The Footnote– Part III

Copyright 1998 Jack M. Balkin. All rights reserved.



The metaphors of Carolene Products are metaphors of inclusion and exclusion – of filled milk, footnotes, minorities, and defects in the democratic process. Yet Carolene Products is also exemplary of an even more basic form of inclusion and exclusion – the inclusion and exclusion that is necessary to theoretical conceptions expressed in our imperfect language. In identifying the pluralist, institutional features of Carolene Products, we have discovered that Carolene Products is a partial reading of West Coast Hotel, a particular path of doctrinal development chosen over other unarticulated alternatives. Like the development of Christian theology, the sense and direction of Carolene Products is partially identical to and partially different from the sense and direction of the texts it builds upon and seeks to interpret, including some aspects of these texts while excluding others.

This inclusion and exclusion, this partiality, is a characteristic feature of law conceived of as writing, as the creation of a series of texts through the reading and rereading of previous authoritative materials. Each successive reading is both a partial reading and a partial misreading of what has gone before, each understanding both a partial understanding and a partial

misunderstanding. And by misreading and misunderstanding, we do not mean merely mistaking the author’s intentions, but rather a selective grasping (or inclusion) of the unspoken possibilities of previous texts, which channels future interpretations and yet also creates new possibilities for understanding (and misunderstanding). (69) Yet to say that law is a kind of writing is also to say that Law is a kind of composition. Law, like religion, is a way of describing the world to ourselves, and in so describing, it helps to create that which it describes. Law remakes the world and ourselves as we use it to regulate and direct our activity. Thus we might say that law is constitutive – not only of itself, but also of our selves, our goals and values, and the society in which we live. (70) Yet the constitution and composition of our texts (and of our lives), is always both inclusion and exclusion; for it is determined both by what is absorbed and what is omitted, what is central and what is peripheral, what is emphasized and what goes unnoticed.

Carolene Products exemplifies these facets of law as writing. Carolene Products carries with it a dominant conception of political life, which is, of course, partial. This conception is both an inclusion (of a conception of politics, of a practice of judicial review, of a theory of institutional roles) and an exclusion (of alternative conceptions). Moreover, this partiality is not accidental. A deconstructionist would say that the very act of intellectual conception inevitably involves exclusion. The word “conceive” itself, derived from the Latin concipere, “to take in,” connotes the simultaneous inclusion of some while leaving others outside (just as biological conception can only occur where an egg is fertilized by one sperm to the exclusion of others).

Yet simultaneously deconstruction argues that – unlike the process of biological conception – the marginalization and exclusion necessary for intellectual conception is never complete. Traces of banished and deemphasized alternatives lurk within the dominant conception, supporting it and at the same time calling its dominance into question. This is the problem of the footnote writ large, the non-trivial lesson to be learned from the trivial example with which we began this Article. Indeed, our ability to glean important insights from the trivial is itself an example of the deconstructive point.

We have already seen how Carolene Products excluded the impurities of Lochner only to introduce new impurities in its own conception of judicial review, how it demarcated areas in which the political process was defective only by refusing to see deficiencies in the process as a whole, how it banished value choices from the judicial role only to see them reemerge in other contexts. The history of Carolene Products is the history of our discovery of its partiality, of its intellectual marginalizations. For history deconstructs – revealing that the dominant conceptions we use to understand the world at a particular point in time are increasingly inappropriate for solving the problems of later years. As events progress, altering our awareness of social reality, we discover the importance of what our theories marginalized or neglected, and how our conception has sown the seeds of its own destruction, adulteration, and putrification in what it has overlooked, in what it has excluded. Ironically, the impurity of a dominant conception – the source of its eventual decomposition and decay – is due less to what it lets in than to what it leaves out. Thus, as we enter the next decade, the next century, we will find that, for all the good Carolene Products has done us, its shortcomings will become more and more apparent. (71)

For example, in the innocent division of the world into text and footnote cases, between the heightened scrutiny of footnote four and the non-scrutiny of “Third,” we can see the beginnings of the bankruptcy of equal protection jurisprudence in the 1970s. We cannot place the blame for these difficulties entirely on Justice Stone, who offered footnote four as the beginning of a search for a modern constitutional jurisprudence. Rather, the fault lies in the shortsightedness of his successors, who allowed his suggestions to ossify into an unthinking paradigm of judicial practice. Thus, by the 1970s, Stone’s revolutionary approach had hardened into an equal protection doctrine featuring two tiers of scrutiny, where government action either received no review at all or a virtually irrebuttable presumption of unconstitutionality. The only issue in most cases was which level of scrutiny applied, which in turn reduced to the question whether a suspect class or a fundamental right was affected.

At first, this ossification did not appear to create any significant disadvantages. During the last years of the Warren Era and the first years of the Burger Era, the Supreme Court continued to add new fundamental rights, such as the right to travel in Shapiro v. Thompson, (72) the right to marry in Loving v. Virginia, (73) and the right of access to courts in Boddie v. Connecticut. (74) Yet because the two-tier theory seemed to offer no alternative between total deference and total unconstitutionality, the recognition of new fundamental rights and suspect clauses was made increasingly difficult. This difficulty suited conservatives like Justice Rehnquist perfectly well, for it guaranteed that the Court would be loath to introduce new theories requiring strict scrutiny. In such situations, the federal courts were left with only the alternative of rational basis review, which in most cases would do little to disturb the economic status quo.

This is not to say that the Supreme Court would not revise the list of suspect classes later on, as it did in the case of gender, or that it would not devise alternative theories of scrutiny, as it did in cases like Plyler v. Doe (75) or City of Cleburne v. Cleburne Living Center. (76) Nevertheless, the two tier system of equal protection of the 1970s created an orthodoxy, a doctrinal hurdle, that any new theories of discrimination or essential rights had to overcome. The gender and illegitimacy cases, for example, were regarded as exceptional situations that did not fit easily into the “normal” mode of constitutional analysis. And in those cases where the Court was unable or unwilling to abandon theoretical normalcy – Dandridge v. Williams, (77) Rodriguez, (78) Harris v. McCrae (79) and Beazer (80) – it relegated many of the most important types of flaws of the democratic process to the nonprotections of the rational basis test. (81) In this sense, later cases like Plyler and Cleburne must be understood as sports that reflect a profound dissatisfaction with the gradual stagnation of equal protection doctrine. Moreover, as cases like these become less and less exceptional, they announce the incipient rejection and transformation of the two-tier system, a transformation that we are living through even now.

The process of doctrinal ossification culminated in the 1973 decision in San Antonio Independent School District v. Rodriguez, (82) in which the Court not only held that education was not a fundamental right but reiterated that poverty was not a suspect classification. With that decision the Court seemingly closed off, for a time, the list of rights and classes that would be counted as

“footnote cases” – and would thus receive the benefits of heightened judicial scrutiny. In Rodriguez, Justice Powell surveyed the list of fundamental rights and decided arbitrarily that the Court would accept no more of them. In support of this decision, he announced that henceforth no right would be considered fundamental unless it had been explicitly or implicitly recognized in the Constitution. (83) This statement was itself an ossification of doctrine, a decision to consolidate constitutional principles at a particular point in history. For the right to procreate in Skinner v. Oklahoma (84) and the right to travel in Shapiro v. Thompson (85) were no more and no less implicated in the Constitution than the right to education in Rodriguez. The only difference was that the former rights had been mentioned in cases decided before 1973. All Powell could offer in justification of his ipse dixit was that these rights had already been let in the door – and that the Court would not compound that error by adding new rights. Thus, Rodriguez symbolized not only a form of doctrinal stagnation, but, ironically, a new form of exclusion – an attempt to place a hermetic seal on the logic of footnote four, to prevent the accelerating conversion of text cases into footnote cases.

As I have said, we cannot lay the blame for Dandridge and Rodriguez wholly upon Justice Stone. Yet there is a connection between the division of text and footnote and the later problems of the 1970s, between the paradigm of Carolene Products and the words of Justice Stewart in Dandridge v. Williams:

For this Court to approve the invalidation of state economic or social regulation as “overreaching” would be far too reminiscent of an era in which the Court thought the Fourteenth Amendment gave it power to strike down state laws “because they may be unwise, improvident, or out of harmony with a particular school of thought.” Williamson v. Lee Optical Co., 348 U.S. 483, 488. That era long ago passed into history… . To be sure, the cases cited, and many others enunciating this fundamental standard under the Equal Protection Clause, have in the main involved state regulation of business or industry. The administration of public welfare assistance, by contrast, involves the most basic economic needs of impoverished human beings. We recognize the dramatically real factual difference between the cited cases and this one, but we can find no basis for applying a different constitutional standard… . [I]t is a standard that is true to the principle that the Fourteenth Amendment gives the federal courts no power to impose upon the States their views of what constitutes wise economic or social policy. (86)

How did this state of affairs come about? How was the progressive vision of Justice Stone twisted into a heartless orthodoxy? How was revolutionary politics stultified into conservative dogma, the religious fervor of 1937 made pharisaic? The beginnings of the problem of the 1970s may be found in the articulation and division in footnote four itself, an articulation originally designed to remove obstacles in the path of progressive economic and social policies. Yet within this strategy lay the possibility of its own emasculation and deradicalization.

To understand this, one must recall the haphazard way in which even the most progressive Supreme Court of the century dealt with the problem of poverty. Many of the leading suspect class and fundamental rights cases of the late Warren and early Burger Courts were really decisions designed to protect the rights of the poor or the economically powerless. Certainly this is true of leading cases like Boddie v. Connecticut, (87) Shapiro v. Thompson, (88) and Graham v. Richardson, (89) at least before a more conservative Court gradually sapped them of their transformative force. (90) Yet, at the same time, the Warren and Burger Courts refused to accept poverty as a suspect classification, or promise judicial scrutiny where differential levels of economic power were implicated per se.

The resulting doctrinal schizophrenia is too easily explained as the Warren Court’s uncompleted progressive agenda rudely truncated by the Burger Court. It reflected a deeper ambivalence. During the 1960s and 1970s the Court clearly sensed the powerlessness of the poor, but was unable or unwilling fully to accept the claim that the democratic process treated lower income classes unfairly in general. Instead, the Court picked out fundamental rights whose abridgement affected the poor in significant respects, or else relied upon suspect classes that were, in some contexts, proxies for poverty. This is not to say that these suspect classes (race, national origin, gender, alienage, illegitimacy) and these fundamental rights (access to courts, right to travel) were not important and deserving of protection in their own right. Still, the issue of economic inequality hung around them like an unwelcome relative (a poor relation) whom one does not have the heart to expel but who is certainly not going to receive a permanent invitation to stay.

Hence, in all of the fundamental rights and suspect class cases of the 1960s and 1970s, the issue of economic power was both acknowledged and unacknowledged. The Court was willing to remedy the ineffectiveness of the political process to aid the poor in every way except directly. And in so doing, the Court created rationales for heightened judicial scrutiny that were so awkwardly constructed, so piecemeal, that they created sympathy for Powell’s skeptical conclusion in Rodriguez that it was time to stop picking out fundamental rights and suspect classes at random, that it was time to bar the door.

The source of the deep-rooted ambivalence that ultimately led to Dandridge and Rodriguez is not the series of appointments made to the Court between 1937 and 1971, although that undoubtedly had some impact. The source of the problem is in the ideology of democratic pluralism reflected in the intellectual framework of Carolene Products – in footnote four and the text, “Third,” that surrounds it. Within Justice Stone’s list of the causes of failures of democratic process, no mention is made of what seems today to be the most obvious cause of all – disparities in political power caused by differences in economic power. In 1938, Stone saw unreasoning prejudice, censorship, and limited access to the ballot box as the chief factors adulterating the purity of self-rule, and given the experiences of his day, these assumptions cannot be too much faulted. Yet fifty years of history would prove this viewpoint seriously inadequate. Today, in a country where traditional civil liberties are relatively well protected, we can see that disparities in wealth and economic power may poison the democratic process every bit as much as the imperfections Stone identified in his famous footnote.

Footnote four’s blindness to these impurities is not accidental – it is inherent in its logical structure. The importance of this logic consists less in what footnote four says than in what it does not say. In the second paragraph of footnote four, when Stone decries “legislation which restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation,” he does not consider the possibility that relative economic strength also affects the possibility of subsequent repeal. Nor does he imagine that governmental action that creates or reinforces disparities in economic wealth and power might have exactly the same self-sustaining effect as governmental action that more directly affects political rights of speech and suffrage. To use the famous example of Williamson v. Lee Optical, (91) if opthamologists and optometrists use draconian regulations to drive opticians out of business, it is hardly likely that the latter group’s weakened economic condition will enhance their political effectiveness in a subsequent battle for repeal. Moreover, as proponents of campaign finance laws have argued for years, freedom of speech means little if no one can hear you, either because you lack the money to make yourself heard effectively or because your opponents have drowned out your message by means of their superior resources.

Similarly, in paragraph three of footnote four, Stone speaks of “prejudice against discrete and insular minorities [as] a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities.” This implies that minorities are closed out of the process because others will not deal with them in the political arena. Yet paragraph three does not consider that the true cause of political powerlessness of minorities might be disparities in economic status which are the effects of previous prejudice or previous exclusions from the political process. By focusing on pluralist bargaining strategies, paragraph three captures the important insight that majorities can adopt self-reproducing strategies for retaining power. Yet at the same time, paragraph three declines to push the analysis of self-reproduction of status and power back one step further – it does not contemplate that dominant economic and social forces might combine to perpetuate an economic underclass, or create minority subcultures that feature poverty, lack of education, learned helplessness, and self-destructive behavior.

If the footnote is blind to these possibilities, it is because it defines itself in terms of a text (the body of the opinion) which thrives upon that blindness. For the argument of “Third” depends upon the assumption that differences in wealth or economic power cannot by themselves be possible causes of failures of the political process, cannot result in self-reproducing stratifications of political power. This assumption was in turn necessitated by Carolene Products’ peculiar interpretation of West Coast Hotel and the particular blindness of the Lochner era.

From the failures of the Lochner era, Carolene Products gleaned two important and interrelated lessons. The first lesson was that regulation with redistributive consequences could be in the public interest in spite of, and indeed because of, its redistributive effects. The second, which appeared to follow from the first, was that the judiciary should no longer concern itself with struggles over economic rights. If the faith of the revolution of 1937 consisted precisely in the belief that the distribution of wealth in society was none of its business (except in the case of a simple taking), how could the Court concern itself with the effect of wealth in other contexts – that is, with its effects on the democratic process? Thus, once again we see how much the logic of footnote four depends upon the logic of the opinion that surrounds it. The agnosticism of Stone, the faith of “Third,” was that a properly functioning political process would regulate the economy and redistribute wealth in the public interest – the Court would only remain concerned with the formal structures of the political process (the subject of the footnote).

In divorcing the structure of procedure from the structure of the economy, Carolene Products proclaimed the essential independence of political from economic liberty, and political from economic equality. That this, too, was the lesson of the Lochner era seemed the most obvious reading of the revolution wrought by West Coast Hotel. Yet here again everything depended upon the partiality of a particular reading and upon its accompanying blindnesses. We have seen that an institutional reading of West Coast Hotel misses its more radical, humanitarian aspects. For by daring to label the common law regime of property and contract a “subsidy for unconscionable employers,” (92) West Coast Hotel affirmed the connection between economic equality and substantive liberty, between economic power and political right. If the legislature was right to alter the economic status quo because that regime violated human liberty, then the distribution of economic power in society had everything to do with the liberty guaranteed by the due process clause. The lesson of Lochner was that courts should not hinder legislatures from pursing human rights through alteration of property rights; and that the state was responsible for the reproduction of disparities in economic power achieved through maintenance of the status quo.

By neglecting this possible interpretation of West Coast Hotel, by understanding it as the strict separation of political and economic liberty (conceived in noncommon-law terms), the pluralist faith of Carolene Products reintroduced, at a new level, the very evil that West Coast Hotel found in Lochner. If the Lochner court had seen differences in economic status as natural and not seriously affecting human rights, so now Carolene Products saw differences in political power stemming from differences in economic power as prepolitical and not seriously threatening the purity of the democratic process. Just as Lochner saw the right of economic participation as unaffected by differences in wealth, so now Carolene Products assumed that the rights of political participation were unaffected by these differences. Indeed, the pluralist credo of Carolene Products is hardly different than the famous Lochnerian credo of Coppage v. Kansas, (93) with the notion of political liberty substituted for that of liberty of contract:

No doubt, wherever the right of private property exists, there must and will be inequalities of fortune; and thus it naturally happens that parties [participating in the political process] are not equally unhampered by circumstances. This applies to all [political bargains]… . Indeed a little reflection will show that wherever the right of private property and [political liberty] co-exist, each party when [participating in the political process] is inevitably more or less influenced by the question whether he has much property, or little, or none; for [the political bargain] is made to the very end that each may gain something that he needs or desires more urgently than that which he proposes to give in exchange. And, since it is self-evident that, unless all things are held in common, some persons must have more property than others, it is from the nature of things impossible to uphold [political liberty] and the right of private property without at the same time recognizing as legitimate those inequalities of fortune that are the necessary result of the exercise of those rights. (94)

Why was the abandonment of West Coast Hotel’s more progressive tendencies so easy for a Court that viewed itself as liberal and progressive? Why was it content to exchange the wheat for the chaff? We must remember that in West Coast Hotel, the Court believed that the Washington legislature had acted to protect the rights of the poor. The Court was optimistic that the democratic process usually would recognize the need to move towards egalitarian measures, if only the heavy hand of the federal judiciary were removed. In this sense, the paradigm of West Coast Hotel gave the Court a false optimism; it led the Court to believe that there was something about the institution of democratic legislatures that made them more likely to regulate the economy and redistribute wealth in the public interest. Conversely, the Court assumed that the Lochner Court’s reactionary tendencies were inherent in the structure of an unelected judiciary, even though the result in West Coast Hotel itself belied this very assumption. Thus, from the perspective of a particular moment in history, the new liberal Roosevelt majority believed that institutional and substantive concerns were tied together in a much more permanent way than later historical experience would confirm. If the assumption of a connection between institutions and values seems particularly naive today, we should remember that it is a mistake frequently made. The Warren Court led many liberal thinkers to precisely the opposite institutional conclusions, and it was not until the rise of the Burger Court that liberal commentators once again began to recognize the ambivalent relationship between substantive and institutional considerations.

The historical situation in which the Roosevelt Court acted led to its institutional delusions, its value-free rhetoric of deference to democracy. Yet by disregarding the effects of differences in economic power on the proper functioning of the political process, the Carolene Products Court betrayed the revolutionary ideas of West Coast Hotel. In the very act of casting out Lochner, it depended upon Lochner-like premises. In its exclusion of impurities from the democratic process, it left untouched the strongest source of adulteration. The Carolene product of Carolene Products is, and always has been, adulterated – an impure substance created by a dual substitution: The replacement of the substantive interpretation of West Coast Hotel by an institutional one, and the exchange of the theory of judicial review in “Second” for that in “Third.” And even as these substitutions are made, they are concealed behind a myth of purity: An ideal legislature that has never sat, moved by artificial purposes that were never expressed, in the service of an artificial public good that has never existed.


If there is a recurrent theme to our discussion of Carolene Products, it is the exclusion that is never fully exclusive, the purity that remains always slightly impure. We have seen the themes of purity and impurity, inclusion and exclusion, so obviously the concern of the subject of this opinion (the shipment of filled milk in interstate commerce) follow us doggedly as we investigate its more theoretical aspects (the protection of civil liberties and the theory of judicial review). There is a peculiar irony and self-commentary even in this. For lawyers do not generally think that the metaphorical, figurative, or

syntactical features of legal texts should play any role in discussions of the meaning of these texts. Such concerns are not only extraneous to good legal analysis, but indeed, may tend to confuse the issue. Yet the more we profess our lack of interest in the metaphorical, figurative, and syntactic features of the text, (as yet another form of impurity to be excluded from legal analysis) the more assuredly they return to trouble the logic of the opinion. Once again, our ability to exclude and marginalize these elements is never complete; our attempt at purifying legal discourse of its literary aspects (leaving only reliable logic and fact) is always impure.

But surely, you will object, this conclusion follows only because I have deliberately tortured the text of Carolene Products, deliberately sought out its figurations and metaphors. In short, I have treated this legal text as if it were a poem! Yet why should anyone treat this opinion as a literary text, focusing on its plays on words, puns, and verbal associations? What are such matters doing in a law review article?

Does not this very question pose the issue of marginalization all over again? Does it not depend upon a vision of legal writing defined in terms of exclusion? Perhaps, only half jokingly, we should raise a Carolene Products defense in response: We need special protection for these figural aspects of the legal text, which are unlikely to be given due consideration by the normal process of legal analysis. Prejudice against discrete and insular metaphors may be a special condition which tends seriously to curtail those scholarly processes ordinarily to be relied upon for intellectual inquiry. (And yet – here I introduce a dangerous thought half whispered – perhaps metaphors in legal texts are not discrete and insular, something so easily excluded. Perhaps as

Professor Ackerman tells us, the groups most seriously discriminated against are diffuse and anonymous groups hidden within the body politic itself. (95) Perhaps legal analysis is itself inescapably figural, metaphorical, even as it denies this.)

But let us return to the question at hand: Why should one make so much of the literary characteristics we find in Carolene Products? Surely they are accidental to the structure of the argument – to the real issues of civil rights, judicial power and legislative discretion. These theoretical issues just happened to arise in a case involving the exclusion of impure milk from interstate commerce – they could have arisen in a case involving regulation of hours and wages, or even the manufacture of ball bearings. Nor are the syntactic features of the opinion any less contingent: the Court’s discussion of minority rights could easily have been made in the body of the opinion, rather than in a footnote. Why, then, pay attention to such contingencies, such unessential details?

These objections disregard, overlook, and exclude what literary scholars term an economy, a lucky condensation of features in a given text. A poem or an aphorism communicates effectively because of its play upon words, because of its witty connection between expression and intention. The literary critic does not turn up her nose at the happy associations between surface features of discourse; she knows that the connections of sound and style among the various elements of a poem are rich sources of meaning. She knows as well that such meaning cannot always be captured through a prose description of a poem’s methods and processes of signification. Nor does a politician eschew aphorisms when they elucidate a point, and yet such aphorisms gain their persuasive force from puns, rhymes, or other surface features of language.

We have indeed been fortunate to discover, in this opinion about the exclusion of impure milk, a happy confluence of metaphors that aid us in exposing the flaws and difficulties of the modern theory of judicial review first articulated in Carolene Products. But this providence is merely a symptom of a deeper logic at work in Stone’s opinion, a logic that would have doubtless manifested itself if Stone had chosen different words, or even if we had used another post-1937 opinion instead. Even if we accept the distinction between surface features of a text and its logic, we do not shun those surface features if they condense the problematics of the text in a provocative or imaginative way.

Thus my use of metaphor and figuration must be understood as a shortcut to understanding the structures and tensions of logic in modern constitutional jurisprudence, by exploiting the textual economies in a particular legal opinion. If I had been given, not Carolene Products, but Williamson v. Lee Optical, an opinion that depends upon the same pluralist conception of politics, I would have found the same difficulties, the same troubling features, even though the words and metaphors used would be different ones. Yet, I have no doubt, other surface features of that opinion would call attention to themselves and bring us to the same conclusions about the limitations of post-1937 jurisprudence.(96) As Derrida explains, if we had chosen a different opinion, a different description, “the effect would be the same, the only loss being a certain economic condensation or accumulation, which has not gone unnoticed.” (97)

By celebrating the textuality of a legal text, we are, in Jonathan Culler’s phrase – “betting with words” – seeking to find a happy economy or condensation of meaning in the ways in which the text refers to itself and to its metaphors and processes of signification. (98) For example, in his famous essay, “Plato’s Pharmacy,” (99) Derrida focuses on the use of the Greek word pharmakon (remedy) in Plato’s Phaedrus. Derrida concludes that it is a “lucky” word – it condenses much in its philological derivations and its resemblances to other words, like the Greek words for poison (also translated as pharmakon), magician, sorcerer, and wizard (all translated as pharmakeus), and scapegoat (pharmakos). Thus armed, Derrida gives us an elaborate critique of Plato’s philosophical system. Similarly, the words in the text of Carolene Products are also “lucky” words. They possess a strange economy that is equally uncanny, that troubles the substantive argument of the opinion, unveils its hidden assumptions, and silently rebukes it for its exclusions. Like Derrida, our bet with words has indeed unearthed a useful condensation (and not only of milk).

At the same time, there is another, equally important, and yet almost contradictory reason to be concerned with the surface features of legal language. To neglect the economies of expression within a text is to deny its writtenness, its textuality, its text-ture. Disregarding the contingent or surface features of legal language features is symptomatic of the very issue of marginalization, exclusion, and false preservation of purity that we have noted before. We believe that we preserve the real, pure meaning of the text we call Carolene Products by excluding superficial, contingent features of the text. We believe that we thus purify it, claiming that the same ideas could have been stated another way, unlike a poem, which resists paraphrase. Yet in so doing, we deliberately place blinders on the ways in which textuality (metaphor, expression, and so forth) affect our ways of thinking about legal problems.(100)

Does this mean that the lawyer should become a literary critic? Note the use of the word “become” – as if the lawyer were not one already. Has no one noticed that recent commentaries on Carolene Products are already brilliant examples of textual criticism? I am thinking, in particular, of the contributions of Professors Brilmayer and Ackerman. (101) Professor Ackerman shows us that Stone’s famous metaphor “discrete and insular minorities” misleads us about the ways in which the political process fails us. By demonstrating that diffuseness and anonymity can as easily lead to underrepresentation, Ackerman reveals the partiality and incompleteness of Stone’s original language. (102) Professor Brilmayer explains how the logic of Carolene Products, pursued to its conclusion, defeats its assumption that the political process adequately represents minorities except where legislation directly discriminates against them. She argues convincingly that the impurity in the process caused by underrepresentation cannot be cabined in, that it infects even neutral statutes. She concludes that the value-free rhetoric of representation in Carolene Products undermines itself, ultimately revealing its value-laden content.(103)

Are Professors Ackerman and Brilmayer legal critics or literary critics? Why ask such a question? Can we not admit that they are simply critics, or if one must have an introductory adjective, textual critics? They too, rely upon the processes of signification in this text (which we call an “opinion”). They too, are equally interested in the problematics of its language, except that at first glance they appear to investigate only a particular aspect of that problematics that we are used to hearing about in a particular academic context. Yet the problems of signification that we see as essentially “legal” in nature (whether such and such is a good distinction, a sound argument, a good description of the relevant facts), are affected by the writer’s choice of language, and more importantly, by those linguistic and cultural factors that she did not choose, but were chosen for her. Is it not clear that the difficulties Ackerman locates in the ideology of Carolene Products are reflected in the choice of an adventitious metaphor? Is it not obvious that the logical flaws Brilmayer unearths find their source in the rhetoric of value neutrality and pluralist democracy?

Perhaps you are impatient with this line of reasoning, this deconstructive dancing around the point. You reiterate your objection as follows: “Look here. I am reading a law review article in a law review. What you are suggesting does not fit the format of good legal analysis, it is not appropriate, it does not fit within an acceptable framework.” Yet why is it impossible for legal analysis to look further than the cut and slash of logical argument? Why do we believe that logical argument is hermetically sealed off from the vagaries of language – from the subtle but insidious influence of surface characteristics of

language, such as metaphor, rhyming, etymological and phonic similarity? Derrida argues that this is the very mistake that philosophers make in their insistence that philosophy is not simply another form of writing, that it can be shorn of its metaphorical baggage.

Why could a system of legal argument not rely on literary features of language, on puns and plays on words? Consider the Rabbis of the Talmud and the Midrash, for example. They saw their task as infinitely more important than merely the development of a system of secular regulation – for they were expounding the Divine law as received in the Torah. Yet in interpreting the Holy Scriptures they had no scruples about making a clever play on words or even misreading or mispronouncing a word in order to make a legal or ethical point. They well understood that such aphorisms effected an economy, or condensation, that was available in the text and could be exploited for the purpose of understanding the text. These same Rabbis made much of mistakes and errors in spelling and grammar in the Torah. Far from regarding mistakes as surface characteristics of a legal text to be disregarded, they assumed instead that no such mistakes existed without a hermeneutical purpose. Hence, they devised elaborate explanations for why errors were placed in the text, even making use of numerology in order to explain their deeper meanings. And such arguments were accepted not because they convinced logically, but because they revealed important facets of the text not discoverable through logic alone.

If we reject such wordplay and argumentation in our own legal texts, is this because we are so much wiser than the Rabbis of the Talmud and the Midrash, so much more advanced in our legal reasoning processes, or is it because of our cultural background, our self-imposed blinders and limitations, our decision to seal off the legal from the literary? We assume, without argument (based, one might think, on political and social factors of modern Western life) that good “legal” or “philosophical” reasoning cannot be the same as good literary analysis, that the textual analysis involved in one case is not the same textual analysis involved in the other. We assume, again without argument, that the problematics of texts must be divided up into those problematics that are suitable for lawyers to investigate (how well the argument is constructed, its reliance on logic and facts) and those that are suitable for the literary critic. Yet the division of these problematics is itself problematical.

Perhaps you will concede that if Carolene Products had been a religious document, or a Talmudic text, the type of analysis I have presented here might seem more acceptable. Yet, you may say, Carolene Products is not a religious text, and therefore this type of analysis is inappropriate. But this argument reveals its own cultural dependencies, its own nonlogical basis, the moment it is stated. It demonstrates that proper textual analysis (even for legal texts like the Torah) depends upon context, upon history, upon a tradition of hermeneutical practices, and is therefore not simply given, not simply immanent in the text itself. Anglo-american lawyers, and, I dare say, Western lawyers in general, have isolated a number of modes of dealing with texts, called them proper legal analysis, and discarded all other approaches. You may say that is because these modes of analysis deal with the intention or logic of the writer’s argument. And I respond that the intention of the author, or the reasoning of her argument, is often irrelevant to legal analysis in the West, because we often read texts in ways contrary to the intention or the reasoning of the persons who created them. What we call legal analysis or the logical analysis of a text, then, is simply the creation of a specific hermeneutical tradition of reading and rereading. And it is the height of arrogance for us to claim that a text can teach us no more than what we can glean from it using the limited tools that constitute Western Legal Analysis.

I must note here parenthetically that one of Plato’s great achievements was his struggle to create a separate discipline of philosophy, one that divorced philosophical argument from the figural language of the poets and the wordplay of the Sophists. Thus, in the Euthydemus, he has Socrates confront and defeat two brothers, who are known as eristics, or “fighters with words,” and who routinely use puns in an effort to trip up their opponents. (104) The brothers’ practice seems so odd to us, who work in Plato’s shadow, that we cannot understand why anyone ever thought their arguments convincing. This more than anything else demonstrates the extent of Plato’s victory in establishing a paradigm of philosophical argument. But in his great philosophical

achievements, has Plato saved us from error, or has he merely intervened in the linguistic field by dividing the world into philosophical discourse, which is literary without knowing it, and literary discourse, which is self-consciously literary?

Viewed in this light, it is no accident that Plato distrusted poetry, for poetry posed the greatest threat to a self-contained discipline of philosophy. Plato was suspicious of poets both for their capacity to persuade people through non-philosophical means and for their capacity to lead people astray. Yet poets could both persuade and mislead precisely because they reveled in theungainliness, the irrepressibility of language, while philosophers viewed language only as an unfortunate necessity, a wild beast that needed to be tamed. For the poet, the blessing of language was its infinite equivocality, its rich evocativeness. For the philosopher (and today for the lawyer) the curse of language was its opacity – for ideally, language should be transparent to reason.

In this essay, I have attempted to question the unspoken assumptions we make in reading legal texts. I have taken seriously the deconstructive claim that a text, even a legal text, may contain an economy: a savings, condensation, or husbanding of resources. I have noted some of the many economies at work in Carolene Products: the ability of this text to say many different and often contradictory things on many different levels at the same time, depending upon our focus. For my part, I do not think that we should turn up our noses to such economies, as one does to a bottle of spoiled milk. Rather, we should exploit these economies, examine the processes of signification in legal texts more closely, in the hope that we can understand the unconscious and unarticulated ways in which we create our world through our use of language. We should, in short, become rhetorical or literary economists – studying the ways in which our language conveys more than is meant, in which our language traps us, and in which our language talks about itself, reinforcing or contradicting what it says.

It is appropriate that there should be an economy in Carolene Products, a case about the regulation of the economy, and that this economy should remain unregulated, that is, without boundaries, uncanny and uncontainable. The opinion, which concerns the need for economic regulation, itself demonstrates the unregulatability of the economy of a text. For what is orthodox legal analysis but an attempt to regulate this economy – to delimit a certain set of features about a text that are appropriate for the reader to consider? Yet the very act of regulating the economies of a text robs the text of its richness (thus destroying its economy) and at the same time guarantees that its economies of expression will be on the outside, (like businesses that move away from a heavily regulated state) peering in and mocking the utopia that the regulator has sought to create. Shall we accept this regulation as the only possible regime, or shall we dare to reap the hidden economies of our legal texts? The texts await us, offering unbounded opportunities. We lack only the entrepreneurs.(105)

Go to Part I
Go to Part II
Go to Part III
Return to Writings Online


69. See Balkin, Constitutional Interpretation and the Problem of History, 63 N.Y.U. L. REV. 911, 931-42 (1988); Balkin, Deconstructive Practice and Legal Theory, 96 YALE L.J. 743, 774-85 (1987).


71. See Ackerman, supra note 16, at 744-46.

72. 394 U.S. 618 (1969).

73. 388 U.S. 1 (1967).

74. 401 U.S. 371 (1971)(recognizing right of access to courts in specific context of divorce proceedings).

75. 457 U.S. 202 (1982)(applying equivalent of middle level scrutiny to state’s denial of free public education to children of illegal aliens).

76. 473 U.S. 432 (1985)(applying heightened scrutiny under the guise of rational basis review with respect to discrimination against mentally retarded persons).

77. 397 U.S. 471 (1970)(upholding maximum welfare grant of $ 250 a month regardless of family size).

78. San Antonio Indep. School Dist. v. Rodriguez, 411 U.S. 1 (1973)(upholding unequal system of public school financing).

79. 448 U.S. 297 (1980)(upholding federal decision to finance expenses of childbirth under Medicaid program but not medically necessary abortions).

80. New York City Transit Auth. v. Beazer, 440 U.S. 568 (1979)(upholding policy of not hiring persons receiving methadone maintenance treatment even where applicants had successfully undergone treatment for at least a year).

81. The partiality and incompleteness of Carolene Products’ conception of judicial review became apparent in the rapidly evolving social and economic circumstances that followed the Second World War. By the 1970s the nature of the economy and governmental regulation had changed sufficiently that few economic equal protection and due process cases before the federal courts involved direct regulations of freedom of contract. Instead, the very success of the New Deal in fostering redistributive social welfare programs had set the stage for important constitutional questions concerning the distribution of government entitlements like social security, medical benefits, and education. Similarly, many of the so-called economic due process and equal protection cases of the 1970s really involved unsuccessful attempts to give protection to new types of suspect classes and fundamental rights that a changing society viewed as increasingly significant. See Balkin, Federalism and the Conservative Ideology, 19 URB. LAW. 459, 489-91 (1987).

82. 411 U.S. 1 (1973).

83. Id. at 33-34.

84. 316 U.S. 535 (1942).

85. 394 U.S. 618 (1969).

86. 397 U.S. at 484-86 (citations and footnotes omitted) (emphasis added).

87. 401 U.S. 371 (1971).

88. 394 U.S. 618 (1969).

89. 403 U.S. 365 (1971) (holding unconstitutional denial of welfare benefits to aliens).

90. See also Goldberg v. Kelly, 397 U.S. 254 (1970) (creating due process protections for recipients of welfare benefits).

91. 348 U.S. 483 (1955).

92. 379 U.S. at 399.

93. 236 U.S. 1 (1915).

94. Id. at 17 (substitutions in brackets).

95. Ackerman, supra note 16, at 724.

96. In Williamson v. Lee Optical, 348 U.S. 483 (1955), the Court allows a lobby of opthalmologists and optometrists to place serious restrictions on the practice of opticians. The rationale of the law is that opticians need regulation to avoid injury to their customers’ vision. Yet this case, which concerns a regulation designed to help the public see more clearly, is also a case about blindness and distorted vision. The restrictions on optical services (as a potentially deceptive trade practice) involve their own deceptions, as raw political power disguises itself in the form of the public interest. The optician may not fit lenses to glasses without a prescription, a requirement created by a previous prescription by the legislature, which distorts the public’s vision even as it attempts to protect it. At the same time, the rational basis test requires the Court to blind itself to the political struggle behind the statute, the victory of one lobby over another. In holding as it does, the Court assumes that it is not responsible if the legislature cannot see what is really in the public interest, or even if the legislature deliberately chooses to deceive the public. Perhaps this is because, as the Court seems to suggest, that there is no preexisting public interest to see in the first place.

Yet in the blindness of Williamson, all have trouble seeing – the Court as well as legislatures and the public. In Williamson, a case about the protection of vision, we witness the virtual elimination of vision (scrutiny). Not only does Justice Douglas not scrutinize the statute closely, he deliberately looks the other way. He refuses to see what is obvious on the face of the statute –that this regulation was designed to favor opthalmologists and optometrists over opticians, and in particular, to curtail the growth of cut-rate volume optical services in department stores. (Today, of course, we might look at some of the restrictions – those on advertising, for example – with different eyes.) The rational basis test of Williamson is not a neutral act of judicial deference. It is a lens through which one views the political process – rose colored glasses, if you will. It is a distortion of vision in which the pluralist bargaining process is made to appear fair and not in need of judicial supervision. The Due Process game envisioned in Williamson requires the Court to make up reasons to justify the statute, to see things that aren’t really there, or at least to view them in their best light. The Court, faced with claims of naked redistribution, refuses to see the nakedness. It declines to correct the legislature’s shortsightedness. Its credo, in this and every other economic regulation case, is that the majority rules – the eyes have it.



99. J. DERRIDA, Plato’s Pharmacy, in DISSEMINATION 61 (1981).

100. As Justice Cardozo warned us long ago: “Metaphors in law are to be narrowly watched, for starting as devices to liberate thought, they end often in enslaving it.” Berkey v. Third Ave. Ry. Co., 244 N.Y. 84, 94, 155 N.E. 58, 61 (1926). Yet Cardozo’s admonition does not go far enough. It does not warn us that our language is already and always inescapably metaphorical and figural; that a logically pure language, divorced from metaphor, has never existed. Moreover, our attempts at excluding metaphor result in a dual impurity – first, because the figural can never be completely banished from discourse, only neglected or overlooked, and second, because any claim that legal thought can rest upon a wholly nonmetaphorical ground involves its own adulteration and deception.

101. Brilmayer, supra note 16; Ackerman, supra note 16.

102. Ackerman, supra note 16, at 724-31.

103. Brilmayer, supra note 16, at 1330-34.

104. PLATO, Euthydemus, in THE COLLECTED DIALOGUES OF PLATO 385 (E. Hamilton & H. Cairns eds. 1961).

105. [Note to the editors of the Law Review – place the entire text of this Article (including this footnote with these instructions) in this footnote. If this causes a problem of infinite regress, improvise].