Every constitutional order must settle one question before it can settle any other: what, exactly, binds the interpreter? For the textualist, the answer is disarmingly simple. The law is what the words say. It is not what their drafters privately hoped, what present majorities would prefer, or what judges regard as just. The enacted text, and only the enacted text, carries the authority of ratification.

This position has reshaped American constitutional discourse over four decades. Justice Scalia's insistence that the text is the law moved from dissent to orthodoxy. The Supreme Court now speaks a vocabulary of ordinary meaning, canons, and semantic context that would have seemed idiosyncratic in the Warren era. Yet the same Court that professes fidelity to text routinely divides, sharply and often predictably, over what that text requires.

That divergence is not incidental. It exposes a structural question that textualism's slogans tend to obscure: how much work can constitutional language actually do? This essay first states textualism's case at full strength. It then traces the systematic ways in which text underdetermines constitutional outcomes. Finally, it asks whether a more sophisticated textualism, one candid about its limits, can preserve what was valuable in the original promise. The answer concerns more than interpretive method. It bears directly on the legitimacy of judicial review itself.

The strongest case for textualism begins with political legitimacy rather than semantics. A constitution claims authority because it was adopted through an extraordinary act of popular will: ratification, and thereafter the demanding procedures of Article V. What was ratified was a text. Purposes, expectations, and the moral convictions of the framers were never put to a vote. Only the words were. On this view, privileging anything over the enacted language substitutes an unratified source for a ratified one.

The second argument is determinacy. Legal rules serve coordination functions that depend on their being knowable in advance. Citizens, officials, and lower courts must be able to read the constitution and predict, within tolerable margins, how it will apply. A method anchored in publicly accessible meaning offers a shared reference point. A method anchored in purposes or values invites each interpreter to construct their own. Textualism promises that the same words will yield the same answers regardless of who sits on the bench.

The third argument, and perhaps the most compelling, is constraint. Alexander Bickel named the counter-majoritarian difficulty, and it is real. When unelected judges invalidate the acts of elected legislatures, someone must explain why the judges' view should prevail. Textualism supplies an explanation. The judge does not override democracy. She enforces a prior, higher democratic act against a later, ordinary one. That account works only if the judge is genuinely bound, so that the text rather than her own commitments does the deciding.

Textualists also make a subtler point about compromise. John Manning has argued, in the statutory context, that enacted language often records bargains struck among competing factions. Reading text in light of a supposed overarching purpose unravels those bargains, handing one faction a victory it could not secure through the process itself. Many constitutional provisions are compromises of exactly this kind: the Senate's equal suffrage, the Electoral College, and, at its darkest, the Fugitive Slave Clause. Their wording marks where agreement stopped.

Together, these arguments form a coherent theory of interpretive authority. The text binds because it was ratified, guides because it is public, and confers legitimacy because it disciplines those who apply it. Any serious critique of textualism must reckon with this structure rather than with caricatures of literalism. The question is not whether these values matter, because they plainly do. The question is whether text alone can deliver them.

Takeaway

Textualism's deepest claim concerns legitimacy rather than language: only the words were ratified, so only the words can carry the authority of ratification.

The difficulty is not that constitutional text lacks meaning. It is that the meaning reliably runs out before the cases do. H.L.A. Hart called this the open texture of legal language: every general term has a core of settled application and a penumbra of uncertainty. Constitutional provisions, drafted to endure across centuries, are unusually penumbral. Several distinct mechanisms produce this underdeterminacy. They are worth separating, because each calls for a different kind of supplementation.

The first mechanism is vagueness. Phrases such as unreasonable searches, excessive fines, and cruel and unusual punishments are not ambiguous between two meanings. They are evaluative standards, and applying them requires judgment the text does not supply. The second mechanism is generality. Equal protection of the laws can be read at many levels of abstraction, from a narrow ban on racial caste legislation to a sweeping principle of equal citizenship. The text is compatible with each reading and does not say which level governs.

The third mechanism is silence. The Constitution nowhere expressly grants courts the power to invalidate statutes. Marbury v. Madison inferred judicial review from structure and from the logic of a written constitution. State sovereign immunity, developed in Hans v. Louisiana and Alden v. Maine, reaches far beyond the Eleventh Amendment's terms. In Alden, the Court acknowledged that the doctrine neither derives from nor is limited by those terms. Federalism and anti-commandeering are likewise structural inferences rather than textual commands.

The fourth mechanism is disagreement among competent readers about what meaning itself is. Bostock v. Clayton County is instructive because both the majority and the principal dissent professed textualism. Justice Gorsuch read the statutory phrase because of sex through but-for causation. Justice Alito read it through the ordinary public understanding of 1964. Each argument was recognizably textualist. They differed on a prior question: which conception of meaning the method should privilege, whether literal, ordinary, or contextual. Bostock is a statutory case, but the lesson carries over to constitutional interpretation.

Bruce Ackerman presses the point further. If the New Deal and the civil rights era transformed the constitution without formal Article V amendments, then the written text may not exhaust the constitution's content. One need not accept Ackerman's theory of constitutional moments to feel its force. The text we interpret is embedded in practices, precedents, and settlements that it does not itself record. Interpreters must look beyond the words. The open questions are how they do so, and under what discipline.

Takeaway

Underdeterminacy is a feature of constitutional language, not a drafting failure. The question is never whether interpreters go beyond the text, only which extra-textual sources they invoke and whether they admit it.

None of this defeats textualism. It defeats one version of it. Naive textualism treats interpretation as a single operation, reading, whose output is the legal rule. Sophisticated textualism distinguishes two operations. Lawrence Solum's influential formulation separates interpretation, which recovers the communicative content of the text, from construction, which gives that content legal effect. Where the communicative content is determinate, construction simply implements it. Where that content runs out, construction must draw on something more.

This distinction clarifies what textualism should and should not claim. It should claim that communicative content constrains: no construction may contradict what the text actually communicates. It should not claim that communicative content is sufficient. Solum calls the remaining space the construction zone. There, vagueness, generality, and silence leave the interpreter with legally consequential choices. A textualism that denies the zone exists is not more faithful to the text. It merely hides the choices being made inside it.

Here the contrast with strict constructionism becomes decisive. Justice Scalia wrote: I am not a strict constructionist, and no one ought to be. Strict construction reads words as narrowly as possible. Textualism reads them reasonably, in light of context, structure, and established conventions. Scalia's dissent in Smith v. United States shows the difference. To use a firearm, read in context, means to use it as a weapon. Likewise, asking whether someone uses a cane is not asking whether they display one in the hall. Context belongs to meaning; it is not an escape from it.

Sophisticated textualism therefore accepts a second-order obligation: the principles that govern construction must themselves be justified. William Baude and Stephen Sachs locate those principles in the positive law of interpretation, meaning inherited rules of precedence, default, and closure. Others look to structural purposes or to the constitution's evident commitments. Rawls supplies a further criterion. Where constitutional essentials are at stake, construction should rest on grounds all reasonable citizens could be expected to endorse. In his terms, it should proceed through public reason rather than through any comprehensive doctrine.

The resulting position keeps textualism's core virtues while relocating them. Democratic authorization attaches to the text's communicative content and to the boundaries it sets. Determinacy is honestly limited to what the text actually fixes. Constraint no longer depends on pretending that words decide everything. It depends instead on requiring that every move beyond the words be stated, justified, and open to contest. The text remains the center of gravity in the system, but it no longer claims to be the whole system.

Takeaway

A mature textualism treats the text as a boundary rather than a complete instruction. It binds wherever the text speaks and demands public justification wherever the text falls silent.

Textualism's enduring achievement is to have made the enacted words the indispensable starting point of constitutional argument. Its persistent temptation is to treat that starting point as the destination. Most constitutional adjudication takes place in the gap between the two.

For constitutional design, the implication is sobering. Drafters cannot write their way out of underdeterminacy. More specific text buys determinacy at the cost of adaptability. Broader text buys endurance at the cost of greater judicial discretion. Every constitution chooses a point along that spectrum, and every choice delegates something to its interpreters.

The legitimacy of judicial review therefore cannot rest on the claim that judges merely read. It must rest on the quality of the reasons judges give when reading is not enough. Those reasons should be public and principled, and they should remain answerable to the text even where the text cannot answer for them. Words bind. They do not, by themselves, decide.