Arkansas Law Review
School of Law
Waterman Hall 107
Fayetteville, AR 72701


Phone: (479) 575-5610

School of Law
1045 W. Maple St.
Robert A. Leflar Law Center
Waterman Hall
University of Arkansas
Fayetteville, AR 72701

Phone: (479) 575-5601

Law School Directory

Facebook Twitter Linkedin Youtube

Antipoverty in Constitutional Law: Some Recent Developments

By Frank I. Michelman, Robert Walmsley University Professor Emeritus, Harvard University. For a session on constitutional consideration for the poor, held at the University of Arkansas School of Law on March 7, 2013, I offered these early thoughts about a connection between: (1) the idea that an antipoverty commitment should rightly compose a part of a country’s constitutional law; and (2) debates around the world about the forms of judicial protection for constitutional rights. A fuller, more advanced presentation of the thoughts I outline here will appear as Chapter 15 in Social and Economic Rights in Theory and Practice (Helena Alviar, Karl Klare & Lucy Williams eds., forthcoming 2014). For this publication of my talk as the Hartman Hotz Lecture, I have freely rewritten the oral transcript to make it presentable in written form (and also supplied a few documentary references) while endeavoring to leave the substance and style of the talk unchanged.


[[213]]How, if at all, should conditions of poverty, or the plight of a country’s poor, become a topic of concern in that country’s constitutional law? I will point to some recent responses to that question in places around the world, and then I will offer some thoughts about how those responses may be shaking up some traditional, long‑standing alignments in understandings of what we mean by ideas of constitutionalism, constitutional law, and constitutional rights.

I said “around the world.” What about here at home? While American legal scholarship has surely been at the forefront of the movements in thought that I will discuss, 1 one cannot say that these movements are especially noticeable in the current operations of United States constitutional law—as they are, however, in the laws and legal operations of quite a few other of the world’s constitutional democracies. 2

Sometimes, not always, these appearances start with express commitments, in constitutional texts, to the fulfillment [[214]]of everyone’s basic economic needs, at social expense where necessary. For example, Section 26 of the Constitution of South Africa provides that everyone has the right to have access to adequate housing. 3 Section 27 provides that everyone has the right to sufficient food, water, healthcare services, and social security. 4 Those are laid down as constitutional rights. You do not see anything like them in the Constitution of the United States, and anyone familiar with decisions like that in DeShaney v. Winnebago County Department of Social Services will understand that claims to positive state assistance will not be judicially read into our national constitutional law any time soon. 5

In saying so, I have in mind that courts elsewhere have, in fact, sometimes found that their country’s constitutional law supports such claims even though they are not expressly or extensively spelled out in constitutional texts. Here is an example. The Basic Laws of Israel (which, as construed and applied by that country’s Supreme Court, serve as a body of constitutional law for that country 6 ) include a guarantee to everyone of “protection of their life, body, and dignity.” 7 The Israeli Supreme Court has construed the protection-of-dignity [[215]]clause to include a responsibility of the State to ensure that everyone is able to obtain access to levels of nutrition, shelter, education, and the like required for a minimally dignified human life. 8

You might start to think that including guarantees of that kind in a constitution as legal guarantees—and, thus, as rights that people can go to court to claim—might present those courts with some difficult challenges. We will come to that soon enough, but the first point to settle is just this: that, as a matter of fact and whatever the pros and cons, the idea has apparently taken hold in many countries that the state’s basic charter and highest laws should include commitments to the fulfillment for everyone of the basic economic necessities of a humanly dignified existence, perhaps even as a prerequisite for the general moral supportability of the state’s exercise of its powers of legal rule. 9


Against that background, my next step will be to put before you two contrasting positions regarding the manner in which constitutional rights and commitments are given effect by a country’s political and judicial institutions. The two positions have usually been perceived as opposing, but I will also be pointing to some common ground that they share.

Consider, then, two parties of thought. I will call one of them the “Liberal Constitutional Mainstream” or “LCM.” It includes me and, I feel it safe to say, many of you as well. The other party I will call the “Democratic Left” or “DL.” I am tempted to say that also includes me and many of you, but none of us can be fully and completely both LCM and DL—not in the ways I am going to present the two positions.

A. The Liberal Constitutional Mainstream

We will spend some time first with the LCM. The LCM has been concerned above all for the observance of certain [[216]]preconditions to the moral legitimacy of a coercive state legal order, and it views the constitution as a legal codification of those preconditions. It thus sees the practice of writing, recognizing, and heeding a constitutional bill of rights as a crucial part of a response to a basic question of political morality, as this party understands political morality.

To say just a bit more about that: The state is powerful and always potentially coercive. The state creates and recognizes laws, expecting and calling upon all citizens normally to abide by those laws just because they are its laws, standing ready when necessary to back that expectation with force. Now, the LCM takes the position that if these demands and powers of the state are to be justifiable to everyone subjected to them, then there are certain preconditions for the conduct of political and social life that will have to be fulfilled. Some of these preconditions are institutional and procedural: The state’s decision‑making will have to be in some basic, broad sense democratic. Where not by the people directly, it will have to be accountable and responsible to the people. Those are process‑related concerns.

In the view of the LCM, there are further preconditions having to do with the substance of political aims and outcomes. You can think here about the kinds of commitments we see mentioned in constitutional bills of rights: to freedoms from physical abuse and detention; to freedoms of expression, association, religion, and the like; in short, the traditional classical catalogue of liberal civil rights or, as they are sometimes called, negative liberties—“negative” because the traditional catalogue takes mainly the form of a series of “thou shall nots” directed to the state.

But perhaps not all the preconditions for a morally sustainable legal order can be negative in that sense. We envisage (let us say) a society where economic life is organized largely through markets—not exclusively so and not unregulated markets, but where economic freedoms and economic markets are primary institutional components. We have good reasons for that, but we also know in that kind of system and in the absence of a protective commitment to the contrary from organized society, there are going to be people for whom avenues to self-support, including by work on terms consistent with human [[217]]dignity, are lacking. Poverty, in other words, can be structural, even perhaps in some ways inheritable, and not just personal.

If you think about poverty in that way, as an ever-present structural potentiality in a liberal‑market‑organized society, you will easily see how liberal thought—the LCM—can come around to the idea that in some way, in some form, guarantees against denial of access to basic economic necessities and a fair chance to obtain them would become a part of that package of commitments on the basis of which we say: “On condition that those commitments (among others, of course) are satisfied, the free play of democratic politics and the free play of market economics not only can be allowed to go forward but does so justly and appropriately.” And if you get that far (while still remaining a part of the LCM), you might well also think that the constitution and constitutional law are the institutional sites where the guarantees are publicly declared and the commitments take hold. And so, you will think, maybe antipoverty commitments belong in constitutional law.

Now obviously what we are concerned about, morally and practically, is not just what is nominally guaranteed, it is also what kind of responsive effort really takes place on the ground. But of course that holds as well for all of the classical liberal constitutional guarantees regarding freedoms of expression, religion, property, privacy, and so on. That is why the LCM historically has stressed the need for some socially recognized, institutionally dedicated forum in which arguable violations can be brought to public attention and correction can be obtained. We call those forums courts of law. We call their service (in the context now under discussion) judicial constitutional review.

And so, in sum, the LCM—concerned (as I have said) for the observance of those preconditions to the moral legitimacy of a coercive state legal order—has traditionally stood in support of a strong role for politically insulated courts to serve as a guardian for those basic rights whose fulfillment we think of as required for the moral legitimacy and of the entire state legal order. The courts thus acquire an authority to override or countermand actions of the government—and that could extend also to governmental failures to act if the constitution contains an antipoverty commitment—when those actions or failures are judicially found to be in violation of one or more constitutional rights. But those rights are typically named by abstract terms—[[218]]”liberty,” “equal protection”—and so when courts decide on their more concrete applications (say, the application of clauses on “liberty” and “equality” to state restrictions on abortions of pregnancies), they get to decide, in the place of society as a whole, disputable questions regarding matters that, to many of us on many occasions, are of the deepest possible individual and public moral concern.

B. The Democratic Left

That is where the Democratic Left—the DL—comes in, so let us now turn for a while to that side of the picture. The DL is not so happy with the sort of strong-form judicial constitutional review that the LCM has tended to support. That is not because the DL does not share with the LCM the kinds of ideas I have been putting before you about the preconditions of the moral legitimacy of the state. What partly puts off the DL is the entrustment to courts of the specification of what those preconditions are. Constitutional clauses, to repeat, are usually written in general, abstract terms that do not supply clear and direct answers where the strongest political disagreements tend to arise. “Free exercise” [or “no establishment”] of religion does not tell you directly whether or by what means the state may or may not support denominational schools. Neither “free expression” nor “equality,” nor both in combination, directly answer hard questions about the state’s powers and duties to regulate racist and other hate speech. In different countries across the world, competent and sincere judiciaries answer those questions differently from the ways our courts do, and the differences are not persuasively explainable by differences in the constitutional texts. I make that observation not to show that our answers (or theirs) must, therefore, be wrong, but rather to remind us of how the questions posed by the application of these constitutional abstractions to contested cases can, and often will, be fairly and reasonably debatable, and so whoever decides the debates, thus, exerts a good deal of political authority.

That is a serious problem for the DL. The DL does not doubt that there truly are substantive preconditions for the moral legitimacy of the state’s power to rule by law, or that there are better and worse answers to questions about what those preconditions are and how they should be applied in debatable cases. What the DL doubts is that the resolution of these [[219]]questions should be left to a small, politically non‑accountable body of judicial officeholders to decide in a legalist frame of mind, rather than left to be worked out by deliberations of the people or the people’s elected representatives.

And now here is a further consideration. The DL—the Democratic Left—is, after all, a left. It is egalitarian and redistributivist. It would exert political control, if it had political control, to do its best to make sure that antipoverty goals are vigorously pursued, quite aside from anything the constitution might or might not have to say about this matter. And the DL does not entirely trust the likelihood that those who ascend to high judicial positions in one or another country will be fully sympathetic to antipoverty as a truly top-level constitutional concern. The DL worries that an elite judicial body will weigh, say, property rights too heavily as against antipoverty commitments.


Everything I have said up to here—both about the rise of antipoverty as a constitutional commitment and about the tensions between the LCM and the DL over the ways and means of constitutional application and enforcement—may be old news to most of my audience. What comes next may be a little less so. What I want to suggest is that, at least partly in response to an accelerated emergence of antipoverty commitments in constitutional law, signs appear within both the LCM and the DL of a possible shift towards a middle ground of so-called “weak-form” (as opposed to “strong-form”) judicial constitutional review. 10

Within the LCM, you see signs of movement away from single-minded attachment to strictly strong-form judicial guardianship of constitutional rights towards a distinctly modified—in some sense softened—conception of the role of the judicial branch in that department of a country’s politics. Within the DL (having in view countries where antipoverty guarantees have made their way into constitutional law but political leadership is perceived as failing in a consistency of [[220]]commitment to those guarantees), you see developing a sense that maybe some judicial muscularity would not be so bad. 11 Is there not some good way, the DL is prompted to ask, of putting our courts of law to work towards the end of getting the state’s efforts and results more effectively in line with the constitutional commitments?

At a first and even a second look, you may say the answer is no. To set the courts to work on behalf of antipoverty commitments (you might say) is to ask from them more than they are well set up to deliver. We would be calling on courts of law to assume some measure of positive direction over state policy in the antipoverty field, as if these law-trained judges should know better than the responsible ministries, parliamentary committees, and their staffs what is affordable and what might work best. What is more, the courts would be assuming this directive authority not in the name of crisp and decisive legal standards but, rather, in the name of the vague-seeming, open-ended guides that constitutional antipoverty guarantees must all but inevitably adopt. Why all but inevitably? Because in a country like South Africa (or actually just about any country you can name), a constitutional commitment to everyone having access to adequate housing cannot be construed as promising everyone a house by tomorrow. You cannot be inviting everyone in genuine need to come to court and plead: “I do not have adequate housing. I do not have and cannot get the money to acquire adequate housing on the market. Where is my house? The Constitution says I get one now.” That might work for the first case or two, but not for long.

Not surprisingly, then, constitutional guarantees in the antipoverty field, where they are found, are normally written and construed to impose a kind of best‑efforts obligation on the part of the legislative and executive branches of the state. 12 The commitment is to move the society over time towards satisfaction of antipoverty guarantees as widely and fully as possible, but subject to budgetary constraints and also to constraints imposed by a due regard for other constitutional [[221]]rights, such as rights to freedom of expression, freedom of association, personal liberty and security, and so on.

A constitution’s antipoverty commitment thus poses a complex challenge. It may very possibly engage every aspect of public policy you can think of: monetary policy; jobs policy; industrial policy; education policy; health policy; trade and import policy; family policy. Everything is implicated. Over the course of every legislative session, there will be plausible occasion to complain that the state has passed up an opportunity to make better antipoverty progress than it has. With a guarantee of that kind on the table, you can understand that courts of law might feel some caution about getting in over their heads.

What courts, in fact, have tended to do is fall back on an idea of “reasonableness review” or something similar. The courts undertake to rule on constitutional antipoverty challenges, but not to the point of deciding positively on a required line of action by the state; rather, only to the point of deciding whether the state’s challenged policy or course of conduct can be called an unreasonable choice, given the constitution’s antipoverty commitment. But then what kind of remedial order does the court issue, in case its answer is yes? The court might just declare the state’s performance, or its policy, to be constitutionally unsatisfactory and leave it at that—but where does that leave the constitution as the country’s supposed highest expression of the rule of law? The court might issue what is sometimes called a structural injunction or interdict. It might, in other words, say to the state: “Would you kindly go away and think about this, and then come back and talk to me, and we will see.”

The DL watches and thinks to itself:

These courts do not respond in that way to free-speech rights or property rights. When an otherwise homeless person with no lease is occupying someone’s property and the lawful owner comes to court and says, “I need an eviction,” the court gives the owner the eviction. Same with restrictions on freedom of speech, and on abortion, where that is considered a constitutional right. The courts cut the ice.

And a part of the DL then goes to the next step, saying to itself: “I want to see something more like comparably muscular [[222]]judicial enforcement of the Constitution’s antipoverty rights.” The DL, responding to the emergence of constitutionalized antipoverty rights, finds itself moving in the direction of support for strengthened judicial authority in the constitutional field.

And now we cut back to the LCM. It also is feeling the pressure, but from a somewhat different direction. We have just been seeing how the strong form of so-called judicial supremacy in the constitutional field runs into trouble with antipoverty rights when those have been constitutionalized. If, therefore, from within the LCM, you feel a strong pull of fidelity to constitutional antipoverty guarantees (where those exist), you may also feel yourself opening up to conceptions of constitutional legality and bindingness that do not always and inevitably depend on strong-form judicial review. And so, in fact, it has been happening. Over the past ten or fifteen years, from various sources, including the Liberal Constitutional Mainstream, there has been flowing an increasingly rich and sophisticated advocacy for an alternative, weaker-form conception of judicial constitutional review. 13

Meaning what? Meaning the courts give their considered opinions, but those do not necessarily serve as the decisive last word. The exact institutional forms and protocols of weak-form judicial review are numerous, and I cannot describe them all. But the idea at which they all aim is that what a constitutional guarantee binds the country to is a commitment to serious and focused deliberation on behalf of certain named values and principles. The guarantee obligates the government to focus its mind and to deliberate committedly about the best ways right now to proceed with implementation, subject to the kinds of material and normative constraints I mentioned earlier. It correspondingly binds the courts to respond to credible citizen complaints of a governmental default on that obligation. The court then, in some way, still officiates. The court calls attention to shortfalls in the state’s efforts as it may see them. It may kick things back to the government or legislature for a try-again or think-again. But it does not undertake to dictate a final, conversation-stopping answer.

[[223]]You may well think, and it may well make you uneasy to think, that what you see here taking shape is a potentially deep alteration of the LCM’s traditional understanding of what a constitutional right is, what the rule of law is, what it means for the constitutional right to be law. One might try to imagine how this could spill over to approaches taken by courts to classical liberal constitutional rights, the negative liberties. Say, a new-model New York Times Co. v. Sullivan: 14 Forget about judicial development of a strict rule of law or constitutional law (no liability for defamation of a public figure without an evidence-based finding of actual malice). Instead, the Supreme Court says to the State of Alabama: “We think you wrote your defamation laws with too little deference to freedom of political expression. So we want to kick that back to you to think again, and you come back and tell us what you did, and we will see how it goes from there.” Now, that little fantasy is not a prediction. I do not believe that is where matters are headed any time soon. I mean only to suggest how the widening appearance of antipoverty commitments in constitutional law is doing its part to complicate thought about the basic institutional forms and arrangements of the constitutional state.


Given (where it occurs) the constitutionalization of antipoverty, the Democratic Left newly and freshly sees the point both of having the preconditions for the moral legitimacy of the state written into constitutional law and of having some dedicated institutional setting in which public accountability of those who are responsible for performing the conditions takes place. Embracing (insofar it does) the constitutionalization of antipoverty, the Liberal Constitutional Mainstream starts to yield ground on judicial supremacy and to think its way toward some modified, softened form of judicial contribution to the debates of the commonwealth. The two old adversaries move towards a meeting.

So there is my story. The poor can no longer be called the forgotten of constitutional law. Much to the contrary, their claims to constitutional consideration now stand at the center of [[224]]a network of debates of potentially great significance for the project of constitutionalism taken as a whole.


  1. See, e.g., Mark Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law (2008).
  2. By “United States constitutional law,” I mean, specifically, the law set down in the Constitution of the United States along with judicial and other official interpretations of it—what lawyers commonly call “federal” constitutional law. The case with regard to state constitutional law is somewhat different. See, e.g., Helen Hershkoff, Forward: Positive Rights and the Evolution of State Constitutions, 33 Rutgers L.J. 799 (2002).
  3. Section 26 of the South African Constitution provides, in pertinent part: “(1) Everyone has the right to have access to adequate housing. (2) The state must take reasonable legislative and other measures, within its available resources, to achieve the progressive realisation of this right.” S. Afr. Const., 1996 § 26.
  4. Section 27 of the South African Constitution provides, in pertinent part: (1) Everyone has the right to have access to (a) health care services, including reproductive health care; (b sufficient food and water; and (c) social security, including, if they are unable to support themselves and their dependents, appropriate social assistance. (2)The state must take reasonable legislative and other measures, within its available resources, to achieve the progressive realisation of each of these rights. S. Afr. Const., 1996 § 27.
  5. See 489 U.S. 189, 195 (1989) (holding that United States constitutional law imposes no obligation on state governments to offer or provide protection against private violence).
  6. The leading decision is CA 6821/93 United Mizrahi Bank Ltd. v. Migdal Coop. Vill., 49(4) PD 221 [1995] (Isr.), available at
  7. Basic Law: Human Dignity and Liberty, 5752–1992, SH No. 1391 art. 4, available at Article 8 of the Basic Laws provides: “There shall be no violation of rights under this Basic Law except by a law befitting the values of the State of Israel, enacted for a proper purpose, and to an extent no greater than is required.” Id. art. 8.
  8. The leading decision is HCJ 366/03 Commitment to Peace & Soc. Justice Soc’y v. Minister of Fin. 60(3) PD 464 [2005] (Isr.), available at
  9. For summary of an argument to that effect, see Frank I. Michelman, Poverty in Liberalism: A Comment on the Constitutional Essentials, 60 Drake L. Rev. 1001 (2012).
  10. The terms are a contribution of Professor Tushnet. See, e.g., Tushnet, supra note 1.
  11. See supra notes 7-8 and accompanying text (illustrating this development in Israel).
  12. See, e.g., S. Afr. Const., 1996 §§ 26(2), 27(2).
  13. See, e.g., Stephen Gardbaum, The New Commonwealth Model of Constitutionalism: Theory and Practice (2013); Tushnet, supra note 1.
  14. See 376 U.S. 254, 279-80 (1964).