Amar 1999 Intertextualism

2023-10-04: Amar Intertextualism (PDF) 1999

Here is another feature of the Constitution: various words and phrases recur in the document. This feature gives interpreters yet another set of clues as they search for constitutional meaning and gives rise to yet another rich technique of constitutional interpretation. I call this technique intratextualism. [emphasis added]
[Part I. Cases and Commentaries]
By confining this survey to what are generally deemed the most canonical cases and commentaries, I aim to avoid stacking the deck.
After we have seen a wide variety of its uses, abuses, and nonuses, we will be better positioned to theorize about what we have seen (in Part II) and to consider how powerful the
Together these four cases mark the basic outlines of conventional constitutional doctrine, characterized by judicial review of both legislative and executive action (Marbury); broad but theoretically finite federal power (McCulloch) that states may not obstruct (McCulloch, again, along with its cousin, Martin v. Hunter's Lessee); emphatic norms against governmental efforts to subordinate or stigmatize racial minorities (Brown and its companion Bolling); and broad protection of judicially defined fundamental rights that may or may not be clearly stated in constitutional text (Roe) [moved from being inside McCulloch below] technique can be at its best (in Part III).
[Part I A. Cases]
[Part I A. Case 1.] McCulloch's [Marshall 1819] claim to canonical primacy, however, rests on more than its doubly significant substance affirming generous congressional power in the first half of the opinion and important limits on state governments in the second half. [emphasis added]
To read McCulloch is to see (and for many beginning students, to learn) how to do constitutional argument.
[Part I A. Case 1. (a)] McCulloch and constitutional argument generally.
Maryland apparently claimed that the Necessary and Proper Clause, requiring the Court to construe the various enumerated powers in Article I more strictly than it otherwise would in the absence of this clause.
In response to this claim, Marshall trumpets the text. Instead of affirmatively declaring ... the clause would have been negatively written ... 'no laws shall be passed but such as are necessary and proper.' [emphasis added]
Marshall buttresses this narrow textual argument with a narrow historical argument. [emphasis added]
Intertwined with Marshall's appeals to text and history are sturdy structural arguments rooted in federalism and populism. [emphasis added]
Thus Marshall argues that the nature of the document repels the idea that every conceivable federal power - such as the power to create a bank - must be spelled out in minute detail rather than implied by and subsumed within the general structure of broadly crafted enumerated powers.
Structural arguments also loom large in the second half of McCulloch, in which Marshall proclaims that Maryland may not tax the charter of a lawfully established federal bank. ... as a matter of general structural logic, surely the part cannot control the whole. (To the colonial patriots, the power of Parliament to pass a tiny tea tax implied the power to tax without limit, which in turn implied plenary parliamentary power in America.)
... structural argument often goes hand in hand with a certain kind of pragmatic argument. Stingy construction of the Constitution, Marshall argues, would offend the nature of the Constitution not merely as a suitably nationalist and populist document, but also as an inherently practical document.
... doctrinal argument, based on judicial precedent. Many readers have noticed that Marshall cites no cases by name.
The rule to be laid down in the case at bar must be capable of being followed by lower courts and a later Court.
Text, history, structure, prudence, and doctrine - these are the basic building blocks of conventional constitutional argument.
Philip Bobbitt suggests that a sixth form of constitutional argument exists and merits attention -- what he calls "ethical [ethos] argument." In the vernacular, an ethical argument might declare a practice unconstitutional because it is "unAmerican," or might affirm the constitutionality of a contested practice because it is part of "the American way."
[Part I A. Case 1. (b)] McCulloch and intratextualism.
But Maryland's counterargument, interpreted most charitably, is that the various enumerated powers should be construed far more strictly than Marshall has proposed, and that the specific words of the Necessary and Proper Clause confirm the imperative of strict construction. Congress should enjoy only those implied incidental powers that are logically "necessary" to carry out its express powers.
But, Marshall counters, the word does not invariably (a wag might say "necessarily") mean this: "... we find that it frequently imports no more than that one thing is convenient, or useful ... to another."
Thus far, Marshall's analysis seems methodologically unremarkable. It is a standard clause-bound exegesis appealing to plain meaning, ordinary language, and (perhaps implicitly) original intent.
[Marshall] turns to another passage in the Constitution itself, in effect using the Constitution as its own dictionary: "It is, we think, impossible to compare the sentence which prohibits a State from laying "imposts, or duties on imports or exports, except what may be absolutely necessary for executing its inspection laws," with that which authorizes Congress "to make all laws which shall be necessary and proper for carrying into execution" the powers of the general government, without feeling a conviction that the convention understood itself to change materially the meaning of the word "necessary," by prefixing the word "absolutely." " [emphasis (bolding) added. italics in the original text.]
Marshall shows that the Constitution itself proves that "necessary" is not always a term of math or logic; that it sometimes takes an adverb that can modify its strictness; and that without an adverb such as "absolutely," the word as used in the Necessary and Proper Clause can be read flexibly not strictly, practically not mathematically.
Here then we see a classic example of intratextualism: establishing the meaning of a word in one constitutional clause by analyzing its use in another constitutional clause. [emphasis added]
And when we accept Marshall's invitation to inspect the Necessary and Proper and Territorial Clauses side by side, we see further parallels of style and substance at work. Stylistically ... ("The Congress shall have power") ...
[Part I A. Case 2.] Martin v. Hunter's Lessee [Story 1816]
The first important clause confronting Story is the opening sentence of Article II, Section i declaring that "[t]he judicial Power of the United States, shall be vested in one supreme Court, ..." ....
The language of the article throughout is manifestly designed to be mandatory upon the legislature ... shall be vested (not may be vested) in one supreme court, ... shall hold their offices during good behaviour, and shall ... receive ... compensation ...."
The same expression, "shall be vested," occurs in other parts of the constitution, in defining the powers of the other co-ordinate branches of the government. ... "all legislative powers herein granted shall be vested in a congress of the United States." ... "the executive power shall be vested in a president ..."
The key intratextual claim here is that the same phrase (in this case, "shall be vested") should be similarly construed in each clause in which it appears .... [emphasis added]
[Therefore shall extend in "The judicial Power shall extend to all Cases, in Law and Equity, arising under this Constitution, the Laws of the United States ... the supreme court shall have original jurisdiction." ... "in all the other cases before mentioned the supreme court shall have appellate jurisdiction ...."] The very exception here shows that the Framers of the constitution used the words in an imperative sense.
... judicial power shall extend to all cases; but in the subsequent part of the clause which embraces all the other cases of national cognizance, and forms the second class, the word "all" is dropped .... to leave it to congress to qualify the jurisdiction, original or appellate, in such manner as public policy might dictate.
[Story] is emphasizing not merely the presence of the word "all" in some parts of the menu, or the absence of the word "all" in other parts of the menu ... but the strikingly selective use of the word.
And so it seems clear that Marshall was not alone in prominently deploying intratextualism.
[Part I A. Case 3.] Marbury v. Madison [Marshall 1803]
As a matter of simple logic, however, there are two obvious possible alternatives to Marbury's reading of the Original Jurisdiction Clause as a constitutional maximum.
First, the language in Article III could be read to prescribe only a minimum, not a maximum, amount of original jurisdiction.
Second, the language might have been designed to establish neither a minimum nor a maximum but simply a starting point - a default rule - from which Congress may depart by statute in either direction.
Marshall does not confront these other possible readings; he simply assumes that the clause must be a maximum or nothing and then chants over and over that it cannot mean nothing.
The Appellate Jurisdiction Clause explicitly authorizes Congress to subtract from the Supreme Court's appellate docket; but the Original Jurisdiction Clause contains no comparable language authorizing Congress to add to the Court's original jurisdiction docket.
The point here is not a standard textual point about the Original Jurisdiction Clause, but a Story-like intratextual point that emphasizes the variation in language between this clause and the next one. [intertextualism]
There are many arguments for judicial review, but perhaps the most elegant and forceful is the simple two-pronged notion that the Constitution is supreme law, and that judges must apply this law in cases within their jurisdiction.
Even though one clause appears in Article III and the other in far-away Article VI, intratextual analysis suggests that they were indeed designed to be read together, as interlocking parts of a coherent whole, two prongs meant to affirm one result: judicial review.
Article III vests federal courts with jurisdiction over all cases arising under "this Constitution, the Laws of the United States, and Treaties made, or which shall be made, under their Authority;" and Article VI proclaims as supreme law "[t]his Constitution, and the Laws of the United States ... and all Treaties made, or which shall be made, under the Authority of the United States."
[Part I A. Case 4.] Brown v. Board of Education and Bolling v. Sharpe [Warren 1954 for both]
As the Bolling Court admits, "The Fifth Amendment, which is applicable in the District of Columbia, does not contain an Equal Protection Clause as does the Fourteenth Amendment which applies only to the states."
"In view of our decision that the Constitution prohibits the states from maintaining racially segregated public schools, it would be unthinkable that the same Constitution would impose a lesser duty on the Federal Government." But where exactly does the Constitution say that?
Harlan ... "the Constitution of the United States, in its present form, forbids, so far as civil and political rights are concerned, discrimination by the General Government, or by the States, against any citizen because of his race." [Gibson v. Mississippi, 162 U.S. 565, 591 (1896)] [emphasis added]
Harlan also said that even in cases involving civil rights - like access to government-run schools - neither state nor federal officials can discriminate "against any citizen because of his race." [emphasis added]
... Fourteenth Amendment: "All persons born or naturalized in the United States and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside." Here is a clause that plainly limits the feds as well as the states. [emphasis added]
All are declared citizens, and thus all are equal citizens. As Harlan himself put the point in the sentence immediately following the one that Bolling quotes: "All citizens are equal before the law."
Standing alone, the Citizenship Clause gives us a clean textual argument that is enough to carry the day in Bolling. But nothing in this argument is particularly intratextual.
In response, a critic could say that Warren's basic intuition - that it is "unthinkable" that the "same Constitution" would regulate the state governments and the federal government differently - is way off base. Under the Founders' Constitution, the rules were emphatically different for the two governments. (examples: obligation of contract; self-incrimination)
It is precisely at this point that an intratextual analysis rides to Warren's rescue. But there are indeed a handful (and only a handful) of intratextually paired clauses that - when read together,as their similarwording invites - implicitly do so.
No Bill of Attainder or ex post facto Law shall be passed [by Congress] .... No Title of Nobility shall be granted by the United States .... No State shall ... pass any Bill of Attainder, [or] ex post facto Law ...or grant any Title of Nobility.
In Bolling's phraseology, as to these three things at least, it is indeed "unthinkable that the same Constitution would impose" different duties on the two sets of governments. Remarkably, the result in Bolling could comfortably be defended by reference to two of the three items in this handful of clauses.
... if no government can name some (light-skinned) Americans "lords" and other (dark-skinned) Americans "commoners," surely it cannot do the same thing through racially segregated schools whose purpose and effect and social meaning is to create a blood-based and hereditary overclass and underclass.
... the legislature may not target human beings for disfavored treatment because of who they are as opposed to what they do. And there is a special historical link between attainders and "corruption of the blood" in which legislatures tried to taint or stain a person's bloodline.
If we take the nonattainder principle seriously, it bars lawmakers - federal no less than state - from passing laws designed to humiliate or demean all persons descended from slaves, or all persons with black (corrupt) blood.
Note that intratextualism itself does not compel a broad reading of the Title of Nobility and Attainder Clauses.
As we shall see in Parts II and III, intratextualism often merely provides an interpretive lead or clue, the full meaning of which will only become apparent when other interpretive tools are also brought to bear on the problem.
... it remains to consider more carefully the most obvious intratextual issue in Bolling itself: the triadic relationship between the Due Process Clauses of the Fifth and Fourteenth Amendments and the Equal Protection Clause of the latter Amendment.
But there is an evident embarrassment in this argument, and it is an intratextual embarrassment: "due process" in the Fifth Amendment should presumptively mean the same thing as "due process" in the Fourteenth Amendment, and these latter words are apparently contra-distinguished from the equality idea, which appears in a different Fourteenth Amendment clause.
The framers of the Fourteenth Amendment believed that due process of law meant a suitably general evenhanded law. "Law" in its nature was general, equal, impartial, and the "process" that generated law must respect that nature.
Thus, for the framers and ratifiers of the Fourteenth Amendment, the words of its Equal Protection Clause were not expressing a different idea than the words of the Due Process Clause but were elaborating the same idea: the Equal Protection Clause was in part a clarifying gloss on the due process idea.
And here is the intratextual kicker: if equal protection really was implicit in the Fourteenth Amendment concept of "due process of law," as its framers believed and said, then after the ratification of this Amendment, equal protection should also be seen as implicit in the Fifth Amendment phrase "due process of law."
[Harlan] proclaims that the "guarantees of life, liberty and property are for all persons within the jurisdiction ...without discrimination against any because of their race. Those guarantees ...must be enforced in the courts, both of the Nation and of the State, without reference to considerations based upon race." [Gibson v. Mississippi, 162 U.S. 565,59i (1896).]
In McCulloch, Marshall suggests that we read "necessary" differently from "absolutely necessary," yet here, I suggest that we read "due process" no differently from "due process plus equal protection." How can both be right? My answer: Where a particular clause is best read as declaratory, glossing earlier words and making explicit what these words only implied, the presence or absence of this gloss should generally make no difference.
[Part I A. Case 5.] Roe v. Wade [Blackmun 1973]
In one noteworthy but often unnoticed passage in his opinion, Justice Blackmun lavishes attention on the document itself, invoking a cluster of constitutional clauses in a single paragraph. Even more remarkably, the passage is not simply a standard example of clause-bound textualism, but a classically intratextualist exegesis.
The question that prompts this exegesis is simple: is a fetus a "person" within the meaning of the Fourteenth Amendment?
[Blackmun:] The Constitution does not define "person" in so many words. Section Iof the Fourteenth Amendment contains three references to "person." .... The word also appears both in the Due Process Clause and in the Equal Protection Clause. ... in ... listing of qualifications for Representatives and Senators; in the Apportionment Clause; in the Migration and Importation provision; in the Emolument Clause; in the Electors provisions; in ... qualifications for the office of President; in the Extradition provisions; the superseded Fugitive Slave Clause; in the Fifth, Twelfth, and Twenty-second Amendments, the Fourteenth Amendment.
[Blackmun:] But in nearly all these instances, the use of the word is such that it has application only postnatally. None indicates, with any assurance, that it has any possible pre-natal application.
[Blackmun:] ... persuades us that the word "person," as used in the Fourteenth Amendment, does not include the unborn.
He places particular emphasis on the intratextual link between the Fourteenth Amendment and the Census Clause, noting in a footnote that "[w]e are not aware that in the taking of any census under this clause, a fetus has ever been counted."
A Roe critic, however, might challenge this approach. Sometimes the same word does sensibly mean different things in different contexts, and nowhere is this more plausible than with a chameleon word like "person."
What is needed in Roe, but what is arguably missing, is an implicit or explicit argument that the context of, say, the Census Clause is sufficiently similar to that of the abortion question at hand.
[Marshall in McCulloch] sought to prove only that "necessary" could mean useful, not that it must mean that.
Blackmun by contrast is seeking to convince us that "person" must mean a postnatal human.
As a precedent-follower, Roe simply stringcites a series of privacy cases involving marriage, procreation, contraception, bedroom reading, education, and other assorted topics, and then abruptly announces with no doctrinal analysis that this privacy right "is broad enough to encompass" abortion. Ipse dixit.
In the end, Roe's many methodological lapses are simply evidence that the best theory supporting its result - a theory emphasizing gender and the particular ways in which abortion laws burden the liberty and equality of women -- was not easily accessible to the Court in 1973.
Justice Blackmun's use of intratextual argument is, methodologically speaking, one of Roe's more impressive moments. Though hurried and unreflective, Roe's exuberant use of intratextualism is considerably more clever than its rather clumsy use of other types of constitutional argument elsewhere.
[Part I B. Commentaries]
[Part I B. 1. Ely]
Ely's proposed solution to this puzzle is to read the broad clauses in light of the general themes of the Constitution as a whole. But if his general methodological prescription is not quite intratextualism, as I have defined it, it is rather close.
Ely makes at least six intratextual moves.
First, he places the similarly worded Due Process Clauses of the Fifth and Fourteenth Amendments side by side for careful analysis.
[Second], he analyzes the Fourteenth Amendment clause protecting "privileges" and "immunities" of "citizens" alongside the Article IV clause containing the same three clustered words, noting both the linguistic similarities and the important syntactical differences between these two nonadjacent clauses.
[Third,] leads him to contrast the substantive rights language of the Fourteenth Amendment's Privileges or Immunities Clause with the equal rights language of its Equal Protection Clause.
[Fourth,] Ely's intratextual comparison of the use of the word "citizens" in one Fourteenth Amendment clause with the use of the different word "person" in neighboring Fourteenth Amendment clauses.
[Fifth,] Ely juxtaposes the language of the Fifteenth Amendment - which explicitly limits "the United States" as well as states - with the omission of similar language in the Fourteenth Amendment.
[Sixth,] Ely reads the language of the Ninth Amendment in light of the language of the Tenth.
[Part I B. 2. Langdell and Lowell]
... what about legal scholarship in its early years? Consider a pair of articles that ap- peared in the pages of this Law Review exactly one hundred years ago, in volumes 12 and 13. Their authors were eminent figures: Christopher Columbus Langdell, and A. Lawrence Lowell ....
Langdell asserts that to properly assess "the status of our new territories," we must "ascertain the meaning of the term 'United States."'
What follows over the next six pages is an exhaustive - and exhausting - intratextualist extravaganza in which Langdell tries to identify every single constitutional clause (more than fifty in all) using the phrase "United States." Over the course of these six pages, Langdell makes several fine- grained intratextualist claims.
In his words, "when the same phrase is used in different parts of the Constitution, a strong presumption arises that it is always used in the same sense.' This is intratextualism with a vengeance.
More generally, the words "United States" may well be an un- promising prospect for intratextualism - or at least for an intratextu- alism that seeks a definitive answer to the territories question.
Rather, we should examine the document as a whole, and its animating ideals. And when we do, the idea that the Constitution would not apply at all to territories on the road to statehood - such as Michigan in 1820 and Arizona in 1899 - seems outlandish.
But even in the course of rejecting Langdell's outlandish thesis, Lowell himself repeatedly makes intratextualist arguments.
Lowell's use of intratextualism is considerably more adroit than Langdell's. For starters, the clauses that Lowell chooses to yoke to- gether seem to share more context, making his clausal comparisons seem less contrived. More generally, Lowell seems to recognize the limits of intratextual analysis as well as its strengths.
In the end, the Langdell-Lowell exchange evidences the promi- nence of intratextualism at the dawn of constitutional scholarship in places like the Harvard Law School. The exchange also reminds us that, as is true of all other legitimate techniques of constitutional inter- pretation, intratextualism can often be used on both sides of contested questions, and with varying degrees of deftness.
[Part II Theory]
[Part II A. The Distinctiveness of Intratextualism]
Intratextualism simply reads the Constitution as if these implicit linking clauses existed. Clause-bound textualism reads the words of the Constitution in order, tracking the sequence of clauses as they appear in the document itself.
[Constrats intratextualism with clause-bound textualism, history and intent methods, structuralism, doctrinalism, ...]
Like doctrinal argument, it seeks to promote a cer- tain coherence in interpretation and avoid the appearance of ad hoc adjudication; absent a good reason for doing otherwise, similar constitutional commands should be treated similarly for reasons analogous to the doctrinal principle that like cases should be treated alike.
Indeed, instead of viewing intratextualism as one distinct form of argument apart from six others, it may be useful to consider intratextualism as a cluster of at least three different kinds of constitutional claims.
[Part II B. The Types of Intratextualism.]
[Part II B. 1. Using the Constitution as a Dictionary: Intratextualism as Philology]
If Marshall is seeking to establish only that "necessary" can mean convenient, a single constitutional example suffices to prove his point, and powerfully.
By contrast, if Blackmun is seeking to prove that "person" must mean post-natal humans, or if Langdell is seeking to establish that "United States" can only mean the several states, even a slew of examples from the Constitution may prove unavailing.
When we seek to prove that a word could mean X, a single example from the Constitution illustrating this is stronger than an entry from a standard dictionary because the example proves that the authors of the Constitution itself - and not simply some "approved authors" some- where - understood usage X.
But if we try to prove that a word cannot mean Y, examples drawn from the Constitution are weaker than entries in standard dictionaries.
[Part II B. 2. Using the Constitution as a Concordance: Intratextualism as Pattern Recognition]
Intratextualism allows the Constitution to function not merely as a special kind of dictionary, but also as a special kind of concordance, enabling and encouraging us to place nonadjoining clauses alongside each other for analysis because they use the same (or very similar) words and phrases.
[Part II B. 3. Using the Constitution as a Rulebook: Intratextualism as Principle-Interpolation]
A final species of intratextualism demands that two (or more) similarly phrased constitutional commands be read in pari materia.
Here we are dealing not merely with a recurring word, or even a recurring word-cluster, but with a complete, carefully elaborated command that appears in identical language with a single variation that (presumptively) should make no legal or moral difference: "The [fill in the blank] power shall be vested ..." and "The right of citizens of the United States ... to vote ... shall not be denied or abridged by the United States or by any State on account of [fill in the blank]."
... we read the commands as if a metacommand clause existed telling us to con- strue parallel commands in parallel fashion. ... this form of interpolation must remain open to the possibility that, upon reflection, there are sound constitutional reasons not to treat the individual commands as in pari materia.
[Part II C. Some Strengths of Intratextualism]
Perhaps the greatest virtue of intratextualism is this: it takes seriously the document as a whole rather than as a jumbled grab bag of assorted clauses. To modify Marshall, it is a (single, coherent) Constitution we are expounding. [emphasis added]
In brief, clause-bound textualism tries to do justice to the Constitution's writtenness but not its wholeness; various other forms of argument try to do justice to wholeness but not writtenness; whereas intratextualism tries to do justice to both, and at the same time.
Writing fifteen years later, Ely raised questions about Douglas's substantive result in Griswold, but embraced - indeed extended - Douglas's methodology, offering a textual "tour" not simply of the Bill of Rights but of the entire Constitution. Democracy and Distrust at 88-101, 221 n.4.
When extended beyond paragraphism to encompass the entire document, holistic textualism has an obvious virtue: it invites readers to ponder connections between noncontiguous clauses that have no textual overlap, yet nevertheless cross-illuminate. But truly holistic textualism also has an obvious weakness: there are so many clauses to consider and an almost infinite number of interclausal comparisons that could be performed. [emphasis added]
Intratextualism has equal and opposite virtues and vices. It tells us when reading clause X to pay particular attention to similarly worded but nonadjoining clauses Y and Z. This focus narrows the field of view, but also gets us going and gives us direction. [emphasis added]
... intratextual tool alerts us to possible self-dealing in judicial interpretation, as the Court pumps up clauses that it favors or that empower it while deflating clauses that it disfavors or that empower others. [emphasis added]
[Part II D. Some Weaknesses of Intratextualism]
Carried to extremes, intratextualism may lead to readings that are too clever by half .... [emphasis added]
Given that sensible use of intratextualism will require us to consider the limits of the technique, the technique will not so much dictate results as suggest possible readings.
It highlights the document's intratextual links, but casts no light on its possibly illuminating intertextual links to other documents, such as the English Bill of Rights, state constitutions, the Declaration of Independence, and the Articles of Confederation. [emphasis added]
Another possible weakness of intratextualism is that it invites strong inferences about constitutional meaning from the document's grammar and syntax. For example, interpolation-style intratextualism presumes that two clausal commands should receive identical treatment because they feature the same basic grammar and syntax. But ... if their underlying problems have evolved in different ways, something must give. [emphasis added]
If changed circumstances sometimes call for disregarding an intratextual linkage that no longer makes sense, the same may be true of ordinary clause-bound textualism.
In the end, these cautions remind us that no tool of interpretation is a magic bullet.
[Part III Cases Again]
[Part III A. Morrison and Starr]