Rationality is dead! Long live rationality! Saving rational basis review
Author | Michael Bishop |
DOI | 10.10520/EJC153241 |
Published date | 01 January 2010 |
Date | 01 January 2010 |
Pages | 312-345 |
Rationality is dead! Long live rationality!
Saving rational basis review
*
Michael Bishop
**
[T]he word ‘test’ is inappropriate, at least insofar as it suggests some meaningful
analytical framework to guide judicial decision-making, because the rational basis
test is nothing more than a Magic Eight Ball that randomly generates different
answers to key constitutional questions depending on who happens to be shaking
it and with what level of vigor.
1
1Introduction
Rational basis review has been in trouble for a while.
2
For at least the past 35
years it has been constantly criticised for being ‘empty’,
3
‘toothless’,
4
inconsistent
5
and incoherent.
6
Despite its doubtful pedigree, litigants continue to rely on it and
*
I would like to thank Alistair Price for agreeing to reply to this paper and for the many conversations
which I can only hope were as enlightening for him as they were for me. This paper was originally
written for a seminar on public law at Columbia Law School run by Prof Gillian Metzger and Prof
Trevor Morrison. I would like to thank them both for giving me the opportunity to write the paper and
for their helpful comments on the original draft. I would also like to thank Stu Woolman, James
Fowkes, the anonymous referees and all the participants at the SAIFAC seminar for their comments.
All errors, of course, remain my own.
**
BA(Law) LLB LLM (Pretoria) LLM (Columbia). Visiting Research Fellow, Columbia Law School;
Extraordinary Lecturer, Department of Public Law, University of Pretoria.
1
Neily ‘No such thing: Litigating under the rational basis test’ (2005) 1 New York Journal of Law and
Liberty 898 at 898.
2
See, eg, Linde ‘Due process of lawmaking’ (1975) 55 Nebraska LR 197; Gunther ‘Foreword: In
search of evolving doctrine on a changing court: A model for a newer equal protection’ (1972) 86
Harvard LR 1; Note ‘Legislative purpose, rationality, and equal protection’ (1972) 82 Yale LJ 123;
Bennett ‘’Mere’ rationality in constitutional law: Judicial review and democratic theory’ (1979) 67
California LR 1049.
3
See, eg, Note (n 2) 128.
4
See, eg, Gunther (n 2) 18-19; Bice ‘Rationality analysis in constitutional law’ (1980) 65 Minnesota
LR 1 at 3-4; In re Agnew (7
th
Cir 1998)144 F 3d 1013 at 1014.
5
See, eg, Bennett (n 2) 1060; Farrell ‘Legislative purpose and equal protection’s rationality review’
(1992) 37 Villanova LR 1 at 2.
6
Meyers ‘Impermissible purposes and the equal protection clause’ (1986) 86 Columbia LR 1184 at
1184; Sunstein ‘Public values, private interests and the equal protection clause’ (1982) Supreme
Court Review 127 at 144.
Rationality is dead! Long live rationality! Saving rational basis review 313
the courts refuse to alter the test to answer their critics. This essay is one more
in a long lineage of attempts to figure out what is wrong with rational basis review
and propose a way forward.
Although I rely extensively on earlier work in the area, I have taken a different
tack. Firstly, I offer a comparative perspective that looks at the law in both the
United States and South Africa. The formulation of the tests and judicial attitudes
toward their enforcement are virtually identical in both countries. Yet the US has
a much longer history of cases and scholarship that address the problem that can
teach South African courts as much from its failures as its successes. South
Africa also makes an interesting comparison from the US point of view, because
it illustrates how even a system without signs of the extreme excesses of US
jurisprudence is in need of saving and may offer some hints on what the route to
salvation might be.
7
Secondly, my goal is not to identify all the inconsistencies and absurdities of
rational basis theory
8
or practice.
9
Instead I want to show how the test fails to do
what it is meant to do and to suggest that a re-think of how we justify and
conceptualise the test is required. I begin in Part 2 by specifying the form of the
test and noting some of the major differences between US and South African
jurisprudence. Part 3 examines the justifications for rational basis review and
deduces what type of test those justifications envisage and what types of laws
they require to be struck down. The heart of my analysis comes in Part 4 where
I describe how the rational basis test is incapable of uniform or routine application;
its outcomes cannot be neatly deduced from a formula but are almost entirely
dependent on the discretion of litigants, lawyers and, most importantly, judges.
Part 5 demonstrates how the analysis in Part 4 unmoors the rational basis test
from its traditional justifications. Finally, in Part 6, I suggest an additional
justification for the test that is compatible with the reality of rational basis review
and suggest what an appropriate formulation of the test might look like.
7
In his excellent reply, Price notes that there is an important difference between equality analysis in
the two countries: many laws that are tested under rationality review in the US will be tested under
the more exacting unfair discrimination standard in South Africa. Price ‘The content and justification
of rationality review’ (2010) 25 SAPL 346 at 347. This is true and perhaps explains why rational basis
review has attracted much more attention and criticism in the US than in South Africa. However,
there is still a significant class of cases – differentiation on grounds not listed in s 9(3) or analogous
grounds – that is subjected to rationality review in both nations. Those are the primary subject of this
paper. Price also notes that the criticisms leveled at American courts’ treatment of rationality review
cannot be uncritically transposed to South Africa. Again he is correct. South African courts have,
thus far, been far more circumspect in their application and explication of rationality review than their
American counterparts. There are places in the paper where I unfairly lump all the sins of one on
both. If I could rewrite it, I would change that. However, I would argue, and I would hope this paper
demonstrates, that the way the test is structured and the justifications offered in favour of it are
virtually identical on both sides of the Atlantic.
8
See, eg, Note (n 2); Linde (n 2).
9
See, eg, Neily (n 1).
(2010) 25 SAPL314
Three small notes are in order before I begin. First, I am only concerned with the
rational basis test in the context of equality claims. Much here might be applicable
to the other contexts in which rationality review occurs, but I do not directly address
those issues here.
10
Second, Alistair Price has written a fantastic response to this
essay in which he carefully tests my assertion that there is something wrong with the
current groundwork of rationality review.
11
I agree with much of what Price says and
believe his work deepens our understanding of the role rationality review can and
should play in a constitutional democracy. I am deeply indebted to him for his careful
and generous engagement with my ideas. However, there are areas – some more
important, some less so – in which we differ. In order to maintain the call-and-
response character of our conversation, I do not react to Price in the text, although
it certainly would be greatly improved if I did! Avid readers will, however, find my
thoughts on some of his claims in footnotes. Third, since writing this paper, the
Constitutional Court handed down four important decisions on rationality as an
element of legality.
12
I have not incorporated these cases into my discussion but refer
to them because they will be of interest to readers of this article. However, both Price
and I have discussed them and their relation to this paper’s thesis in other forums.
13
In short, I believe that the four decisions – especially when read together – confirm
and support the analysis offered below. But this article should be read as stating the
law at the end of 2009.
2 The test
The formulation of the basic test is remarkably similar in both South African and
10
Price does tackle rationality in all its guises. I think much may be gained from considering all occurrences
of rationality review together. However, I have two partially conflicting reasons for focusing on equality.
One, I think that, ultimately, virtually any case in which rationality arises can be framed as a differentiation.
There are technical reasons why the challenge will not be brought under s 9(1) – if, eg, it is a challenge
to a constitutional amendment – but the substance of the test is the same. Two, although any challenge
can be brought as a differentiation claim, something happens when we invoke the right to equality that
subtly changes the nature of the complaint. Unequal treatment is the starting point. Rationality is the
standard we use to determine whether the unequal treatment – which we intuitively think of as somewhat
unjust – is acceptable. When rationality is applied without the allegation of differentiation the flaw is the
irrationality alone, not any additional underlying wrong. That distinction will affect both the justifications for,
and I think the nature of, the test we use. These reasons justify looking at rationality separately in the
context of equality, even if the vast majority of insights are translatable to, and from, other contexts.
11
Price (n 7).
12
Albutt v Centre for the Study of Violence and Reconciliation 2010 3 SA 293 (CC), 2010 2 SACR 101
(CC), 2010 5 BCLR 391 (CC); Poverty Alleviation Network v President of South Africa [2010] ZACC
5, 2010 6 BCLR 520 (CC); Law Society of South Africa v Minister for Transport 2011 1 SA 400 (CC),
2011 2 BCLR 150 (CC); Glenister v President of the Republic of South Africa [2011] ZACC 6.
13
See, Price ‘Rationality review of legislation and executive decisions: Poverty alleviation network
and Albutt’ 2010 SALJ 580; Bishop and Brickhill ‘Constitutional law’ 2010 (1) Juta’s Quarterly Review
para 2.1-2.2; Bishop and Brickhill ‘Constitutional law’ 2010 (4) Juta’s Quarterly Review para 2.1;
Bishop and Brickhill 2011 (1) Juta’s Quarterly Review para 2.1.
To continue reading
Request your trial