Ambiguity has an odd and oddly underexamined place in law. Judges talk about it constantly, define it carefully and treat it as the trigger for major interpretive moves. Yet when it’s time to decide a case, they’re remarkably reluctant to say that the disputed text is actually ambiguous.
The stock formula is familiar: A provision is ambiguous if it’s susceptible to more than one reasonable interpretation. On paper, that sounds rigorous. In practice, there’s plenty of rhetoric about ambiguity. What’s often missing is a willingness to let it survive into the opinion’s final paragraph.
Invariably a linguistic disease in legal drafting, ambiguity should be distinguished from vagueness—which is often usefully employed for linguistic breadth (e.g., prompt, reasonable, substantial). Ambiguity concerns a failure to denote one meaning as opposed to another, very different one.
Courts routinely acknowledge that a losing argument over interpretation is plausible, substantial or “not without merit.” Yet they often stop just short of calling it reasonable enough to create ambiguity. Ambiguity is described as a threshold concept, but the threshold itself often remains hidden from view.
That reluctance has explanations, even if they don’t flatter the system. To find ambiguity often triggers consequences. In criminal cases, it may invite the rule of lenity. In contract disputes, it may open the door to extrinsic evidence or doctrines such as “contra proferentem,” the doctrine that a document is to be construed against the drafter. In statutory cases, it may require a court to acknowledge that the text alone hasn’t resolved the controversy. Once that concession is made, the opinion can’t as easily portray the result as linguistically inevitable. Judicial judgment becomes more visible. Courts understandably have incentives to avoid that position.
The doctrine’s own language makes this easier. A text is ambiguous if it admits of more than one reasonable interpretation. But reasonable is itself a vague concept. (It isn’t ambiguous, but it’s vague: Please hold up the distinction.) Everyone agrees that some readings fall outside it, and others fall within it. The difficulty is the territory in between. The doctrine almost never says how much support a competing reading needs before it becomes reasonable. Hence the standard that polices ambiguity imports its own fuzziness.
A common response is that ambiguity doesn’t arise merely because lawyers or judges disagree. Textualists often argue that ambiguity exists only when multiple interpretations remain viable after all accepted interpretive tools have been brought to bear. That point is well taken. Disagreement alone doesn’t establish ambiguity; judges and lawyers can be mistaken. But persistent disagreement among careful readers remains evidence that the text supports more than one serious reading. If multiple courts applying the same tools arrive at different conclusions, that doesn’t automatically prove ambiguity. It does suggest that the question isn’t as easy as the eventual winner may later claim.
One often sees a familiar appellate progression. The trial court declares that the text unambiguously means X. The intermediate court reverses and insists, with equal confidence, that the same text unambiguously means Y. The court of last resort eventually announces that both were mistaken because the provision unambiguously means Z. Each court invokes plain meaning. Each denies ambiguity. The highest court may well reach the best interpretation—certainly the ultimate one. But if three nonfrivolous courts can extract three incompatible certainties from the same sentence, the litigation history is at least some evidence that the language supported more than one serious reading.
Part of the difficulty is that legal discussions often use ambiguity to cover nearly every kind of interpretive uncertainty. Strictly speaking, not all these problems are ambiguities in the linguistic sense. Some concern meaning. Others concern application. Still others concern legal significance. Yet courts and lawyers routinely throw them all into the same basket. A more candid approach would separate the principal varieties before using ambiguity as a gatekeeping concept.
The 5 varieties
Linguistically speaking, there are five varieties.
The first is lexical ambiguity, which arises when a word has multiple established meanings. The famous chicken case turned on whether the word “chicken” referred only to young broilers or also included older stewing hens. The problem is the word itself: one word, two accepted meanings. Resolving the dispute requires choosing between those meanings.
The second is syntactic ambiguity, which arises when the same sequence of words can be grammatically organized in more than one way. A statute makes it a crime to “knowingly transport, receive or possess” certain items “in violation of this act.” One reading carries knowingly throughout the phrase. Another limits its reach. Every word stays exactly where it is. The uncertainty comes entirely from the sentence’s structure.
The third is referential ambiguity. Here, the meaning of the language is perfectly clear, but the language doesn’t uniquely identify what it’s talking about in the real world. The classic example is a contract requiring shipment on the vessel Peerless when two ships bear that name. The problem is figuring out which Peerless the parties had in mind. The uncertainty lies not in meaning but in reference.
The fourth is pragmatic ambiguity. Here, the uncertainty concerns what the speaker or writer is doing with the words. Consider a letter of intent stating: “The parties agree to the following terms and intend to execute a definitive agreement on or before Dec. 31.” The words and grammar are clear. The uncertainty lies elsewhere. Have the parties already bound themselves? Have they merely promised to negotiate in good faith? Have they simply announced an intention to keep talking? The dispute isn’t about what the words mean but what legal act they perform.
The fifth is contextual ambiguity. Here, even the speech act may be clear, but the provision’s role within the larger legal instrument remains uncertain. Consider a statute providing: “Notwithstanding any other provision of this act, the court may award attorney fees to the prevailing party.” The words themselves are straightforward. The question is how this clause fits into the larger statutory machinery. Is it a master key that overrides every other fee provision in the act? Or is it a modest safety valve that operates only where no other fee rule applies? The uncertainty arises not from the clause itself but from its place within the larger structure.
These categories aren’t mutually exclusive. Real cases often involve several at once. A provision can be lexically ambiguous, syntactically ambiguous and contextually ambiguous at the same time. Their value is that they force us to identify what kind of uncertainty we’re confronting before declaring a text clear or unclear.
The categories also point toward different interpretive tools. Lexical ambiguity asks what a word means; syntactic ambiguity asks how words fit together; referential ambiguity asks what the language refers to; pragmatic ambiguity asks what legal act is being performed; and contextual ambiguity asks what role a provision plays within the larger scheme.
The practical value of these distinctions isn’t taxonomic neatness. It’s intellectual honesty. Lexical ambiguity invites examination of usage, dictionaries and terms of art. Syntactic ambiguity calls for grammatical analysis and a comparison with parallel provisions. Referential ambiguity often requires evidence about surrounding circumstances. Pragmatic and contextual ambiguity frequently require attention to legal practice, institutional context and the structure of the broader instrument.
Identifying the source of uncertainty also makes it harder to obscure what’s actually driving the decision. Courts sometimes invoke the rhetoric of clarity when they’re relying heavily on contextual inferences, institutional assumptions or judgments about how a legal text should operate. There may be perfectly good reasons for doing so. The problem isn’t that such considerations enter the analysis. The problem arises when an opinion treats them as though they were simply the inevitable unfolding of unambiguous language.
None of this requires judges to celebrate ambiguity or to find it whenever disagreement exists. It requires only a measure of candor. Ambiguity isn’t a unitary phenomenon, and legal texts can be uncertain in different ways and for different reasons.
Acknowledging those differences produces a more accurate account of interpretation. The legitimacy of legal interpretation isn’t strengthened by insisting that every hard case contains only one reasonable answer. It’s strengthened by identifying the source of the difficulty, explaining why one interpretation ultimately prevails, and recognizing that reasonable alternatives sometimes existed all along. No canon of construction—official or unofficial—requires judges to pretend that the difficult choice they’ve just made was never a choice at all.
Bryan A. Garner. (Photo by Karolyne H.C. Garner)
Bryan A. Garner is the author of The Winning Brief, Garner’s Modern English Usage and Legal Writing in Plain English.
