are more than this. Sym-
bols exert a control over how ideas are expressed and, more fundamentally, over what
ideas can be expressed. They mould attitudes, structure thought and contribute to the
reconstruction of culture and community. Symbols and symbol systems, being as
varied as human experience, may relate to every kind of human communication and
to every kind of human capacity to interpret the messages of others. They may be vis-
ual (painting, text), aural (music, speech), olfactory (cooking, insence), physical
(dance, ritual) or multiple combinations of these (opera, theatre). The means for hu-
man symbolizing are limited only by the imagination of humans themselves.
In law and contemporary “modernist” legal theory, nonetheless, a summa divisio
of communicative symbols reigns: symbols are characterized as either linguistic or
nonlinguistic.” What is more, within the privileged category of linguistic symbols it is
” See S.K. Langer, Philosophy in a New Key, 3d ed. (Cambridge, Mass.: Harvard University Press,
1957) c. 1, 2.
” For examples of how this might be done, see D. Howes, “In the Balance: The Art of Norman
Rockwell and Alex Colville as Discourses on the Constitutions of the United States and Canada”
(1991) 29 Alta. L. Rev. 475; D. Howes, “‘We Are the World’ and its Counterparts: Popular Song As
Constitutional Discourse” (1990) 3 Int. J. Pol. Cult. & Soc’y 315; D. Howes, “La constitution de
Glenn Gould: le contrepoint et l’ttat canadien” in J.-G. Belley, ed., Le droit soluble: Contributions
quebecoises al l’tude de l’internormativitj (Paris: L.G.D.J., 1996) 95.
” See N. Elias, The Symbol Theory (London: Sage, 1991). I have tried to illustrate an understanding
of legal doctrine as symbol rather than as norm in R.A. Macdonald, “Reconceiving the Symbols of
Property: Universalities, Interests and Other Heresies” (1994) 39 McGill L.J. 761. Compare R. Kev-
elson, The Law as a System of Signs (New York: Plenum, 1988).
‘, See P.L. Berger & T. Luckmann, The Social Construction of Reality (New York: Doubleday,
1966).
” As others have noted, this bifurcation roughly tracks the distinction between made law and im-
plicit law – between the law emerging from specialized social institutions such as legislatures, courts
and law faculties, and the law emerging from human interaction. See e.g. L.L. Fuller, “Human Inter-
.132
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 42
crimes, property, torts –
text (the written) that reigns over speech (the oral). While much of the law immedi-
ately known to the ordinary citizen –
concerns the interpre-
tation and understanding not of what was said or written, but of what was done or not
done, jurists are typically preoccupied by the language of the law. Cases, statutes, jury
instructions, pleadings, negotiations, contracts, wills and marriage vows presuppose
language. As a result, even though the normativity of official law is, in no small
measure, rendered by nonlinguistic symbols,’ much professional legal activity de-
scribes itself as little more than the generation and interpretation of complexes of lin-
guistic symbols.” Law is, on this construction, held to be something close to secular
scripture.”
11. Language apparently exhibits two communicative properties better than other
symbolisms. First, its nominative capacity –
the fact that language permits explicit
denotations –
seems to give us mastery over objects and concepts. Like a totem, a
voodoo doll or a monarch’s crown, a name stands as surrogate for that which is
named, and functions as an instrument of subjugation.”‘ Unlike a photograph, lan-
guage enfranchises whoever bestows the name. In prompting those who name to ex-
plore, organize and take responsibility for this subjugation, the act of naming nor-
mally requires the differentiation of self and other to a greater degree than, for exam-
ple, the invocation of a totem. The act of naming thus reveals one of the communica-
action and the Law” (1969) 14 Am. L Juris. 1, and more recently, B. de Sousa Santos, “Law: A Map
of Misreading. Toward a Post-Modem Conception of Law” (1987) 14 J.L. & Soc’y 279.
“More generally, one might say that much legal normativity is in fact comprised of non-linguistic
symbols. See R. A. Macdonald, “Pour la reconaissance d’une normativit6 juridique implicite et
‘infrrentielle'” (1986) 18 Sociologie et Socidtds 47, for an exploration of the idea from a perspective
internal to law, and W.R. Janikowski, ed., Legality and Illegality: Semiotics, Postmodernism and Law
(New York: P. Lang, 1995), for the same idea from an external perspective. See, finally, P. Goodrich,
“Modalities of Annunciation: An Introduction to Courtroom Speech” in R. Kevelson, ed., Lav as
Semiotics, vol. 2 (New York: Plenum, 1988) 143; P. Goodrich, Oedipus Lex: Psychoanalysis, History,
Lanv (Berkeley: University of California Press, 1995).
41 See J.-G. Turi, “Norme linguistique et norme juridique” (parts 1-3) (1984) 53 Qudbecfranfais
78, (1984) 54 Qudbecfranpais 86, (1984) 55 Qudbecfranfais 74. Compare P Fitzpatrick, The My.
thology of Modern Law (New York: Routledge, 1992). For a rich historical perspective, see D. Man-
derson, “Statuta v. Acts: Interpretation, Music, and Early English Legislation” (1995) 7 Yale J.L. &
Human. 317.
” See J. Goody, The Logic of Writing and the Organization of Society (Cambridge: Cambridge Uni-
versity Press, 1986), especially c. 1, 4. As for constitutions, see M. Golding, “Sacred Texts and
Authority in Constitutional Interpretation” in R. Penode & J. Chapman, eds., Authority Revisited
(New York: New York University Press, 1987) (NOMOS 29) 267; T.C. Grey, “The Constitution As
Scripture” (1984) 37 Stan. L. Rev. 1. As for civil codes, see J.E.C. Brierley & R.A. Macdonald, eds.,
Quebec Civil Law: An Introduction to Quebec Civil Law (Toronto: Emond Montgomery, 1993) at
84ff; P. Legrand Jr., “Strange Power of Words: Codification Situated” (1994) 9 Tul. Eur. & Civ. L.F 1.
“1 Of course, this surrogate function of nouns and representational objects may also be seen in other
symbolisms: the ringing of a bell, the firing of a cannon, the smell of incense and the tasting of wine
perform a like function in various contexts. They can all be both signs and denotators (see Langer,
supra note 37 at 53-78).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
tive advantages of language: nouns appear to be more mobile than totemic objects
both in their denotations and in their metaphoric deployment.’
A second communicative property of language is its capacity, through conven-
tions of syntax and grammar, to facilitate the successive articulation of presumably
limited units of thought and experience. This iterative capability allows speakers and
writers to control chronology, suggest causal relationships and generate what appear
as logical arguments moving from premise to conclusion. English – perhaps less so
than languages that decline nouns and their modifiers, but more so than languages
that do not recognize grammatical distinctions between verbs and nouns – permits its
speakers to differentiate ‘The practice was perfect” from “Practice makes perfect”,
“Practice to perfect” and “Perfect the practice”. Each of these phrases can reasonably
be understood by speakers of English in at least two different senses, depending only
on the grammatical function of the words in question: is “perfect” a verb or a noun?
and is “practice” a noun or a verb?
12. These two features –
distinguish language from
other communicative symbolisms, at least as a matter of degree. But they do not make
language entirely subservient to human intention. Nouns, especially, have the power
to confound as well as to enfranchise. A vocabulary often limits what one wishes to
say and, at times, may even induce one to say what one does not wish to say. Words
contain unsuspected valuations, often by incorporating an unstated context, or even
by encapsulating an entire theory.” For this reason, employing a particular word in a
novel context can lead to confusion. And yet, using a term of art to challenge the ac-
cepted theory it encapsulates can provoke revelation, as did Darwin’s use of the word
“species”.”9 Words limit; words liberate; words confound; and words enfranchise?
vocabulary and grammar –
It is not just vocabulary that has the capacity to imprison; so too do grammar and
syntax. The structure of everyday language typically is linear and episodic, channel-
ling our apprehension of the world and our expression of ideas. For this reason, lan-
guage frequently seems incapable of reflecting the complexities of spatial and tempo-
ral relations. Linguistic presuppositions and stereotypes become so-ingrained that they
intellectual “boilerplate” – which are almost impossible to
create habits of thought-
46 For a development of this point, see F de Saussure, Course in General Linguistics, trans. W.
Baskin (New York: McGraw-Hill, 1966) at 68ff.
4’ These grammatical senses often can be signalled by means of emphasis in pronunciation; they are
much more difficult to convey in a written text without the use of conventions such as quotation
marks, italics, capitals and like devices. On these conventions, and analogous conventions in music,
see C.D. Stone, “Introduction: Interpreting the Symposium” (1985) 58 S. Calif. L. Rev. 1.
“This point is explored in the introduction of M.-C. Prmont, Le langage du droit (Ph.D. Thesis,
Faculty of Law, Universit6 Laval, 1995) [unpublished]; G. Lakoff & M. Johnson, Metaphors We Live
By (Chicago: University of Chicago Press, 1980) c. 1.
‘For a thoughtful elaboration of the theme, see J. Beatty, “What’s in a Word: The Problem of
Coming to Terms in the Darwinian Revolution” (Paper delivered at Concordia University, June 1980)
[unpublished].
‘ On these points generally, see J. Goody, The Domestication of the Savage Mind (Cambridge:
Cambridge University Press, 1977); Goodrich, supra note 35; Ong, supra note 32.
MCGILL LAW JOURNAL / REVUE DEDROITDE MCGILL
[Vol. 42
penetrate without a conscious departure from conventional syntax and grammar.”
More significantly, syntactical structures carry with them assumptions about indi-
viduation, causation and human agency. The power of poetry and other forms of lit-
erature derives largely from a self-conscious “abandonment to metaphor” –
that is,
using the conventions of grammar against its logic. 2
13. Jurists have long recognized that language does not arise in syntactical and con-
textual isolation.” They have, however, tended not to acknowledge in practice the
central role that symbolic analysis plays in apprehending the meaning of legal lan-
guage.’ Following Ernst Cassirer and Suzanne Langer, I distinguish two modes for
deploying and interpreting communicative symbolisms: the discursive and the presen-
tational.” Symbolisms in a discursive mode seem to isolate small units of understand-
ing and arrange thought as chronology; the meaning we ascribe to them seems to con-
front us serially. For example, oral directions given on the street comer to a tourist
seeking to find a particular attraction are most likely to be given and interpreted, at
least initially, discursively. Symbolisms in a presentational mode do not lend them-
selves as easily to fractionation, either temporally or spatially; they confront us, and
we understand them, as an undifferentiated whole. For example, a sculpted piece of
art is initially most likely to be revealed to, and interpreted by, the viewer presenta-
tionally.
Of course, no human communicative symbolism –
language, art, music, dance,
ritual –
is inherently and exclusively discursive or presentational. These two are
modes of symbol structure internal to symbols and modes of apprehension of symbols
particular to the roles of any symbolism in a given culture. Choices about style
(including rhetorical choices) made by the initiator of the communication, and choices
and interpretive strategies made by the target of the communication will, along with
cultural context, shape the communicative frame.’
” This idea is explored using the conventions of classical-music forms in D. Manderson, The Aes-
thetics of Law (D.C.L. Dissertation, McGill University, 1997).
2P. Ricoeur explores this “abandonment to metaphor” at length in Temps et ricit, 3 vols. (Paris:
Edition du Seuil, 1984).
” See e.g. the famous exchange between H.L.A. Hart (“Positivism and the Separation of Law and
Morals” (1958) 71 Harv. L. Rev. 593) and L.L. Fuller (“Positivism and Fidelity to Law – A Reply to
Professor Hart’
(1958) 71 Harv. L. Rev. 630). A whole textbook industry, devoted to constitutional
and statutory interpretation, and a whole subset of legal theory, devoted to deriving the ratio deci-
dendi of cases, depends on the syntactical and contextual features of legal interpretation. See e.g. the
symposia published in “Symposium: Law and Literature” (1982) 60:3 Texas L. Rev.; “Interpretation
Symposium” (1985) 58:1-2 S. Calif. L. Rev.; “Symposium Patterson v. McLean” (1989) 87 Mich. L.
Rev. 1-137. See also M.S. Moore, “The Semantics of Judging” (1981) 54 S. Calif. L. Rev. 151.
4 Two notable exceptions are J.B. White, When Words Lose Their Meaning: Constitutions and Re-
constitutions of Language, Character and Community (Chicago: University of Chicago Press, 1984);
and Klinck, supra note 35.
“See Langer, supra note 37, c. 4.
The thought is from D.C. Freeman, ed., Linguistics and Literary Style (New York: Holt, Rinehart
& Winston, 1970). See also S. Chatman, ed., Literary Style: A Symposium (London: Oxford Univer-
sity Press, 1971).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
14. “Natural languages” are most often deployed in everyday life for instrumental
purposes (e.g., oral statements of future conduct, or written assembly instructions for
household objects). In such usages their discursive properties are brought to the fore-
ground. One presumed advantage of grammar-based language over pictographs and
hieroglyphs is precisely that the rules of grammar and syntax permit writers and
speakers to exploit discursive communication. Conventions about sentence parsing,
about the relationship of subject, verb, direct object and indirect object, and about the
referent of phrases and clauses, in addition to a limited number of conjunctions and
prepositions, permit the juxtaposition of words and define the range of possible rela-
tionships that ordinary usage can accommodate. 7
Yet this property of discursivity is not always capable of management in simple
sentences. For example: it is relatively easy to say that a book is “on” a table, or
“under” a table; more difficult to say simply that a book is being taken from a position
on the table and placed in a position under the table; and very difficult to say simply
that a book is being taken from a position on top of another book on top of a table,
and placed in a position under another book under the table. These situations that
seem to require subordinate clauses and complex grammatical constructions are pre-
sumably the referents of the expression “a picture is worth a thousand words”. Con-
versely, depending on the potential range of the intended audience, a series of nonlin-
guistic pictoral instructions may even display greater discursive capacity than lan-
guage itself.”
15. Discursivity is more than a consequence of grammar and syntax; it flows also
from limitations on the aural and visual capacities of human beings. Orality in the
speech of one speaker is itself linear. The same is true in dialogue if cacophony is to
be avoided. Unlike musical notes that may be played simultaneously to produce a
harmonic even as they progress in temporal fashion as themes and subthemes, differ-
ent words are only with great difficulty made capable of such harmony when spoken
together; their logic is to follow each other in temporal sequences 9 So too with texts,
although their linear properties tend to be spatial rather than temporal. Conventions
for reading texts (e.g., in English but not Chinese, commence at the top left and pro-
ceed in a horizontal line to the right) permit authors to control, in large measure, the
order in which ideas are initially engaged. These same temporal and spatial conven-
tions work together to condition communication in other symbolisms such as film,
7 This is not to say that there do not exist equally elaborate conventions for reading hieroglyphics.
The point is only that the spoken rendition of most hieroglyphs proceeds discursively and not as one
sound.
“‘ Two relatively recent phenomena of the globalized economy evidence the point. Assembly in-
structions for knock-up furniture of the type sold by IKEA and emergency-procedures cards found in
airline-seat pockets both rely on sequentially numbered pictograms as discursive symbolisms.
” I acknowledge the phenomenon of singing in harmony; but here there is no conflict in the words
being sung. More difficult is the case of singing in harmonic rounds. Nonetheless, the temporal dis-
placement of the words of a round between groups of singers is designed for harmonic effect, even
when, in rare cases, entirely different texts are being sung by different groups of singers. We are
trained simultaneously to follow two or more musical themes, but we are not trained to follow two
conversations simultaneously.
136
McGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 42
theatre, mime, or dance. By contrast, most purely spatial visual symbolisms – paint-
and most olfactory symbolisms, ‘ are not as susceptible of simi-
ing, for example –
larly authoritative antecedent discursive control. They are, therefore, symbolisms that
tend initially to be apprehended primarily in a presentational mode.”‘
Just as conventions of space and time enhance the properties of some symbolisms
such as language that conduce to discursivity, it is also possible to reduce the reflex to
discursivity by explicitly transgressing these conventions. Poetic forms, flashbacks
and multiple images in film and theatre, subthemes and counterpoint in music, fore-
shadowing in dance and ritual, or combinations of these techniques can serve to dis-
rupt the linear logic of discursivity –
be it spatial or temporal. It is also possible to
enhance discursivity in symbolisms that are, in most cases, apprehended primarily in
a presentational mode. The use of colour and perspective, series paintings such as
Vassarelli’s C.TA. 102, comic strips, the mounting and display of sculpture to reveal
the work part by part, and the adherence to conventions for the interpretation of cer-
tain nonlinguistic symbols such as a tapestry, can offer clues for a discursive
“reading” of symbolisms that might initially be thought dominated by presentational
forms.
16. Jurists in Western legal traditions have chosen, or at least believe that they have
chosen, language as their primary symbolism of normative communication.’ Because
technical legal language is deployed and apprehended predominantly in a discursive
mode, not surprisingly, many jurists believe that law itself can be fully understood
discursively. One kind of discursivity, formal rationality, was identified by Max We-
ber as the authoritative standard to which Western law should aspire.’ Weber at-
tempted to distance the ideal of Western law from the deferential language forms as-
sociated with “hallowed traditions” and “oracular revelations”, which he character-
ized as “irrational”. Nevertheless, even formally rational language requires deference
to authoritative texts, and this deference ultimately emerges from tradition and reve-
l
60 See D. Howes, ed., The Varieties of Sensory Experience: A Sourcebook in the Anthropology of the
Senses (Toronto: University of Toronto Press, 1991); but contrast the complex rendering of the rela-
tively formalized rituals of eating in M. Visser, Much Depends on Dinner: The Extraordinary History
and Mythology, Allure and Obsessions, Perils and Taboos of an Ordinary Meal (Toronto: McClelland
& Stewart, 1987); M. Visser, The Rituals of Dinner: The Origins, Evolution, Eccentricities and
Meaning of Table Manners (Toronto: Harper Collins, 1991).
“To say that these symbolisms are initially apprehended in a presentational mode raises complex
questions of how it is that we learn a language. Is there a difference between acquiring the ability to
comprehend and speak a language and the ability to read and write a language? Similarly, is there a
difference between acquiring the ability to comprehend and make music, and the ability to read and
write music? Is the difference between presentational and discursive apprehension merely a matter of
education? Is learning discursivity a more complex, hence later, intellectual achievement? If so, why
is it that we also have to learn how to read a poem, listen to music and look at a painting? On these is-
sues, see generally Pinker, supra note 3.
‘ Yet this, it should be recognized, is a choice. For a fascinating exploration of one alternative, see
A. Lowenhaupt Tsing, In the Realm of the Diamond Queen (Princeton: Princeton University Press,
1993).
61 See M. Rheinstein & E. Shils, trans., Max Weber on Law in Economy and Society (Cambridge:
Harvard University Press, 1954).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
lation. The language of deference and authority occupies a significant place in law
beside the language of rational deduction. Both are necessarily commingled, much as
discursive and presentational forms are commingled.’
These two sets of distinctions between forms of legal normativity can be placed
on intersecting axes so as to construct a four-square typology. If one asks what are the
expressive types of legal language created by conventions of grammar, syntax, space
and time, one can distinguish between the discursive and the presentational. If one
asks what are the cognitive forms of legal language, one can distinguish between the
rational and the deferential. The following table plots these possibilities.’
Presumptive Formal Typology of Linguistic Legal Symbols
Discursive
Presentational
Rational
Legislation
Contractual boilerplate
Deferential
Judicial reasons for decision
General principles of law
Positing a bilingual legal order clarifies the interrelation between these kinds of legal
normativity. It helps to reveal both why rationalistic and discursive language has
tended to dominate most approaches to legal normativity and why deferential and
presentational normativity are, nonetheless, always in play.
17. Take, first of all, the archetypal discursive-rational language symbol in official
legislation. It is often said that a statute simply means what its words literally
law –
say. But as soon as an apparently rational and discursive language symbol like a stat-
ute is made in two languages, its deferential and presentational aspects become more
apparent. The enactment of bilingual statutes in Manitoba, New Brunswick and On-
tario has created a presentational symbol far more significant than the discursive
content of any of the legislative texts so presented.’ While bilingualism heightens our
sense of the presentational element of statutes, this element persists even where texts
do not exist in two or more language versions. In the first half of the twentieth cen-
tury, for example, the Civil Code of Lower Canada attained scriptural status among
French-speaking jurists in Quebec.’7 Perhaps the most striking contemporary example
–
supra note 22, 72.
University of Chicago Press, 1986).
“This commingling is examined in J. Vining, The Authoritative and the Authoritarian (Chicago:
65 The irony of deploying a table – a visual symbolism that primafacie suggests presentationalism
to explicate the seemingly discursive text of a legal essay surely bears notice.
“See A. Bradn, “Le bilinguisme dans le domaine Idgislatif’ in Les droits linguistiques au Canada,
67 See the story as presented in Brierley & Macdonald, supra note 44 at 67-73. See further J.-L.
Baudouin, “Confdrence de clture” in Enjeux et valeurs d’un Code civil moderne (Montreal: Th6mis,
1991) (Jourmes Maximilien-Caron 1990) 219; a series of critiques of this reverential fallacy are de-
veloped in P. Legrand Jr., “Civil Law Codification in Quebec: A Case of Decivilianization” (1993) 1
Z. Eu. P. 574; P. Legrand Jr., “Bureaucrats at Play: The New Quebec Civil Code” (1995) 10 Brit. J.
Can. Stud. 52; Legrand Jr., supra note 13; and P. Legrand Jr., “Civil Codes and the Case of Quebec:
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 42
is the Constitution of the United States, which has become an icon for the American
citizen.9
Once a legislative text is understood as having a presentational meaning, its
authority is enhanced and its capacity to function as a deferential symbol enlarged.
Such a text can become, for true believers, a pure object of deference. This is espe-
cially true where, as in the cases of the Torah, the Bible and the Koran, belief is itself
grounded in a deferential context such as religious tradition.’ Correlatively, once a
discursive legislative text is seen overtly as an object of deference, its normative
weight increases and its presentational symbolism comes to predominate. The notion
of “fundamental rights” is among those legal concepts most tributary to presentational
interpretation by most Canadians. Hence, for many, the Canadian Charter of Rights
and Freedoms is emerging as a reverential “sacred text”.”0
18. Modem official law also relies heavily on discursive-deferential texts: arche-
typally, the written judgments of adjudicative bodies. Where “reasons for decision”
are given in more than one language, however, the presentational function of lan-
guage is often engaged. Those who have struggled to ensure that courts entertain ar-
guments and render judgments in both English and French are not simply concerned
with expanding the range of legally cognizable justificatory material available for
deployment by adjudicators. They are concerned as well with ensuring that the courts
and their decisions in some sense belong to minority language communities. This ef-
fort would hardly have been necessary were the business of adjudication merely dis-
cursive and rational. For example, a bricoleur really does not care to retrace the
mathematical operations by which the members of a truss-roof are located. Clearly,
some goal besides the correct deduction of the legal outcome sustains the bilingualiz-
ing endeavour.7′
(New York: Alfred A. Knopf, 1986).
69 This theme is developed in B. Polka, The Dialectic of Biblical Critique: Interpretation and Exis-
Semiotic Musings around an accent aigu” in R. Kevelson, ed., Conscience, Consensus, and Cross-
roads in Law: Eighth Round Table on Law and Semiotics (New York: P. Lang, 1995) c. 13
[hereinafter “Semiotic Musings”].
9 See e.g. M. Kammen, A Machine That Would Go of Itself: The Constitution in American Culture
tence (New York: St. Martin’s Press, 1986).
70 Pejoratively, such zealots may be characterized as Charter-patriots. On Charter-patriotism, com-
pare M. Mandel, The Charter of Rights and the Legalization of Politics in Canada, rev. ed. (Toronto:
Thompson Educational, 1994) with A.C. Cairns, Charter versus Federalism: The Dilemmas of Con-
stitutional Reform (Montreal: McGill-Queens University Press, 1992) and with D.M. Beatty, Consti-
tutional Law in Theory and Practice (Toronto: University of Toronto Press, 1995). Reverential inter-
pretation of the Charter is examined in B. Polka, “The Supremacy of God and the Rule of Law in the
Canadian Charter of Rights and Freedoms: A Theologico-Political Analysis” (1987) 32 McGill L.J,
854.
71 See M. Bastarache, “Le bilinguisme dans le domaine judiciaire” in Les droits linguistiques au
Canada, supra note 22, 129. Of course, there is a different argument to be made for enacting legal
rules in both the French and English language, namely, that citizens have a legitimate expectation of
being able to understand the law that is applicable to them. But this argument simply exhausts itself in
multilingual societies such as Canada. Instrumental effectiveness and moral legitimacy apply just as
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
The reason why it is important to be able to argue before a court in one’s own
language owes much to the rhetorical power of language. If the norm itself is avail-
able in one’s language and if judicial judgment is simply the deductive application of
the norm, pleading in one’s language and receiving judgment in one’s language is a
superfluous luxury. But if a judgment is the rhetorical act of convincing, its presenta-
tional elements are equally important. The process of argumentation in law is more
than a process of rational justification; it is also a process of presentational dialogue.’
19. Symbolisms in an apparently presentational mode, such as a stop sign or contrac-
tual boilerplate, can evoke deference but they can also provoke other responses from
the agent who acknowledges the norm. To respond, “stop signs should be obeyed be-
cause they help to coordinate the behaviour of all motorists,” is to recall how legal
normativity can be revealed in a presentational-deferential text. Rendering a presenta-
tional symbol in two languages highlights the connection between its discursive ele-
ment and its normative meaning. Those who have gone to the trouble of erasing the
word “stop” on signs in Quebec do so because they are not convinced that “stop” and
“arrf’ are simply equivalent. To them, the sign reads very differently if the word
“stop” is blocked out. Indeed, Quebec’s language legislation would be incomprehen-
sible were there not also a presentational meaning to apparently discursive words.’
Most of Canada’s official law seems far removed from the presentational-
deferential mode typically found in literature or religious canon. Nonetheless, its gen-
eral principles, which remain largely unarticulated and undetermined, inform legal
normativity. Unexplored assumptions, the generative ideas of a legal order, permit le-
gal normativity to appear discursive by dispensing with inquiry about certainty. Law
itself symbolizes a rationality that can never be fully achieved. We defer to the law as
sound and principled, but we are not able to express in a definitive way the grounds
for our deference. The presence of a second language at least serves to reassure us
that this deference is not linguistically arbitrary. To claim that norms must be ex-
pressed and that their interpretation and application must be justified in more than one
language is also to claim that whatever the semiotic importance of discursive, ra-
tionalistic modes for their apprehension, they must also be apprehended in presenta-
tional, deferential modes.
20. The object of posing the semantic question has been to show that the mere pos-
sibility of legal bilingualism can reveal something of the interrelationship between le-
gal normativity and the manner of its expression in language. The discursive and ra-
much to aboriginal peoples and to immigrants who speak neither French nor English, yet apart from
aboriginal peoples, few have claimed the need for multilingual legislation. The argument, that is, rests
primarily on symbolic and not on instrumental grounds.
” The locus classicus is, obviously, Ch. Perelman, Traitd de l’argwnentation: la nouvelle rhitori-
que, 5th ed. (Bruxelles: Universitd de Bruxelles, 1988). See also Ch. Perelman, Logique Juridique:
nouvelle rhiftorique, 2d ed. (Paris: Dalloz, 1979).
3 For a further elaboration of this idea, developed in respect of the use of the French word Quibec
in the title of the English language version of the Civil Code of Quebec (sic), see P. Legrand Jr., “Civil
Codes” in Kevelson, ed., supra note 67; J.E.C. Brierley, “Les langues du Code civil du Qu6bec” in P.-
A. C6t6, ed., Le nouveau Code civil: interprdtation et application (Montreal: Th6mis, 1993) 129.
140
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 42
tional is intertwined with the presentational and deferential even in language-
dependent legal symbolisms. Legal knowledge does not exhaust itself in the discur-
sive and rational apprehension of texts, as the contemporary Kabbalists would have it.
Nor does it exclusively depend upon the meaning of those texts being reduced to dis-
cursive rationality, as latter-day realists would have it. Legal knowledge transcends
language, and this invites consideration of law’s epistemologies.
II. The Epistemology of Legal Bilingualism
21. Discursive rationality is not only a mode of symbol structure internal to symbols;
it is also a mode for interpreting communicative symbolisms. Nowhere is its power
more evident than in that symbolism we characterize as language. After all, language
is the symbolism of choice precisely in those situations in which we want to express a
train of limited ideas in a definite order, the instruction manual being the archetypal
use of language in which both structure and meaning seem only discursive.” Discur-
sivity facilitates an economy of expression which, in combination with rules of
grammar and the reification of concepts, seems to make language “logical”.
But the ascription of meaning cannot be equated with the act of naming. Meaning
implies, at the very least, a congeries of relationships, and not simply a denotational
inventory of objects to which a word is appropriately applied. It follows that, while
conventional explicit denotations may be catalogued, together they do not comprise
literal meaning. A literal meaning is no more than a stylized teleological argument in
which the range of factors thought relevant to the discovery of context is conven-
tionally and artificially constrained.” A speaker or writer expresses thought through
speech or text, and a listener will ascribe a meaning to the sound or printed mark. The
quality of communication between them will depend on their ability to engage in a
shared and interactive process of ascribing meaning. ‘
22. Each legal act or judgment involves a complex endeavour of characterization and
nominative ascription. Take the case of a female store clerk, picketing in a lawful la-
bour dispute, who is then charged with trespassing by the manager of a shopping
centre. ‘ Whether this dispute is argued and decided as a crime, tort or labour-relations
issue, or as an issue of private property, freedom of speech or women’s rights is never
decided solely by the explicit denotations of the legal terminology invoked. Quite
7, Of course, in moments of anguish, reflection, sorrow and joy particularly, we directly confront the
impoverishment of discursive rationality. Here our points of reference tend to the presentational and
deference; here we know precisely why “words fail us”; and here we know that formula and ritual are
meaningful precisely because they are not being deployed and apprehended in a mode of discursive
rationality. Once again, see Langer, supra note 37 at 83-88.
” See Klinck, supra note 35 at 91-102. But compare L. Raucent, “Droit et linguistique – une ap-
proche du formalismejuridique” (1978) 19 C. de D. 575.
76 From the voluminous literature, one might profitably consult B.S. Jackson, Law, Fact and Narra-
tive Coherence (Merseyside: Deborah Charles, 1988); D.R. Klinck, Book Review of Law, Fact and
Narrative Coherence by B.S. Jackson (1992) 7:1 Can. J.L. & Soc’y 224.
“The reference is, of course, to Harrison v. Carswell, [1976] 2 S.C.R. 200, 62 D.L.R. (3d) 68,
[1975] 6 W.W.R. 673.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
apart from characterizations of such events resulting from the application of analytic
tools in disciplines that do not depend on law’s conceptual apparatus (such as eco-
nomics, sociology, political theory or whatever), the categories of legal self-ascription
themselves are multiple. The framing of an issue, the interpretation of a case, and the
analysis of legislative instruments all reveal the importance of implicit connotation
arising within the legal order. Yet, legal language also reflects popular language and is
constantly coloured by non-legal denotations and connotations. Because implicit con-
notations cannot be controlled, explicit denotations will ultimately change; changing
explicit denotations will in turn change implicit connotations.7
To recognize that discursive modes necessarily carry meaning associated with
presentational modes is to acknowledge that writer and reader (speaker and listener,
artist and viewer, composer and listener) engage in making explicit the possibilities of
knowledge that were implicit in the symbolism being apprehended. At the same time,
the evolving catalogue of discursive statements replenishes a meaning that can only
be described as inchoate: that which is known, but which the knower does not yet
know is known. Inchoate meaning is partly shared and partly personal: shared, be-
cause the attainment of meaning in any text or symbol draws one beyond oneself; and
personal, because to symbolize is to engage in self-revelation. Meaning does not to-
tally inhere in a text; nor is it definitively given to a text by an author; nor is it
uniquely ascribed to a text by a reader. To read a text is to undertake a “tri-lectic” with
the text as text, with its author and with oneself, in which a constellation of experi-
ences, values, fears and aspirations – many of which are only dimly perceived –
are
brought to bear on the processes of naming, sorting, recognizing and ignoring the
larger part of the world about oneself. 9
23. Each act of interpretation is an act of will by which interpreters impose them-
selves upon their world; but each judgment is also an act of submission, for each in-
terpretative act commits interpreters to a future judgment shaped by past understand-
ing. Seeing the ascription of meaning to communicative symbolisms as active inter-
pretation has two major implications in a legal context. First, the conception of nor-
mativity in legal texts as exclusively a priori cannot be sustained. Neither statutes nor
judicial decisions have one correct meaning. The canonical language of statutes and
the mythology of legislative intention suggest the possibility, but the actual practice of
Except in very highly technical fields – biology and physics come to mind-
scientific words are
either derived from non-scientific usage, or quickly pass into everyday use. Today, this is manifestly
the case with “computer-speak”: hardware, software, abort, interface, download, network and so on.
A brilliant historical demonstration is provided in J. Gordley, The Philosophical Origins of Modern
Contract Doctrine (Oxford: Clarendon Press, 1991), which traces how legal language, popular lan-
guage and philosophical language interpenetrate, and how, once the philosophical context of legal
doctrine is lost, the continued coherent deployment of the concepts central to such legal doctrine is
impossible.
79 I know no better attempt to explore this idea than M. Polanyi, Personal Knowledge: Towards a
Post-Critical Philosophy (New York: Harper & Row, 1964).
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 42
interpretation is to the contrary.”‘ Moreover, since the holding of a case exists only in
relation to other decisions in a related field, and since that field is constantly chang-
ing, it is virtually impossible for someone to pick one case and, having read it, appre-
ciate much of its potential legal meaning.” Absent this larger context, a reader would
have no standpoint from which to discern whether what is said in a statute or case is
different or important.
Second, the addressee of a legal text assumes a critical role in discerning its po-
tential meaning. To claim that interpretation engenders a “tri-lectic” among reader,
writer and text means that a conception of the community of possible readers is es-
sential for any author. In law, unfortunately, the relevant community has rarely been
thought to be universal. While statutes and judicial decisions are widely available,
their usual audience is a restricted group –
lawyers and judges. Again, the transfor-
mation of unwritten presentational symbolisms (ritual, tradition, custom) into written
discursive language often abets the contraction of community.’ The tragedy of much
modem official legal writing –
and even unofficial legal writing such as the present
essay –
is its renunciation of any attempt to expand the community of readers.” The
tragedy is particularly evident in statutes such as the Income Tax Act.”
0 See generally P.-A. Ct6, Interpritation des lois, 2d ed. (Cowansville, Que.: Yvon Blais, 1990) at
3-6, 235-37 and sources cited. See also R.A. Macdonald, “On the Administration of Statutes” (1987)
12 Queen’s L.J. 488.
” The literature on the derivation of the ratio decidendi is both rich and enormous. The best elabo-
ration of the preset point may be found in the introduction to A. Harari, The Place of Negligence in
the Law of Torts (Sydney: Law Book, 1962) at 5-11. See also G.J. Postema, Bentham and the Com-
mon Law Tradition (Oxford: Clarendon Press, 1986).
See Goody, supra note 44, c. 4.
‘3The point, of course, is not that every possible reader will read. Rather, where the moral ambition
is to write so that all might read, the character of writing, the character of what is written and the
character of membership in a community will be changed. For speculations on this theme, see L.L.
Fuller, The Morality of Law, rev. ed. (New Haven, Conn.: Yale University Press, 1969) c. 2, 5. Im-
plicitly, the purpose of “plain language” movements is not just to make law accessible; it is to enfran-
chise a greater proportion of citizens by implicating them in the process of legal elaboration. On the
relationship between literacy and enfranchisement, compare R.A. Macdonald, “Access to Justice and
Law Reform” (1990) 10 Windsor YB. of Access Just. 287 with E. Ryerson, Report on a System of
Public Elementary Instruction for Upper Canada (Montreal: Lovell & Gibson, 1846) and N. Mac-
donald et aL, Egerton Ryerson and His 7imes: Essays on the History of Education (Toronto: Macmil-
lan, 1978).
,Income TaxAct, R.S.C. 1985 (5th Supp.), c. I. Philip Vineberg put it this way:
At first reading, the new tax law defies understanding. The opening words of the Bi-
ble in Hebrew are: Brayshis boroh elohim ess hashomayim vhoretz. These six words
are extended in the King James’ version to nine: “In the beginning God created the
heaven and the earth”. In the style and phraseology of the tax draftsmen, it would read
somewhat as follows:
At a time prior to the first taxation year of the taxpayer, and next immediately pre-
ceding the commencement of the frst taxation year of the world’s first taxpayer, the
Supreme Authority, as defined in section 248(c), and hereinafter referred to as the
‘S.A.’ (whose decision is final and binding and free from any review, whether by judi-
cial determination or otherwise, except only for right of cross-appeal by the Minister of
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
in a particular text, for example –
24. Analytically, in Western legal traditions legal norms may be distinguished along
two formal axes. If one seeks to highlight the way in which their meaning is ex-
pressed in language –
a distinction can be drawn
between formulaic (or canonical) norms and inferential (or metaphorical) norms. If
their mode of elaboration is highlighted, a distinction can be drawn between explicit
(institutionally stated) and implicit (interactional) norms.85 The resulting typology,
which parallels the typology of language forms discussed previously, is also illumi-
nated when plotted against the requirements of legal bilingualism. It reveals that the
model of law as purely explicit and formulaic is implausible, and therefore that inde-
pendent, isolated, parallel language texts cannot really exist in a bilingual or multilin-
gual legal order. Interpretation presupposes that, whatever the presumptive form of a
legal norm, retrieving its substantive meaning is an exercise that depends on working
through each of these forms as applied to that norm.
Presumptive Formal Typology of Legal Norms
Explicit
Implicit
Formulaic
Legislation
Custom, tradition
Inferential
Judicial reasons for decision
General principles of law
25. Explicit formulaic norms, such as those thought to be found in statutes, are meant
to express discrete, authoritative propositions. Such norms lack narrative structure and
assert propositions without supplying an argument. This was not always true of stat-
utes. One striking example, of continuing importance in Canada, is the Royal Procla-
mation of 1763, which reads, in part:
And whereas great Frauds and Abuses have been committed in purchasing
Lands of the Indians, to the great Prejudice of our Interests, and to the great
Dissatisfaction of the said Indians; In order, therefore, to prevent such Irregu-
larities for the future, and to the end that the Indians may be convinced of our
Justice and determined Resolution to remove all reasonable Cause of Discon-
tent, We do, with the Advice of our Privy Council strictly enjoin and require,
National Revenue), produced, grew, mined, created, manufactured, fabricated, im-
proved, packaged, preserved, and constructed, in whole or in part, not more than the
greatest and not less than the least of the aggregate of:
i)
the maximum upper extension of the minimum areas above the universe, as de-
fined in section 251;
ii) the minimum lower extension of the maximum areas below the universe; and
iii) the area in between.
Not many readers would have waited until the exciting part where Adam and Eve are
expelled from the world’s first tax heaven. The Bible wouldn’t have become a best-
seller (“Understanding the New Tax Law” [1972] Meredith Mem. Lect. (Twelve Lec-
tures on Income Tax Laws (Montreal: Wilson & Lafleur)) 7 at 7).
85This typology is derived from Macdonald, supra note 42.
MCGILL LAW JOURNAL! REVUE DE DROIT DE MCGILL
[Vol. 42
that no private Person do presume to make any purchase from the said Indians
of any Lands reserved to the said Indians, within those parts of our Colonies
where, We have thought proper to allow Settlement…
[The French version reads:]
Attendu qu’il s’est commis des fraudes et des abus dans les achats de terres
des sauvages au pr6judice de Nos int6rts et au grand m6contentement de ces
derniers, et afin d’emp&cher qu’il ne se commette de telles ingularit6s A
l’avenir et de convaincre les sauvages de Notre esprit de justice et de Notre r6-
solution bien arr&6e de faire disparaltre tout sujet de m6contentement, Nous
d6clarons de l’avis de Notre Conseil priv6, qu’il est strictement d6fendu 4 qui
que ce soit d’acheter aux sauvages des terres qui leur sont rdserv6es dans les
parties de Nos colonies, ott Nous avons cru A propos de permettre des 6tablis-
sements…
The French version contains two interesting differences from the English version:
(1) the word “Indian” is translated as “sauvage”, and (2) the term “all reasonable
Cause of Discontent” is translated as “tout sujet de micontentement”. Neither of these
differences concerns the explicit formulaic norm announced at the end of the clause,
but each significantly colours the rhetorical effect of the provision. If one reads both
versions on the assumptions that the French version is not a translation of the English
version (admittedly false in fact), and that both do have equal normative weight
(admittedly false in practice), one has a subtler sense of the narrative contained in the
Proclamation. The terms “sauvage” and “all reasonable Cause of Discontent” insert
elements of paternalism not fully present in the other version. The reader need not ask
which version is more authentic (indeed, here the paternalism alternates). Nor is it
possible to find a common meaning for each pair of terms. One must supplement one
version with the other and recognize that the text is incomplete without both. The
presence of an equally authoritative set of propositions in two languages that must be
reconciled can force an analysis of the spirit, intent and objects of an enactment –
precisely those elements that are not explicit or formulaic.
26. Explicit inferential norms, such as those expressed in court decisions, are meant
to state premises, develop an argument and reach conclusions. Insofar as judges are
writing “opinions”, they are attempting to persuade, and persuasion is intimately
linked to stylistic choice.’ Style largely depends upon the language in which it is ex-
pressed: its grammar, idiom, vocabulary and literary tradition. While good, even ex-
cellent, translations are possible, they can never identically capture the persuasive
16Royal Proclamation, 1763 (U.K.), 3 Geo. 3, reprinted in R.S.C. 1985, App. II (Constitutional
Acts and Documents).
” See D.R. Klinck, “Style, Meaning and Knowing: Megarry J. and Denning M.R. in In re
Vandervell’s Trusts (No. 2)” (1987) 37 U.T.L.J. 358 [hereinafter “Style, Meaning and Knowing”];
D.R. Klinck, “‘Criicizing the Judges’: Some Preliminary Reflections on Style” (1986) 31 McGill L.J.
655.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
authority of the original judgment. One sacrifices either normative precision or rhe-
torical impact.’
To the extent that judgments are understood to culminate in a ratio decidendi,
they are assimilated into the model of explicit formulaic norms. If there are such
things as rationes decidendi, one might believe that at least they can be exactly dupli-
cated into another language. A sophisticated understanding of legal normativity, how-
ever, once again serves to challenge this approach by highlighting that one loses part
of the judgment in not knowing the entire original decision. Whatever else it may be,
a case headnote is not an adequate precis of a judgment’s normativity.” The lesson is
that one must simply be able to act through “committed practices” in both languages
if one wants fully to understand legal normativity in a bilingual legal order. Ulti-
mately, this means not only the ability to speak, read and write in both languages, but
also the ability to plead, draft and judge in both languages.
27. Implicit formulaic norms, such as those constituted in and acted upon through
custom or trade usage, are meant to describe a practice so as to make it possible to
engage in that practice. Here, language is not used to construct an authoritative text.
The only “text” is the practice itself. Of course, the practice may become sufficiently
canonical that it is ritualized, as in the formalities for nominate contracts such as emp-
tio-venditio in Roman law, or that it even becomes expressible in a formula. If that
formulaic description of the practice should then become authoritative, as when a
custom is referentially incorporated into a statute or when it is explained in a judicial
decision, the norm is transformed from implicit to explicit. But should the customary
canon remain unauthoritative, the canonical text by which it is expressed also remains
a reflection of an implicit, formulaic norm.”
Where a practice is shared by speakers of different languages, the textual differ-
ences that might emerge through attempts to state the practice canonically are only
latent in the norm. The possibility of shared practices despite differences in the lan-
guage of practitioners holds out the promise of explicit norms that are not arbitrary. A
bilingual legal order brings to consciousness how any given language translates nor-
mativity. Its practice also serves to identify the movement from implicit to explicit
norms, much as it serves to identify the counter-movement from explicit to implicit
norms.
“On this point, see the sensitive unpacking of the work of Lord Denning in D.R. Klinck, “‘This
Other Eden’: Lord Denning’s Pastoral Vision” (1994) 14 Oxford J. Legal Stud. 25.
” The idea that rules themselves are complex narratives is not new to legal theory. See e.g. R.M.
Cover, “The Supreme Court, 1982 Term – Foreword: NOMOS and Narrative” (1983) 97 Harv. L.
Rev. 4; R. West, “Jurisprudence as Narrative: An Aesthetic Analysis of Modem Legal Theory”
(1985) 60 N.YU. L. Rev. 145. See especially Jackson, supra note 76 at 97ff.; its implications for the
normative impossibility of headnotes are obvious.
” This point is developed in R.A. Macdonald, “Custom Made” (Research paper prepared for the
Law and Society Programme of the Canadian Institute for Advanced Research, initially presented as
the Lansdowne Lecture at the Faculty of Law, University of Victoria, 18 January 1990)
[unpublished].
MCGILL LAW JOURNAL! REVUE DE DROITDE MCGILL
[Vol. 42
28. Implicit inferential norms, such as those which constitute the general principles
underlying a particular legal order (i.e., public-policy norms, political and economic
ideologies, tacit principles of justice and conceptions of the possibility of human
agency) are typically made patent in dialogue and debate centring on what these
principles are. One might go so far as to say that the substance of the principles is in-
separable from the question as to what they actually might be in a given case.” It is
worth stating the obvious: jurisprudential arguments about the foundations of law are
also of this character. 2
While the problems of jurisprudence have stimulated various programmes for
their solution in different places at different times, debate and dialogue have not been
precluded by a difference in language between various intervenants in these debates.
Ideas of law and legal theory typically cut across linguistic boundaries.” This suggests
that the “arbitrariness” of any single or even all natural languages, in and of them-
selves, is not the source of theoretical debate. The coexistence of multiple languages
in debates about law’s finalities and about general legal principles points away from
radical scepticism about the possibility of normative meaning opened by reference to
language.’ A bilingual legal order reinforces this insight: the multiple languages exist
not across legal orders, but within a single legal order.
29. The object of posing the epistemological question has been to suggest that the
possibility of legal bilingualism can reveal something of the way legal knowledge is
apprehended, rendered and transmitted through different forms of language symbol-
ism. To acknowledge legal bilingualism is to reveal why norms expressed in texts is-
sued by authoritative institutions are inseparable from norms having neither authori-
tative nor textual expression. To work through how these norms can be understood
bilingually is to describe legal bilingualism itself.
Ul1. Theorizing about Legal Bilingualism in Canada
30. The general considerations of the previous two sections now permit some an-
swers to be given to the question raised in the Introduction to this essay: Given legal
” See “Cinqui6mes Journaes juridiques franco-hongroises” (Symposium: les Principes g6n6raux du
droit, Budapest, 13-17 October 1980) [1981] R.I.D.C. 180-91.
2R. Cotterrell, The Politics of Jurisprudence: A Critical Introduction to Legal Philosophy (London:
Butterworths, 1989) nicely explores why these so-called “academic” debates matter to the politics of
legal practice. See also L.L. Fuller, The Problems of Jurisprudence, temp. ed. (Brooklyn: Foundation
Press, 1949).
” This is not, of course, to claim that language is irrelevant to the relative weight of different ap-
proaches and that these approaches are identical regardless of the language in which they are adum-
brated. Compare D. Davidson, “Reply to Foster” in G. Evans & J. McDowell, eds., Truth and Mean-
ing: Essays in Semantics (Oxford: Clarendon, 1976) 33.
, There is, of course, the response that even the deconstructive writing is itself contingent and
makes no claims to intersubjectivity. See L. Hutcheon, A Poetics of Postmodernism: History, Theory,
Fiction (New York: Routledge, 1988). But the point being made in the text is not that one can ascribe
non-contingent rationality to language. It is only to note that communicative symbolisms are not life-
less –
they are communicative.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
bilingualism, how is it possible? Bilingual legal texts will normally be created either
as two originals which are then reconciled as against a more general expression of
ministerial intention formulated (orally or in writing) in one language, or, as more of-
ten has been the case historically, as one original which is then translated.! Hence, it
is tempting to begin with a brief excursus on the theory of translation. There are, of
course, many “theories” of legal translation as literary (and legal) practice.’ Regard-
less of the actual mechanics of translation in any given context, inevitably Steiner’s
Law applies: all interpretation is an act of translation, and all translation is an act of
interpretation.’ This is why theorizing about legal bilingualism in a particular legal
order is helpful in rendering patent the epistemological and semantic foundations of
deploying language to apprehend legal normativity.
31. Recall the two paradigmatic types of discursive legal norm: legislation and rea-
sons for judgment given by an adjudicative body. Despite their differing authoritative
status and textual mode of expression, as texts, both rest on identical epistemic and
semantic grounds. Legislative drafting involves the same kind of interpretive exercise
as decisional writing by courts and other officials. Both require the translation of a le-
gal idea, into language or into a new language. Sometimes this exercise of translation
implies the initial drafting of a piece of legislation or the discovery and application of
legal norms that previously had no canonical expression in language (customary and
common-law rules). In these cases, the point of entry for the constitutive or reconsti-
tutive effort is practice, not text. Action, relationships and even entire lives have to be
apprehended as normative artifacts that contain both explicit and implicit meanings.
Conversely, sometimes this normative exercise implies the amendment of an ex-
isting text in one or more natural languages (or even the translation of a legislative
text from one natural language into another); likewise, sometimes it implies the dis-
covery and application of legal texts already expressed in canonical form in one or
more natural languages (or even the translation of reasons for judgment from one
natural language into another). In these cases the reconstitutive process is a search for
” Until recently, the practice of legislative drafting at the federal level was of the latter type (see
L.-P. Pigeon, “La r6daction bilingue des lois f&6rales” (1982) 13 R.G.D. 177), whereas, since the
mid-1980s, it has been one of co-drafting. See P. Johnson, “Bilingual Drafting: The Government of
Canada System- Part r’ (Paper presented to the National Seminar on Legislative Drafting and Inter-
pretation, Ottawa, 19-21 August 1987) [unpublished]; L. Levert, “Le bijuridisme 16gislatif’
(Manuscript, 25 March 1996) [unpublished]. See also A. Morel, “La r~daction de lois bilingues har-
monis6es avec le droit civil” (Manuscript prepared for the Federal Department of Justice, September
1996) [unpublished].
96 For a sampling, see J. Kerby, “Problmes particuliers a la traduction juridique au Canada” (1979)
12:2, 3 R. de l’Universit6 de Moncton 13; “La traduction juridique” (Symposium, Australia, August
1986) (1987) 28 C. de D. 735-859; G6mar, supra note 7; PW. Schroth, “Legal Translation” (1986) 34
Am. J. Comp. L. 47 (supp.); A. Covacs, “Bilinguisme officiel et double version des lois” (1979) 24
META 103; J.-C. G6mar, “La longueur des textes en traduction juridique- domaines anglais et fran-
qais” in P Pupier & J. Woehrling, eds., Language and Law (Montreal: Wilson & Lafleur, 1989) 599.
See also G6mar, ed., supra note 9.
” See Steiner, supra note 2, preface and passim.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 42
the meaning of a legal norm whose apprehension is already partially controlled by a
text (be it an enactment or a judgment).
32. A legal norm comprises more than just the words that serve to give an explicit
point of reference for its apprehension. Words, phrases, and even entire statutes, codes
and judicial decisions are normative symbols that contain both explicit and implicit
meanings.’ The problem of legal bilingualism is one of legal interpretation/translation
conceived in this more embracive sense. This problem is at once lesser and greater
when an authoritative text rather than an action is the symbolic referent –
lesser be-
cause of discursive conventions of language already noted, and greater because these
conventions truncate the search for meaning. Authoritative texts in two or more lan-
guages raise three discrete interpretive challenges. These flow from: uncertainty about
the relative authority of the two versions; doubts about the degree to which either re-
flects the norm; and the embedded character of language.
33. Take first those challenges relating to the manner in which texts are created. Here
the focus is on legislation. If legal bilingualism presupposes equal authority of both
versions of a text, how ought the interpreter to react when one such version is patently
a derivative translation of the other?’ Given the draftsperson’s necessary fidelity to
textual symmetry, the implicit and symbolic meanings of the primary version will in-
evitably be lost, ignored or compromised for pragmatic reasons in any translation.”‘
For this reason, it is doubtful that legally significant one-to-one translations are even
possible, even where the principal nouns in the norm are transliterated terms of art:
contract, offer, acceptance, capacity, cause. Such an exercise is predicated upon the
dubious proposition that words (and especially legal terms of art) carry with them
detachable, fixed meanings that can be derived from a bilingual dictionary.”
34. As to the second challenge, even where both language versions have been drafted
as originals, the ministerial directive giving rise to the drafting exercise inevitably will
be expressed in one or the other language. Consequently, the implicit problem being
addressed, and the implicit solution being sought, will be shaped by the perspective of
the person requesting the legislative text.” Again, regardless of the language of the
” The point is nicely developed in the special case of legal fictions by R.A. Samek, “Fictions and
” This, for example, is the case of the Civil Code of Qubec (see Brierley, supra note 73, especially
the Law” (1981) 31 U.T.L.J. 290.
at 144-45).
“‘ The point is developed in L.-P. Pigeon, “La traduction juridique – L’quivalence fonctionnelle”
in G~mar, ed., supra note 9, 271; see also R Newmark, “The Translation of Authoritative Statements:
A Discussion” in Grmar, ed., ibid., 283. See below, text accompanying notes 146-62, for examples of
changes to the manner of drafting the form, style and presentation of federal legislation that would be
required to make a text semiotically “language-neutral”.
‘0’ For a subtle rendering of these problems, see N. Kasirer, “Dire ou d~finir le droit?” (1994) 28
RJ.T. 141; N. Kasirer, “What is vie commune? Qu’est-ce que living together?” in Brierley et al., eds.,
supra note 27 [forthcoming]. See also S. Sarcevic, “Bilingual and Multilingual Legal Dictionaries:
New Standards for the Future” (1988) 19 R.G.D. 961.
” The counsel of the so-called mischief rule of Heydon’s Case (1584), 3 Co. Rep. 7a, 76 E.R. 637
(K.B.), thus speaks not only to the task of the judge-interpreter, but also to the task of the drafter-
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
ministerial request, the resulting text may constitute a better reflection of normative
intention and extrapolation in one version rather than the other.” More to the point,
clarity is not fundamentally the criterion of fidelity to intention. A legal rule that
seems clear in its expression in one language and not so straightforward in its expres-
sion in the other may actually be a more faithful rendering of ministerial purpose in its
less clear version. Still again, the meaning of a rule cannot be reduced to its lowest
common denominator of overlapping meaning.” Finally, when a bilingual text is not
created in two original versions, definition, grammar and syntax in both languages
will still exercise significant influence over its import. A close reading of two texts
may point to a norm only imperfectly rendered in each and which, like the elephant
touched by several blindfolded persons, is much greater than the immediate extension
of any of their reports.'”5
35. This suggests a third challenge that inheres in all legislative drafting and interpre-
tation: the language of law is embedded. As noted above, it is embedded first in eve-
ryday usage: no matter how autonomous certain technical legal concepts are from
everyday usage, they cannot be disjoined from it.” More than this, the language of
law is also embedded in legal tradition: in Canada, there are not only two official lan-
the common law and the civil
guages, but also at least two official legal cultures –
law.’ A statute that translates “mortgage” as “hypothque” fails to acknowledge how
much legal language presumes legal culture.” Similarly, a statute entitled the Federal
Real Property Act/Loi sur les immeubles fjddraux”‘ presumes that an “immeuble” in
common-law legal French is equivalent to “real property”, and that “real property” in
civil-law legal English is equivalent to “immeuble”. Both are, obviously, inexact pre-
interpreter. And this is true whether the text is to be produced in several linguistic versions or in only
one. All drafting, like all interpretation, is translation.
‘O Art. 2674, para. 3 C.C.Q. is an example where the translated English text (“proceeds” for
“sommnes d’argent”), though inexact, offers a more coherent normative statement than the original
French text.
” Compare “Litigating the Meaning”, supra note 9 at 329, where Beauprd calls this method the
search for the “highest common meaning”, and advocates its adoption.
’05 Compare the essays from the symposium “L’interpr6tation des lois et des conventions plurilin-
gues” ((1984) 25:4 C. de D.) with C.B. Kuner, “The Interpretation of Multilingual Treaties: Compari-
son of Texts Versus the Presumption of Similar Meaning” (1991) 40 I.C.L.Q. 953.
See J. Darbelnet, “Niveaux et r6alisations du discours juridique” in G6mar, ed., supra note 9, 51.
107 In addition to the common-law and civil-law traditions, there are also numerous aboriginal legal
traditions that are part of “official” Canadian law (see e.g. J. Borrows, “With or Without You: First
Nations Law (in Canada)” (1996) 41 McGiU L.J. 629); and there is also a federal legal tradition that
by its very nature is neither “common law” nor “civil law” (see J.E.C. Brierley, “Quebec’s ‘Common
Laws’ (droits communs): How Many Are There?” in E. Caparros, ed., Miflanges Louis-Philippe Pi-
geon (Montreal: Wilson & Lafleur, 1989) 109; see also Macdonald, “Encoding Canadian Civil Law”,
supra note 27).
0 Minist~e de la justice, Section du Code civil, Le traitement des crianciers dans la loi sur la
faillite et l’insolvabilitd et les mtcanismes de garantie dans le droit civil du Quibec: rapport, problW-
mes et solutions possibles (December 1996) [unpublished].
‘9 S.C. 1991, c. 50.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 42
sumptions.”‘ Here, the aspiration to legal bilingualism highlights the different dimen-
sions of the embedded character of all legal language.”‘
36. Over the past few decades, various tentative responses to the challenges of legal
bilingualism have emerged from official circles in Canada. Some years ago, the Law
Reform Commission of Canada issued a study paper entitled Drafting Laws in
French.”2 Its authors drew inspiration from French theorists who had argued that
“[t]he expression of the law, to be well understood, ought to be written in a language
directly accessible to the mind.””‘ If one pardons the overstatement about how the
mind apprehends language, one can nonetheless discern much wisdom in the affmna-
tion. The vocabulary, grammar and syntax of the law ought to respect, at any given
time and in any given place, the conventions and usages of the natural language in
which the legal norm is being expressed. But the Law Reform Commission appears to
have accepted a further, unstated, assumption that is less defensible.
This assumption is that language and culture are so integrally connected that to
communicate law adequately, only the “indigenous” language of a given legal system
is appropriate. For the Law Reform Commission, “[i]n order for [a] rule to be ac-
cepted, it must be, above all, in keeping with the customs and cultural reflexes of
those persons to whom it is addressed.””‘ Translations are inadequate because the
“sound-associations” within a language cannot usually be translated, though they are
of immense importance in building up interpretive responses. This view of legal
translation presumes that one is creating a second legal order, as much as possible the
mirror of the other, but necessarily separate from it.
The late Jean Kerby, a linguist formerly with the Federal Department of Justice,
held that “la traduction comporte non seulement le passage d’une langue une autre,
mais encore la transposition du message d’un syst~me de droit A un autre.””‘ In this
o See J.E.C. Brierley & N. Kasirer, “Review of Federal Real Property Act/Loi sur les immeubles
fidraux in light of the bijural and bilingual character of federal statutory instruments” (Manuscript
prepared for the Federal Department of Justice, 31 March 1996) [unpublished].
.. The point is explored in E. Smith, “Peut-on faire de la common law en frangais?” (1979) 12:2, 3
R. de l’Universit6 de Moncton 39. For an affirmative answer to this question, see P. Legrand Jr., ed.,
Common Law: d’un sikcle l’autre (Cowansville, Que.: Yvon Blais, 1992); M. Bastarache & D.G.
Reed, “La n6cessit6 d’un vocabulaire frangais pour la Common law” in G6mar, ed., supra note 9, 207.
See also “Legal Bijuralism”, supra note 21. Other aspects of the theme are elaborated in J.-M. Bris-
son, “L’impact du Code civil du Qu6bec sur le droit f&6ral: une probl6matique” (1992) 52 R. du B.
345; J.-M. Brisson & A. Morel, “Droit f16dral et droit civil: compl6mentarit6, dissociation” (1996) 75
Can. Bar Rev. 297.
“2 Law Reform Commission of Canada, Drafting Laws in French (Study Paper) (Ottawa: Minister
of Supply and Services Canada, 1981).
“3 bid., epigraph (trans. quote from Report preceding Decree No. 71-740 of 9 September 1971 in-
stituting new rules of procedure intended to constitute part of a new Code of Civil Procedure
(France), cited in J.-L. Sourioux & P. Lerat, 12 langage du droit (Paris: Presses Universitaires de
France, 1975) at 64).
14 Ibid at 279.
“‘ Kerby, supra note 96 at 13. See also J. Kerby, “La traduction juridique, un cas d’esp~ce” in
G6mar, ed., supra note 9, 3.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
conception of the enterprise, a direct connection between a language and a legal tra-
dition is implied: the common law and the English language are inexorably connected
such that the possibility of “common law in French” is fatally compromised; and the
civil law and the French language are intimately connected, such that the possibility
of “civil law in English” is also fatally compromised.
37. These observations are deeply troubling. While they apparently rest on a sophisti-
cated view of the relation of language to meaning, they “essentialize” both language
and legal culture.”‘ In so doing, they ignore several instances of the successful prac-
tice of legal bilingualism. Nonetheless, this approach is not surprising, given three key
theoretical tenets that have accompanied the movement to official legal bilingualism
at the federal level in Canada over the past few decades: positivism, statism and mo-
nism.
38. In the present context, legal positivism is not meant to signal a particular jurispru-
dential theory about the pedigree of legal norms.”‘ Rather, it points to the widespread
belief that it is desirable to reduce all law to written enactments. Many of those who
preach legal bilingualism today are naively continuing the Napoleonic and
Benthamite codification projects in their unvarnished form.”‘ Of course, most con-
temporary proponents of codification (whether of the private law in a civil code, or of
civil-liberties guaranties in a charter of rights and freedoms) do not pretend for a mo-
ment that legislation either precedes or exhausts legal normativity.”‘ They do, none-
theless, assume that “made law” overtakes the legal norm that it purports to express.
When custom is thus made, a form of legal transubstantiation occurs; thereafter, law
is text. Because constitutionalized legal bilingualism requires legislative instruments
The literature on such themes tends to emanate from writers who perceive a legal tradition under
threat, such as civilians in Scotland, Louisiana or Quebec, although it is also present in dominant cul-
tures. See e.g. C.R. McDiarmid, Scots Law: How Can and Why Should It Survive? (LL.M. Thesis,
Faculty of Law, McGill University, 1995) [unpublished]; A.N. Yiannopoulos, “Louisiana Civil Law:
A Lost Cause?” (1980) 54 Tul. L. Rev. 830; D. Lemieux, “The Quebec Civil Law System in a Com-
mon Law World: The Seven Crises” [1989] Juridical Rev. 16. For a general discussion, see Brierley
& Macdonald, eds., supra note 44 at 67-73 and sources cited. But it also has a long pedigree in cases
of “threatened” legal languages. See e.g. M. Sparer & W. Schwab, Ridaction des lois: rendez-vous du
droit et de la culture (Quebec: Editeur officiel du Qu6bec, 1980); P Issalys, Langage et syst~me des
lois (Quebec: Fditeur officiel du Qu6bec, 1980).
“‘ For a contemporary account of the necessary elements of such a theory, see J. Raz, The Authority
of Law: Essays on Law and Morality (Oxford: Clarendon Press, 1979).
“‘ Compare J.B. Proudhon, TraitJ des droits d’usufruit, d’usage personnel, et d’habitation, 2d ed.
(Dijon: V. Lagier, 1836) and J. Bentham, An Introduction to the Principles of Morals and Legislation
(London: Athlone, 1970).
“+ See P.-A. Cr6peau et aL, eds., Codification: Valeurs et langage (Quebec: Service des communi-
cations du Conseil de Ia langue franalse, 1985); on charters of rights, see R.A. Macdonald, “The
New Zealand Bill of Rights Act: How Far Does it or Should it Stretch?” in The Law and Politics:
1993 New Zealand Law Conference – Conference Papers, vol. I (Wellington: N.Z. Law Society,
1993) 94 at 113-41.
152
MCGILL LAW JOURNAL/REVUE DE DROIT DE MCGILL
[Vol. 42
to be enacted in both French and English, the legal order itself will be more bilingual
the more it is written.'”
No doubt, the theory of legal bilingualism requires that all legislative texts be
rendered and that verbal communication in law flourish in two languages: that is, in
Canada, in both French and English.’2’ But what of non-literary and non-verbal nor-
mative symbolisms –
customs and those practices, understandings and symbolic
constructions that do not depend on preexisting legal language? Since they are, in the
crudest sense, not directly translatable, proponents of legal bilingualism feel impelled
to discount or minimize them as legal artifacts. Demonstrating a commitment to legal
bilingualism and to promoting its official practice therefore tends to privilege a theory
of law that equates legal normativity with its expression in language.
39. A second assumption shared by many proselytizers for legal bilingualism is that
legal normativity can be equated with the activity of the state. On this view, all law is
either made, or referentially incorporated, by the political state. The rationale for con-
flating law and state is that if the enacted product of official legislatures –
the federal
Parliament or provincial Legislatures – can be rendered in both languages, then one
of the basic requirements of legal bilingualism has been achieved. Ensuring that other
state institutions –
operate bi-
lingually completes the project.” This dominant conception of law is increasingly
being contested. Legal normativity may be understood as grounded in several sites
besides the political state.'” Within the geography of a single state, many unofficial
legal orders compete for the fidelity and attention of citizens. Each of these – corpo-
rations, unions, universities, clubs, and so on – produces its own legislative and ju-
dicial texts.’24
courts, administrative tribunals, the public service –
If legal bilingualism is about textuality, adopting a non-statist view of law might
seem to require that each of these so-called peripheral legal orders also operate bilin-
20 For a critique, see R.A. Macdonald, “Understanding Regulation by Regulations” in I. Bernier &
A. Lajoie, eds., Regulations, Crown Corporations and Administrative Tribunals (Toronto: University
of Toronto Press, 1985) 81.
“2, See Constitution Act, 1867, supra note 23, s. 133; Charter, supra note 25, ss. 16, 20; Official
Languages Act, R.S.C. 1985, c. 0-2, s. 1. See generally the two volumes edited by Bastarache, supra
note 22.
, For an inventory of the obligations of various governments in Canada, see Hogg, supra note 26
at 1197-1226; Tremblay, supra note 26.
‘
‘” See generally J. Griffiths, “What is Legal Pluralism?” (1986) 24 J. of Legal Pluralism I; B. de
Sousa Santos, Towards a New Common Sense: Law, Science and Politics in the Paradigmatic Transi-
tion (New York: Routledge, 1995).
24 For a lengthy excursus in the Canadian context, see R.A. Macdonald, “Recognizing and Legiti-
mating Aboriginal Justice: Implications for a Reconstruction of Non-Aboriginal Legal Systems in
Canada” in Royal Commission on Aboriginal Peoples, Aboriginal Peoples and the Justice System:
Report of the National Round Table on Aboriginal Justice Issues (Ottawa: Minister of Supply and
Services Canada, 1993) 232.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
153
gually.'” For this reason, the claim that law arises only in state institutions permits
proponents of legal bilingualism to circumscribe the scope of necessary translation.
Dispute resolution institutions other than courts – not being official institutions of the
state – need not operate in both English and French.”” More than this, these propo-
nents seek to draw the sources of law narrowly. Textbooks, treatises, monographs and
other commentary – not being official products of the state legal order – need not be
produced in two languages. Adopting a statist view of law permits those who are
committed to legal bilingualism to claim success as soon as the artifacts of state insti-
tutions are rendered bilingually.
40. Those who promote legal bilingualism typically also assume that there is within
the state legal order only one normative system. They cleave to a monistic view of
law. The official legal order comprises a single system of norms that can be arranged
in hierarchical fashion such that conflicts between them are capable of a “correct”
resolution. On such a view, legal bilingualism is simply the double representation of a
single legal reality: whether apprehended in English or in French, there is only one
legal norm – a legal norm that is totally confounded with either of the texts of its ex-
pression.’ 7 This formal conception of law focuses on the identification of the legal
norm.
The possibility that interpretation is an act of constructing meaning puts assump-
tions about the transparency of words into question. The goal of preventing the two
language versions of a norm from being ranked lexically – of asserting the equality
of the two texts –
leads proponents of legal bilingualism to claim the discreteness of
each version. To recognize that the law does not operate as a single normative system
and that any particular legal rule may have a different outcome in different situations
seems to undermine the project of legal bilingualism; it suggests that what really
matters is individuality and not language affiliation.’28 In this sense, the challenge of
legal pluralism is that it seems to dismiss the mediating role of English and French as
official modes of apprehending law in Canada. Whatever the language of a text, a
person may translate it into a personal language which may or may not be one of the
official natural languages of the legal bilingualism project.
41. These three assumptions and commitments induce those who argue for legal bi-
lingualism to advance a formal conception of law: law is reduced to its discursive ex-
“‘ For an explanation of alternate strategies to avoid this result, see R.A. Macdonald, “Critical Legal
Pluralism as a Construction of Normativity and the Emergence of Law” in A. Lajoie, ed., Thdories et
dmergence du droit (Montral: Th6mis, 1997) [forthcoming].
‘2 On the implications of deregulation, privatization and alternative dispute resolution for the re-
configuration of state normativity under a non-pluralist regime, see R.A. Macdonald, “Prospects for
Civil Justice” in Ontario Law Reform Commission, Study Paper on Prospects for Civil Justice
(Toronto: Ontario Law Reform Commission, 1995) (Chair: J.D. McCamus).
‘.. The impossibility of conflating text and norm is often claimed by postmodernists to be their key
insight: see Douzinas & Warrington, supra note 18. Compare Fuller, supra note 53 for a similar af-
firmation a generation earlier.
‘ See J. Boyle, “Is Subjectivity Possible? The Post Modem Subject in Legal Theory” (1991) 62 U.
Colo. L. Rev. 489.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 42
pression in language. The thirty-year official-languages saga in Quebec – Bill 63,
Bill 22, Bill 101, Bill 178, and Bill 86″0-and its obverse, the tribulations in the prai-
rie provinces over the language of education, ‘” are classic manifestations of the misdi-
rection of energy by many supporters of legal bilingualism. Their case for legal bilin-
gualism is grounded in the principle that the law should be made manifest in the lan-
guage of the citizen to whom it is addressed.
But what difference does it make to the average citizen that official legal rules are
available in two languages? Few citizens ever read legislative instruments. Fewer still
pore over reports of judicial decisions or doctrinal writings. One suspects that institu-
tional bilingualism is conceived and understood as a professional exercise: it requires
the creation of more official written law for the benefit of practitioners of official
written law. Far from producing proof of how law transcends language, the develop-
ment of legal bilingualism in Canada has contributed to a renaissance of legal formal-
ism. In other words, the theory of legal bilingualism one adopts directly influences
what one is prepared to consider as law.'”‘ Not surprisingly, the mantra of official legal
bilingualism in Canada is now: if it cannot be made bilingual, it cannot be law.
42. This renewed formalism consequent upon the attempt to achieve a bilingual legal
order has the potential to exacerbate two other pathologies of modem law. There is a
strong possibility, first, that current understandings of legal bilingualism will produce
defacto legal dualism (the practice of two legal unilingualisms), and second, that le-
gal texts will become even more bureaucratic than they have been previously. Evi-
dence in support of the former proposition can be found in the history of orthodox
civil-law practice in Quebec. Legal practice in post-confederation Quebec has always
been carried on in English and French and has always been centred upon at least two
sets of written materials. Except at the highest levels of legal endeavour, until recently
these practices remained largely insulated from each other.”‘ The English-speaking
bar has developed specific case-law dossiers quite distinct from the francophone bar.
9 Bill 63, An Act to Promote the French Language in Quebec, 4th Sess., 28th Leg., Quebec, 1969
(assented to 28 November 1969, S.Q. 1969, c. 9); Bill 22, Official Language Act, 2d Sess., 30th Leg.,
Quebec, 1974 (assented to 31 July 1974, S.Q. 1974, c. 6); Bill 101, Charter of the French Language,
2d Sess., 31st Leg., Quebec, 1977 (assented to 26 August 1977, S.Q. 1977, c. 5); Bill 178, An Act to
amend the Charter of the French Language, 2d Sess., 33d Leg., Quebec, 1988 (assented to 22 De-
cember 1988, S.Q. 1988, c. 54); and Bill 86, An Act to Amend the Charter of the French Language,
2d Sess., 34th Leg., Quebec, 1993 (assented to 18 June 1993, S.Q. 1993, c. 40). See generally J.E.
Magnet, Official Languages of Canada (Cowansville, Que.: Yvon Blais, 1995), especially at 35-65.
“‘ For the relevant decisions of the Supreme Court on language-of-education questions, see Hogg,
supra note 26 at 1219-38; see also P. Foucher, “Les droits scolaires des minorits linguistiques” in
Beaudoin & Mendes, eds., supra note 26, 16-1.
” The causal connection is, obviously, not unidirectional. Commitments about what counts as law
derive from many sources, and not just a concern about bilingualism. Nonetheless, within a broadly
statist and instrumentalist conception of law, a concern for bilingualism reinforces the tendencies
noted.
“‘ For a discussion of this point, albeit mainly about the ignorance of English-speaking common
lawyers, but by implication also English-speaking civil lawyers, see J. Deschanes, The Sword and the
Scales (Toronto: Butterworths, 1979) c. 3 (“On Legal Separatism in Canada”).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
Even the accepted interpretations of statutes, codal articles and cases have frequently
been dual.’3 Secondary-source materials (texts, law-review articles, standard-form
contracts, precedents) have tended to be deployed by those who practice law primar-
ily in the language in which these doctrinal sources were composed.’3 ‘
Under the Civil Code of Lower Canada, the relative frequency of deployment of
particular legal institutions, such as the fiduciary substitution by contrast with the
trust, showed a divergence grounded in linguistic and cultural dualism.’3 Similarly,
recourse to, and acceptance of, different legal institutions –
community of property
with a contractual appointment of an heir by contrast with separation of property with
inter vivos matrimonial gift –
reflected differing perceptions of the role of courts or
public officials such as notaries.’36 Judges themselves revealed methodological diver-
gences that can be traced, at least in part, to questions of language and culture.’37
43. These are only examples of the scope and character of legal dualism. The larger
questions are: What is the basis for claiming that legal bilingualism has the potential
to produce de facto legal dualism? And what is wrong with legal dualism? The latter
‘. See e.g. the differing interpretations given to Matamajaw Salmon Club v. Duchaine, [1921] 2
A.C. 426 (P.C.) (contractual dismemberment of ownership/theory of estates) and Banque canadienne
nationale v. Lefaivre (1950), (1951] Que. K.B. 83, 32 C.B.R. I (ownership sui generis/mortgage-type
transaction) and to art. 127 C.C.L.C. (obligatory impediment to marriage/impediment only binding on
the duty of the official solemnizing the marriage) and art. 2022 C.C.L.C. (absolute prohibition of dis-
guised non-possessory security over moveables/prohibition not reaching double-sale transactions).
” An unscientific sampling in 1981 of memos and briefs prepared on three sets of topics – wills,
contract remedies and latent defects -by
two Montreal law firms (one largely English-speaking, one
largely francophone) revealed that less than 50 percent of the authorities referred to (cases, doctrine)
were common and that each tended to cite material in the predominant language of the firm in a ratio
of 3:1.
“‘ Much of this is a direct result of how law is taught, and with the relative ease with which the in-
stitution finds a reflection in the non-legal literature, in cultural mythology and in religious dogma.
Compare, on this point, different treatments of family-property arrangements in Ringuet, Trente ar-
pents (Paris: Flammarion, 1938), H. MacLennan, Two Solitudes (Toronto: Collins, 1945) and M.
Richler, St. Urbain ‘s Horseman (New York: Knopf, 1971).
6 Compare M. Ferron & R. Cliche, Quand le peuplefait la loi: la loi populaire a Saint-Joseph de
Beauce (Montreal: Hurtubise, 1972); C.-E. Gagnon, Mefmoires d’un notaire de campagne (Quebec:
Septentrion, 1990); and E. Kimmel, “The Notarial System and its Impact in Canadian Law” in R.
Landry & E. Caparros, eds., Essays on the Civil Codes of Qudbec and St. Lucia (Ottawa: University
of Ottawa Press, 1984) 109, for different perspectives on the notarial profession reflecting cultural-
linguistic dichotomies.
“” These divergences frequently manifest themselves in debates about preserving the “purity” of
civil-law institutions from the corrupting influence of the common law, although it has never been the
case that such a polemic was produced by an English-speaking judge. For comparative interpretive
perspectives, see S. Normand, “Un th~me dominant de la pens~e juridique traditionnelle au Qu6bec:
La sauvegarde de l’int~grit6 du droit civil” (1987) 32 McGill L.J. 559; J.-G. Castel, “Le juge Mi-
gnault d6fenseur de l’intdgdt6 du droit civil Qudbdcois” (1975) 53 Can. Bar Rev. 544; D. Howes,
“Nomadic Jurisprudence: Changing Conceptions of the ‘Sources of Law’ in Quebec from Codifica-
tion to the Present’ in Glenn, ed., supra note 21, 1. See, on the linguistic point, J. Deschenes, supra
note 130 and the comments of A. Monet J.A. in Central Factors v. Imasa Ltd (12 March 1979),
Montreal, 09-000-245-779, J.E. 79-318 (C.A.).
MCGILL LAW JOURNAL! REVUE DE DROITDE MCGILL
[Vol. 42
is more easily answered. From a practical point of view, the problem of legal dualism
is that it enhances justificatory confusion by a continual process of self-referential ci-
tation. This self-reference necessarily truncates the search for meaning. Many of the
possible interpretations of legal rules are lost, or at least are not consciously present,
in the mind of the interpreter who looks to the rules in only one language. Rather than
seizing upon a second language in order to enrich legal understanding and to gain
more sophistication in legal interpretation, the jurist who looks to only one text ac-
cepts incomplete normative descriptions and relies on jejune interpretive methods.”
Numerous factors contribute to the apparently inexorable decay of legal bilingual-
ism into legal dualism: intellectual laziness among legal professionals; rampant uni-
lingualism among legal elites; a proliferation of mediocre translations of texts; an
educational system that privileges information over understanding; and, not least, a
plethora of secondary sources and computerized finding tools.'”” Given this, and ac-
cepting the ambition of legal bilingualism as a primary goal, one might well conclude
that comprehensive translation of legal materials is counterproductive. Legal re-
sources in two-language versions have the perverse consequence of limiting the
catholicity of a lawyer’s search for meaning. The more materials are available in
one’s own language, the less one tends to look outside that language for ideas and in-
spiration, and the less one tends to look outside the language for criticism and inter-
pretation of legal rules.”‘ The more legal materials are available in both languages, the
less interpreters will have to confront the assumptions, nuances and metaphors of
their own language.”
44. This last observation points to the second concern, noted earlier, about the new
formalism consequent upon the quest for legal bilingualism. The conception and
drafting of bilingual texts risk becoming even more bureaucratic. Bureaucratic texts
are those that are not necessarily the work of one person and that are highly formu-
“‘This theme is elegantly explored in N. Kasirer, “Dire ou d6finir le droit”, supra note 101.
“‘I have attempted to illustrate how similar factors have begun to transform legal practice in R.A.
Macdonald, “Images du notariat et imagination du notaire” [1994] Cours du Perfectionnement du
Notariat 1.
” Here one returns to the theory of bilingual education. See K. Hakuta, Mirror of Language (New
York: Basic Books, 1986) c. 9. In law, some of the best theoretical writing on this point is by David
Howes (see D. Howes, “From Polyjurality to Monojurality: The Transformation of Quebec Law,
1875-1929” (1987) 32 McGill L.J. 523; D. Howes, “La domestication de la pens6e juridique qu6b6-
coise” (1989) 13:1 Anthropologie et Socidt6s 103; D. Howes, “Dialogical Jurisprudence” in W.W.
Pue & B. Wright, eds., Canadian Perspectives on Law and Society: Issues in Legal History (Ottawa:
Carleton University Press, 1988) 71; Howes, supra note 135; and D. Howes, “Faultless Reasoning:
Reconstructing the Foundations of Civil Responsibility in Quebec Since Codification” (1991) 14
Dalhousie L.J 90.
‘z, See Brierley & Macdonald, eds., supra note 44 at 147-49 for discussion in the context of a civil
code. Compare Beaupr6, Interpreting Bilingual Legislation, supra note 9. Of course, the whole exer-
cise of comparative law is predicated on the assumption that “exteriority” enriches understanding. See
e.g. P. Legrand Jr., “Comparative Legal Studies and Commitment to Theory” (Book Review of A
Modem Approach to Comparative Law by P. de Cruz) (1995) 58 Modem L. Rev. 262.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
laic.”‘ One already sees, in cases where enactments or judgments are not initially con-
ceived in two originals, legislative drafters or judges writing for a translator. In the
latter situation, sometimes the author does not even verify the translated version of his
or her judgment.” The more a text aspires to formulaic rationality for ease of transla-
tion, the more difficult becomes the task of ascribing meaning. In systems that privi-
lege such forms of communication, an interpreter is less concerned with the meaning
of a text than with its practical application. It follows that in a jurisdiction with two
unilingual sectors of the bar, the pressure will increase to produce materials whose
meaning is intended to be self-evident.
The devaluation of legal meaning reflected by a devaluation of legal symbolism
parallels a similar reconfiguration of society’s most pervasive communicative sym-
bols. Harlequin romances, verse masquerading as poetry in greeting cards, Muzak,
“art” sold in shopping-centre parking lots, plastic renditions of David, television soap
operas, sports-stadium rituals and Reader’s Digest are only some examples of the
commodification of form. Ultimately each of these derivatives takes on an iconic and
original life of its own – becomes its own original. This is the character of popular
culture, a culture that feeds into the very communicative symbolisms against which it
is constructed.'” While “low culture” beneficially calls “high culture” to account, in
law, the syntactical form of “popular culture” is usually misappropriated into a con-
text where professional practice purges it of its substance. This misappropriation
grounds the contemporary crises of law: the curse of masses of indecipherable legis-
lation, volumes of detailed statutory instruments, bureaucratic judgments, and head-
notes routinely cited as expositions of the law, even by judges themselves.
IV. Practicing Legal Bilingualism in Canada
45. The two foregoing consequences of the official theory of legal bilingualism in
Canada today – dualism and bureaucratic writing – might lead one to conclude that
the project is inherently flawed. Yet neither is a necessary concomitant of the endeav-
our. Each is only a consequence of particular semantic and epistemological presup-
positions bound up in the official theory. It is possible to conceive of a legal theory
and a legal practice that accommodate and welcome the possibility of a bilingual legal
order rather than dual unilingual legal orders, and that contest and resist the contem-
porary drift towards bureaucratic discursivity in legal texts. In short, if the objective is
‘ See Vining, supra note 64, part I; and I Vining, “Justice and the Bureaucratization of Appellate
Courts” (1982) 2 Windsor YB. Access Just. 3.
‘”The point is obvious in cases where judges are unilingual. Of course, much effort now goes into
ensuring high-quality translation of Federal Court and Supreme Court judgments in Canada. For a
striking example of the pitfalls of not attending to translation, see Tamarak Construction Inc. v.
United Services Club Ltdi, [1971] C.A. 334, where “change of heart” in a judgment was translated in
a headnote as “transplantation de cmur” (ibid. at 334).
‘See Law and Popular Culture (special issue) (1995) 10 Can. J. L. & Soc’y. See especially A.
Hunt, “The Role of Law in the Civilizing Process and the Reform of Popular Culture”, ibid., 5.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 42
to generate the preconditions for and the practices necessary to sustain legal bilingual-
ism, the primary task lies in translating theory into action.
The antidote to legal dualism cannot lie in endless exercises of bureaucratic nor-
malization of legal terminology (at its most asinine, ought one to render “the common
law” by “le droit commun”, “le common law” or “la common law”?). 5 Terminologi-
cal normalization may well facilitate the production of texts in both French and Eng-
lish,'” but it is no guarantee that the equivalent expressions will necessarily compel
engagement with each other. An increased recourse to legislated legal norms is also
no assurance that law will be read bilingually. Finally, the practice of translating all
forms of legal texts from French into English and from English into French offers no
certainty that both will actually nourish interpretation. All the above, paradoxically,
facilitate and encourage legal dualism.
The antidote to legal dualism does, however, lie in the promotion of legal bilin-
gualism, where language serves as a point of access for legal knowledge rather than
as an exhaustive expression of that knowledge. In a bilingual legal order there will be
little place for “official” translation understood as the discursive rendering of legal
texts generated in one natural language into legal texts expressed in another.”‘ In such
a legal order the specific practices of legal bilingualism will vary according to the dif-
ferent modes of expression –
through
which the legal norm in issue may be apprehended.
as enactment, as ratio decidendi, as custom –
46. In imagining the practices required by legal bilingualism, one may begin by con-
sidering the legislative text. Most twentieth-century enactments are attempts to render
inferential legal norms into explicit language or to induce new inferential legal norms
by providing them with a canonical textual referent.’ ‘ These explicit norms can never
be dissociated from their implicit context, and their formulaic (instrumental) meaning
will always be nourished by inferential (symbolic) understanding.’ ‘ Nevertheless, the
creation of a textual deposit of law is a central feature of the legal enterprise, espe-
cially in contemporary, democratic, pluralistic societies.'”‘
“‘ On this parenthetical point see P Legrand Jr., “Le droit de Blackstone et la langue de Mollre:
une valse A mille temps” in P Legrand Jr., ed., supra note 111, vii (Preface).
‘,* See Ministry of the Attorney General for Ontario, Petit guide de tenninologie juridique / Hand-
book of Legal Terminology (Toronto: Ministry of the Attorney General, 1981), and sources cited for
examples of the exercise.
” This is not to say that “official” translations will not be necessary. Not all legal material will be
created simultaneously in both English and French; not all interpreters will interpret legal material si-
multaneously in both English and French. The point, to be explored below, is that these “official”
translations must be understood as “translations”. See Covacs, supra note 96.
,” It was not always so. See Manderson, supra note 43. On the contemporary theory of legislation,
see C. Allen, Law in the Making, 7th ed. (Oxford: Oxford University Press, 1964) at 426-69; 1.
Ghestin & G. Goubeaux, Traitj de droit civil: introduction gingrale, 3d ed. (Pads: L.G.D.J., 1990) at
225-302.
” See Macdonald, supra note 42.
“See L.L. Fuller, Anatomy of the Law (Westport, Conn.: Greenwood Press, 1968) at 57-69, on
“implicit elements in made law”.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
159
There are, of course, many forms of enactment. Legislation is at its most discur-
sive in statutes, regulations, orders-in-council and bylaws. It is less so in most consti-
tutions, codes, policy directives and like instruments.”‘ Even in its most discursive
expressions, however, the aspiration to achieve a one-to-one correspondence between
texts in two natural languages is a false hope. In both the internal logic of legislative
language and its external presentation on a page or computer screen, normative diver-
gence is betrayed.’ The occasions for this divergence and the accommodations that
the structural properties of language impose upon those who would seek legal bilin-
gualism through legislation bear notice.
47. Any attempt to achieve a bilingual statute-book through the translation of legisla-
tion initially drafted in one language into another cannot fully succeed. The inevitable
limits of discursivity are such that translators will be compelled to sacrifice meaning
for textual exactitude, and this sometimes even at the expense of clarity.”‘ These same
limits also compromise the quest for a strict, discrete dualism in texts, even where two
originals are initially produced. Distinct originals are, in other words, the precondition
for legal bilingualism. Bilingual statutes will then be the result of integrating two
separate texts initially crafted and drafted in a manner sensitive to the contexts and
subtleties particular to each language.’5’ This dual craftsmanship would inspire not
only the vocabulary but also the syntax and grammar of legislation. It would also
shape the formal structure and organization of the legislative text. Neither version
would emerge as a “translation” or “rendition” of the other.
The formal as well as substantive consequences of dual drafting and of a com-
mitment to the equality of texts are profound. Some can be immediately perceived. It
could well be that in one language version what is stated as an exception to a principle
“‘. See on these points Macdonald, supra note 120 at 121-25.
,’ For example: texts take up more space in one language than another, subjects in one text become
objects in the other, contingent clauses may precede the noun in one language, and may follow it in
the other. See generally M. Sparer, “R&laction des lois, langage et valeurs: les enjeux” in G6mar, ed.,
supra note 9, 111.
“.. One of the most striking illustrations comes with the Supreme Court’s initial interpretation of s.
28 of the Federal Court Act, R.S.C. 1970 (2d Supp.), c. 10, where “other than a decision or order of
an administrative nature not required by law to be made on a judicial or quasi-judicial basis” is trans-
lated by: “autre qu’une d6cision ou ordonnance de nature administrative qui n’est pas 16galement
soumise It un processus judiciaire ou quasi judiciaire.” The French version implicitly excludes a
common-law requirement to act judicially of the type found in Ridge v. Baldwin (1962), [1963] 1
Q.B. 539, [1962] 1 All E.R. 834. For a consequence of this difference, see the judgments of Pigeon
and Dickson JJ. in Howarth v. National Parole Board, [1976] 1 S.C.R. 453, 50 D.L.R. (2d) 349
[hereinafter cited to S.C.R.]. Dickson J., presumably working from the English version of the legisla-
tion, found the administrative body in question to be subject to a requirement to act on a judicial basis
and therefore subject to the review of the Federal Court of Appeal under s . 28 (ibid. at 459-62). Pi-
geon J., presumably working from the French version, read s. 28 more narrowly, so as to exclude
administrative decisions which are “aucunement [des] ddcision[s] judiciaires” because no legislative
text imposed a duty to act judicially from the review of the Court of Appeal (ibid. at 472).
” See Levert, supra note 95; Morel, supra note 95; Canada, Ministry of Justice, Report of the Leg-
islative Byuralism Committee / Rapport du Comitj sur le Bijuridisme Ligislatif (Study prepared by
the Legislative Services Branch, 4 April 1996) [unpublished].
McGILL LAW JOURNAL/REVUE DE DROIT DE MCGILL
[Vol. 42
might, in the other, be stated as the principle itself. Since the legal norm is not the text,
there would be no necessary reason for statutes to have the same number of sections,
sub-sections, or paragraphs in both language versions.'” There would be no reason
why the typical English-language definitional enumeration of legal concepts (real
property, personal property, incorporeal hereditaments, chattels real, rights of rever-
sion) could not be rendered in the French-language version by a simple general ex-
pression (tous les biens).'”
Bilingualism ought also to be reflected in the material presentation of legislation.
One might ask why the alpha-numerical presentation of, say, the Revised Statutes of
Canada, should not be different as between English and French versions: for exam-
ple, if the Bankruptcy and Insolvency Act is listed as chapter B-3, why should the Loi
sur la faillite et insolvabilitd not be listed as F-something?” Again, one might ask
why one version should always be on the right-hand (or left-hand) side of the page.
Could not two volumes be produced so that in the French volume the French text
would be in the right-hand column and follow the French alpha-numeric order, and in
the English volume the English text would be in the right-hand column and follow the
English alpha-numeric order? In other words, an acknowledgment of the semiotic
importance of all elements of substantive and material presentation'” requires a re-
thinking of more than the language of legislation.”9
48. Of course, legal texts have an instrumental purpose: they are meant to be applied.
Application presupposes interpretation and interpretation presupposes translation.
Hence, a rethinking of the manner of drafting and presenting legislation presumes that
the underlying theory of interpretation of bilingual texts would also be recast.'”
Authoritative interpretation of legislative texts would reconstruct the expansive proc-
ess of their drafting. Just as the drafters of bilingual legislation are engaged in the
” One already sees evidence of this in the non-parallel enumeration of definitions in federal stat-
utes and in the non-parallel expression of paragraphs and sub-paragraphs (see e.g. Crown Liability
and Proceedings Act, R.S.C. 1985, c. C-50, s. 2, as am. by S.C. 1990, c. 8, s. 22; S.C. 1993, c. 28, s.
78 (Sch. III, item 38)).
‘Compare
the two versions of s. 4 of the Federal Real Property Act, S.C. 1991, c. 50, both of
which adopt a generic mode of expression.
… One might overcome objections related to citation confusion by listing English versions in Ro-
man characters and the French-language version in Arabic numbers. Thus, Bankruptcy and Insol-
vency Act R.S.C. 1985, c. B-Ill; Loi sur lafaillite et insolvabilitd S.R.C. 1985, c. F-6 (hypothetically).
‘ The semiotics of the statute book have achieved a signal place in Quebec. In one popular edition
of the Civil Code of Lower Canada (J.-L. Baudouin & Y. Renaud, eds., Codes civils (Bas Canada et
Quebec) – Civil Codes (Lower Canada and Quebec) (Montreal: Wilson & Lafleur, 1991)), the Eng-
lish text of the Code is printed in a typeface demonstrably smaller than the French text. See A. Duval,
“Bilingue A 25%” (Book Review of Baudouin & Renaud, eds., ibid.) (1991) 94 R. du N. 260. See also
“Semiotic Musings”, supra note 67.
‘9 There is a further complication at the federal level in Canada, given that federal legislation must
speak also to two legal traditions: civil law and common law. On these questions, see sources cited
supra note 154.
60 This does not imply that prudential rules of interpretation for dealing with bilingual texts should
be completely abandoned. See generally Beaupr6, Interpreting Bilingual Legislation, supra note 9,
for a listing of analogues, in bilingual context, to ordinary canons of statutory interpretation.
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
translation of a single juridical idea into two natural languages, interpreters would
come to accept that knowledge of one version alone is an insufficient point of refer-
ence for understanding the juridical idea in question. They would understand legisla-
tive texts as fully embracing both English and French connotations and contexts, and
as necessarily meaning what both versions say. No longer would it be possible to
speak of two texts being equally authoritative.’ To the extent that any formulation of
a legal rule can be authoritative, it will be necessary to speak of one authoritative bi-
lingual text in French and English.
For this reason, formal incommensurability,’
or even substantive inconsistency,'”
of legislative texts would pose no special interpretive problems. Interpreters would
treat any apparent incommensurability of material presentation and substantive in-
consistency between English and French versions of a statute no differently than they
would treat any apparent commensurabilities and consistencies between them. In both
situations, interpreters would construct, to the best of their ability, the legal norm im-
manent in the language of the two texts. The act of interpretation will always be an
attempt to apprehend the legal norm of which the text is a contingent formulation.
This point might be misunderstood as a claim that meaning is radically constructed
and that the indeterminacy of language fatally compromises the very project of law.'”
It means no such thing. Once the requirements of legal bilingualism are understood,
the determinacy or indeterminacy of language becomes an irrelevant distraction. The
language of legislation is a point of access for apprehending, constructing and trans-
lating a legal norm; it is never the norm itself.
49. Legal texts that are explicit but more presentational, such as judicial decisions and
other written justifications for official action, raise different problems. For here, it is
generally acknowledged by all manner of legal theorists that a written opinion ex-
pressing a legal norm is not the norm itself.'” Even where the judgment purports to
justify a particular interpretation of a norm, the apprehension of which a legislature
has taken the trouble to direct through the language of an enactment, it is still less
susceptible than the statute or regulation to statement in a discursive, canonical
… Compare the present formulation of s. 18(1) of the Charter, supra note 25.
.. By “formal incommensurability” is meant: section numbers need not be the same; the total num-
ber of sections might differ;, a rule may be cast as a principle in one version and as an exception in the
other;, and so on.
” By “substantive inconsistency” is meant: words and phrases that have the same normative func-
tion in a text but that are manifestly of different import (e.g., “sommes d’argent I proceeds” in art.
2674, para. 3 C.C.Q.); texts that are mistranslations (e.g., “mortgage I hypotlhque” in s. 2 of the
Bankruptcy and Insolvency Act, R.S.C. 1985, c. B-3); and texts with variable linguistic parallels (e.g.,
“enterprise I entreprise” in art. 1525, para. 3 C.C.Q. and “business purposes / entreprise” in art. 1842,
para. 3 C.C.Q.).
‘ Recall the quotation from Peller above, accompanying note 16.
‘6 For a compelling exposition, see E Schauer, Playing By the Rules: A Philosophical Examination
of Rule-Based Decision-Making in Law and in Life (Oxford: Clarendon Press, 1991); Sparer &
Schwab, supra note 116.
MCGILL LAW JOURNAL/REVUE DEDROITDE MCGILL
[Vol. 42
form.” A fortiori, where a judgment seeks to justify a particular interpretation of a
norm, the apprehension of which has remained a matter of successive judicial refor-
mulation, its text cannot be dissociated from its implicit context.67 A judicial opinion
is an exercise in persuasion: it is a subtle interweaving of a statement of a legal norm
and the justification for both the content of its normative pith and the form in which it
is stated.'” Judges must be free to use rhetorical techniques that are central to the per-
suasive force of a text but are not readily translated into discursive form. In addition
to their role in convincing the parties, their counsel and the public about the appro-
priateness of the norm being applied (the stare decisis function ofjudgments), judicial
opinions are also aimed at persuading them of the correctness of the decision reached
(the res judicata function). Modes and means of persuasion, such as explicit argu-
ment, rhetoric, metaphor and syntax, are in large measure language-specific.'”
As a consequence, the ambition of legal bilingualism should not become a pretext
for having judges write more bureaucratically simply to facilitate translation. The re-
cent tendency of the Supreme Court of Canada towards model judgments” betrays a
profound misunderstanding of the opinion-writing enterprise.” More than this, the
practice of employing “judgment writers”, adopted by some Canadian courts, is a
tragic distortion of the justificatory process. Given the function of a judicial decision
(whether in the interpretation of legislation or previous judicial decisions, or in the
apprehension and expression of custom), the case for bureaucratic translation of judi-
cial decisions is not at all compelling. For unlike the case of legislation, here there is
not even the pretence that a canonical formulation of a legal norm will serve authori-
tatively to control its mode of apprehension in the future.'”
50. If it is true that judgments are fundamentally only invitations to normative dis-
covery and reconstruction, then the case for producing judicial opinions in two lan-
guages is diminished. However much appellate judgments are negotiated between
judges, in the common-law tradition they are, nonetheless, not the product of a
pression: e.g., the rule in Hicklin v. R. (1868), L.R. 3 Q.B. 360, 11 Cox C.C. 19.
’66 This is not to deny that some judgments interpreting statutes have achieved iconic canonical ex-
6 See e.g. the rule in Shelley’s Case (1581), 1 Co. Rep. 93b, 76 E.R. 206 (K.B.) or the rule in Tulk
v. Moxhay (1848), 2 Ph. 774, 41 E.R. 1143 (Cli.) or the rule in Donoghue v. Stevenson, [1932] A.C.
562, All E.R. Rep. 1 (H.L.).
” The point is nicely developed in Fuller, supra note 150 at 84-112, especially at 90-94.
‘ See J.B. Wfite, Heracles Bow: Essays on the Rhetoric and Poetics of the Law (Madison, Wis.:
University of Wisconsin Press, 1985).
“o Quite apart from this normalization of form, there is an increasing tendency to normalize style,
See J.O. Wilson, A Book for Judges (Ottawa: Ministry of Supply and Services Canada, 1980) at 79-
92. See also L. Maihot, Ecrire la dicision: guide pratique de ridaction judiciaire (Cowansville, Que.:
Yvon Blais, 1996).
See “Style, Meaning and Knowing”, supra note 87.
The point is simply that the so-called ratio decidendi of any case is never fixed and is always
being reimagined every time the case is cited in a subsequent case. See generally, on divergent ap-
proaches to extracting the holding of a case, R. Cross, Precedent in English Law, 4th ed. (Oxford:
Clarendon Press, 1991).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
committee; they are, in the final analysis, the literary product of a single mind.’ No
doubt, it might be imagined that judges could employ both English- and French-
judgment drafters, directing each upon conclusion of a case to produce a draft
judgment. These two texts might then be compared and a “common” version pro-
duced. Yet here, exactly the same types of arguments about textual fidelity as would
apply to legislation are even more apposite. By contrast, if the two versions of a
judgment must both be original creations of their author, their production will nec-
essarily be serial. For this reason, unless judges are able to write poetry in two lan-
that is, unless they can draft originals in two languages such that the
guages –
previous commitment to a particular expression in one does not commit them to a
parallel expression in the other –
they should not themselves attempt to write
opinions in two languages.
Obviously, few judges are as bilingual or trilingual as Samuel Beckett.'” Yet the
symbolism of legal bilingualism in Canada today requires, at least for Supreme Court
and Federal Court judgments, that a text be issued in both English and French. Con-
sequently, translation, rather than the dual craftsmanship suggested in the case of leg-
islation, is inevitable.”‘ In other words, it is better to maintain a fidelity to the judg-
ment-writing process at the expense of true normative bilingualism than to maintain a
fidelity to normative bilingualism at the expense of judgment-writing. For this reason,
the existence of “official translations” should not lead, as it once did, to the court de-
clining to indicate which text is a translation.'” An explicit acknowledgment of which
version is a translation ought, in fact, to free translators from excessive discursivity. It
should open the way for them to attempt to capture the rhetorical dimensions of the
original, and should encourage them to treat their work as more literary than bureau-
cratic. A citation to Shakespeare in an English-language original judgment should not
automatically lead to its direct translation in the French-language version of the same
‘”See Klinck, supra note 88.
,7,A notable exception was, of course, the late Jean Beetz. On Justice Beetz’s judgment-writing
habits, see G. La Forest, “Jean Beetz – Souvenirs d’un ami” in Milanges Jean Beetz (Montreal:
Th6mis, 1995) 143 at 14647.
… The expedient of translation is not as free from difficulty as one might think. Imagine the situa-
tion of unilingual judges; how will a judge know whether the translation adequately captures the full
flavour of the original judgment? How will the judge even know whether the implicit restrictions or
extensions of a principle announced in a judgment are conveyed by the translation?
,6 Beginning with the volume [1970] S.C.R., all judgments of the Supreme Court were published in
parallel columns in French and English, without an indication of which version was the original and
which was the translation. Beginning with the volume [1980] S.C.R., the law reports indicated either
“version franaise du jugement de …” or “English version of the judgment delivered by …”. Only on
rare occasions, such as the judgment of the whole court in Ford v. Quebec (A.G.) [1988] 2 S.C.R.
712, does the judgment not indicate the language in which it was drafted. The subject matter of the
namely, the constitutionality of certain sec-
case leading the court to depart from its usual practice-
is itself a matter of semiotic
tions of the Charter of the French Language, R.S.Q. 1977, c. C-Il –
significance.
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 42
judgment. Nor, afortiori, should other rhetorical devices, constructions and sayings
be literally translated.’-
Were all judges to be fluently bilingual, dual versions of judgments would be rare.
Typically judges would write only in one language, choosing that language on the
basis of the total context –
language of pleadings, language of litigants, predominant
language of extended audience, character ofjustificatory source material, and rhetori-
cal and syntactical properties of a language – of their literary endeavour.” In such a
world, their equally bilingual readers would reflect on the choice of language as part
of the meaning of the text being interpreted. Until such bilingual dialogue became
possible, however, fluently bilingual judges would be required to produce two origi-
nals, and unilingual judges would require the services of a translator. Whether bilin-
gual or unilingual, judges would frankly resist the temptations of discursive and bu-
reaucratic writing. The drafting of opinions responsive to metaphor and allusion
within a particular language tradition is the temporary and expedient alternative to
lulling unilingual readers into believing that official translation can compensate for
“partial knowledge and partial prophecy”.”
51. Enactments and judicial opinions do not exhaust the repertoire of normativity in
any legal system – even one that explicitly characterizes itself as embracing legal bi-
lingualism. Less discursive and less explicit norms, such as those found in practice,
usage, custom, contract, doctrinal writing, or supereminent or general principles,
might exist in only one language, if they are expressed in language at all. ‘ Moreover,
the practices and processes of official institutions, and of other facilitators of the state
legal order, such as lawyers, notaries and legal educators, also have significant nor-
mative import.”‘ Here again, the relationship of language to norm is often tenuous or
even nonexistent.
This presents the practice of legal bilingualism with a paradox: should these other
forms of normativity be rendered into language, or should they be excluded from
what is defined as “law”. For reasons already given, they cannot all be rendered into
language. Even those that can will often be produced only in one language version
such that the bilingual project can only be achieved through a derivative translation. “‘
‘7’ Thus, ceteris paibus, “putting one’s foot in one’s mouth” ought to be rendered as “se mettre les
pieds dans les plats” and “se noyer dans un verre d’eau” ought to be rendered by “making a mountain
out of a mole-hill”. Indeed, entire literary constructions –
extended metaphors, or the use of irony
rather than paradox- cannot simply be transposed from one language to another.
.Of course, on occasion, judges might also choose to write judgments in both French and English
for similar contextual reasons, or because they believe their ideas can be best expressed through a
metaphoric “autrement dite” rather than the simile “in other words”. Here the aspiration is not to rep-
licate what is said in one language but rather to enrich understanding by saying it differently.
” Paraphrasing 1 Corinthians 13: 9. Of course, given the political dynamics of Canadian law and
the demographics of the Canadian legal profession, this means that, except in Quebec, judgments
written in the French language are rarely read. See Deschenes, supra note 132 at 33-40.
“o See generally Brierley & Macdonald, eds., supra note 44 at 128-30.
… R.A. Macdonald, “Les vielles gardes” in Belley, ed., supra note 38, 233; Macdonald, supra note
137.
,’ This is the semantic point canvassed in Part L
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
In addition, they cannot be banished from the domain of “official” law. Even the most
comprehensive codification of norms ultimately rests on general principles not tex-
tually expressed.” The practice of legal bilingualism requires coming to terms with
these forms of normativity as they are and abandoning the pretence of translating
them into language, or converting single-language representations into bilingual rep-
resentations through bureaucratic translations.”‘
52. Achieving legal bilingualism presupposes legal authors and audiences that would
be willing and able to explore the symbolism of law by reference to legal texts in two
languages. These authors and audiences would not confuse text and norm. They
would see statutes as canonical expressions of a means for apprehending legal norms
and would find meaning through both language versions. They would have no need
for translated judgments of any kind, although were two originals of a judicial opin-
ion to be produced, they would find meaning through both versions, as they would in
the case of legislation. In all cases, the search for legal normativity would be the
product of-bilingual dialogue. The “translation process” would be one of normative
meaning to textual expression and vice versa; it would not be of English to French or
vice versa. Legal bilingualism would ultimately require bilingualism in all its practi-
tioners. Rather than encouraging or even allowing two distinct official legal cultures
to form around two languages, the practice of legal bilingualism would draw on both
languages to construct one official legal culture. In Canada today, that official legal
culture is neither French nor English, neither civil law nor common law; it is all these
together, with the ambiguity that such complexity implies.’
Until such a state is reached, translations of non-legislative and non-judicial ma-
terials from one language to the other will continue. But this temporary bow to the
imperatives of the present should not inhibit ambitions of a different timbre. Produc-
ing a set of materials now, through which the unfortunate practice of legal dualism
may be facilitated, may be a necessary transitional step towards (and a necessary price
to pay for) the achievement of legal bilingualism in the future.
,3 This is the epistemological point canvassed in Part II.
For example, there is currently a superb translation of R. Dussault & L. Borgeat, TraitJ de droit
administratif, 2d ed. (Sillery: Presses de l’Universit6 Laval, 1984-1989) by M. Rankin under the title
Administrative Lnv: A Treatise (Toronto: Carswell, 1985-90). Would the translator’s time have been
better served rewriting the book and refocussing it towards an English-speaking audience? How
much additional effort would this have required? Compare R. David & C. Jauffret-Spinosi, Les
grands systlmes de droit contemporains, 10th ed. (Paris: Dalloz, 1992) with R. David & J.E.C. Brier-
ley, Major Legal Systems in the World Today, 3d ed. (London: Stevens, 1985), an adaptation of the
8th edition of the French-language original and an example of beneficial recasting in the translation of
a doctrinal work.
‘ 5 In talking of bijuralism and bilingualism, I do not mean to imply that the configuration of the of-
ficial legal order in Canada ought to be (or even is today) merely “bi”. Increasingly, the “multi” fea-
tures of Canadian law flowing from a better understanding of Aboriginal justice are being acknowl-
edged (see Macdonald, supra note 124). The observations advanced here in relation to legal bilingual-
ism apply afortiori in multilingual legal environments.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 42
Conclusion
53. In the Introduction, the story of the Tower of Babel was presented, theologically,
as teaching that in the past, when human beings had but one language, they were able
to undertake a conquest of the heavens. Only God’s wrath prevented them from
building upon the unity of their speech.
But now another theological reading of Genesis 11 suggests itself. We were mis-
guided in believing that a single language would give us the power to reach absolute
heights. Far from being wrathful, God saved us from our hubris. To assure that we
would never fall prey to it again, we have been blessed with a multiplicity of lan-
guages. The existence of these languages allows us to reflect on the relationship be-
tween language and knowledge.'”
On this reading of Genesis 11, the problem of Babel is, fundamentally, that of
human individuality. Information and facts are a secondary part of all intersubjective
communication; myth, fiction and metaphor primarily characterize human language
and differentiate it from animal sign systems. Pace Wittgenstein, we can know more
than we can say. Indeed, the lesson of Babel is really one of legal pluralism. It is a les-
son about the capacity of human beings to construct and reconstruct meaningful
communicative symbolisms for the expression of normativity.
54. The problem that initially gave rise to this essay was how to account for the phe-
nomenon of official legal bilingualism. Is it possible to imagine a legal bilingualism
that does not reside in dualism? What does legal bilingualism imply about the role
and character of legal translation? The thesis of this essay has been that the existence
of a multiplicity of legal languages has much to teach us about the relationship be-
tween the different forms of symbolism found in law and the different ways in which
these symbolisms are deployed to generate meaning.
The possibility of legal bilingualism teaches that to pursue the goal of a univer-
sal communicative discourse through formal rationality in law is both futile and
presumptuous: law is more than discursive language; legal knowledge is more than
the rationalistic interpretation of texts. The possibility of legal bilingualism also
teaches that language cannot be equated with law: legal knowledge and legal text
are often mutually constitutive; but law is also grounded in nonlinguistic meaning;
the boundaries of legal normativity extend to the boundaries of human communi-
cative symbolisms.
“. Compare the account of Babel in The Book of J, trans. D. Rosenberg (New York: Grove Weiden-
feld, 1990).
1997]
R.A. MACDONALD – LEGAL BILINGUALISM
These lessons are not, however, just the lessons of legal bilingualism. They are
the lessons of law. Taking the claims of legal bilingualism seriously, and imagining
the semantic and epistemological foundations of legal normativity adequate to meet
its claims, is a heuristic for reconceiving law in a manner that overcomes the strictures
of Kabbalists and critical legal theorists alike. It is to conceive law as hermeneutic
aspiration.'” Far from being a source of despair in their complexity and in the intellec-
tual demands they place upon those who would practice legal bilingualism, these les-
sons are liberating.'” They reveal that legal understanding can only come from accept-
ing and adopting the multiplicity of expression, linguistic and nonlinguistic, as our
own.’1
‘7 For a like conclusion, see de Sousa Santos, supra note 123, who calls for a paradigmatic transi-
tion in the understanding of law.
‘. Of course I acknowledge that this essay has not addressed the question of social power. Yet no
such claims modify the analysis and conclusions presented here. This essay speaks equally to the
concerns of A. Hunt in Explorations in Law and Society (New York: Routledge, 1993), and those of
J. Vining in From Newton’s Sleep (Princeton: Princeton University Press, 1994).
” For further development of this theme, see R.A. Macdonald, “Multiple Selves and Critical Legal
Pluralism” (Law and Society Association Annual Meeting, Strathclyde University, Glasgow, 1996)
[unpublished].
