Original Expected Applications Redux

Guest Blogger

For the Balkinization 20th Anniversary Symposium

Lawrence Solum

Jack Balkin is famous for highlighting the difference between “original public meaning” and “original expected applications.”  This conceptual distinction first came to the attention of the legal academy via “The Meaning of Original Meaning,” a 1998 article by Mark Greenberg and Harry Litman.  As Balkin wrote in 2007, “originalist practices of argument tend to conflate original meaning and original expected applications.” (Jack M. Balkin, “Original Meaning and Constitutional Redemption.”) 

The nature of the distinction between “original expected applications” on the one hand, and “original public meaning,” on the other, requires unpacking.  The word “meaning” is notoriously ambiguous, but in the phrase “original public meaning,” the relevant sense of the word “meaning” is best captured by the idea of “communicative content.”  And “content” refers to the concepts and propositions that are conveyed (communicated by) the constitutional text to its intended audience.  Another way of putting this uses the distinction between “sense” and “reference” made famous by Gottlob Frege.  The original public meaning of the constitutional text is its original sense. 

Communicative content is conceptually distinct from expected applications.  The communicative content of a text is what determines its applications.  “Expected applications” are beliefs (expectations) about what applications the communicative content of a text will produced.  “Original expected applications” are application beliefs that are formed at the time a constitutional text is framed and ratified.

Thus, the original public meaning of the Commerce Clause is a function of the content communicated by the clause to the public at the time Article One was framed and ratified.  That content is determined by the contextualized meanings of the words and phrases that make up the clause at the time it was drafted, proposed, and ratified.  That meaning is not the same thing as the expectations of the public, the framers, and the ratifiers about the application of the clause by Congress.  Those expectations may, for example, have included the belief that Congress would enact statutes regulating navigable waters, but the belief that such statutes are regulations of commerce among the several states is not the same thing as the meaning of the Commerce Clause.  Meaning (communicative content) is one thing; applications are another.  And beliefs or expectations about applications are a third and conceptually distinct thing. 

Original public meaning and original expected applications are conceptually distinct, but they are related in various ways.  The most obvious and important relationship is evidentiary.  Application beliefs formed at the time a constitutional provision is framed, ratified, and first implemented are relevant evidence of its meaning.  If the public, the framers, and the ratifiers expected that the Commerce Clause would be applied in the form of regulations of navigable waters, that is evidence in favor of an interpretation of the clause that permits such regulations and evidence against any interpretation that would render such regulations unconstitutional. 

Application beliefs are strong evidence of communicative content, but they are not conclusive evidence.  Why?  One reason is that expectation beliefs can be mistaken.  Drafting is tricky, and even the author of a legal text can make a mistake about its consequences.  Anyone who has tried to draft a complicated rule is familiar with this phenomenon.  And because the United States Constitution has endured for a very long time, it is also possible that a change in circumstances (facts) will lead to an unexpected application of a constitutional provision.  The application beliefs at the time the Second Amendment was enacted may have been limited to the types of arms that existed in 1791—perhaps, flintlock pistols, muskets, and swords.  With the invention of the revolver, a new type of “arm” came into being.  Assuming that the original public meaning of the word “arm” (in context) was something like “a weapon that can be carried by an individual,” it would follow that pistols are arms, even though no one in 1791 had an application belief about pistols.  For public meaning originalists, it is the communicative content and not the application beliefs that determines whether pistols are covered by the Second Amendment. 

The distinction between original public meaning and original expected applications continues to generate confusion and disagreement.  Why is that?  One important reason for continued disagreement about application beliefs concerns their evidential weight.  Justice Antonin Scalia bears some of the responsibility for this confusion.  Justice Scalia sometimes made arguments that moved directly from original expected applications to the legal content of constitutional doctrine without consideration of the actual meaning of the constitutional text.  Here is an example of how that can happen.  In 1791, when the Cruel and Unusual Punishment Clause of the Eighth Amendment was framed and ratified, the death penalty was a very common punishment, and no one expected that the adoption of the Eighth Amendment would result in the abolition of capital punishment.  So, one might go directly from the original expectations about application to the legal conclusion that capital punishment is not “cruel and unusual” without pausing to ask what the communicative content of “cruel and unusual punishment” actually was. 

Because expectations about application are strong evidence of original meaning, taking a shortcut (skipping the meaning step and going directly to the legal conclusion) may seem reasonable.  And to some judges (perhaps Justice Scalia) may believe that application beliefs provide decisive evidence of original meaning; if this were true, then there would be no need to determine the communicative content of “cruel and unusual punishment” in order to answer the question whether capital punishment violates the Eighth Amendment. 

But taking shortcuts is not good originalism.  When originalists take shortcuts, they can make mistakes.  One reasons for this is that the actual original meaning of the constitutional text can be surprising to contemporary readers.  The Eighth Amendment provides a good example of this phenomenon.  Most discussion of the application of the Eighth Amendment to the death penalty focuses on the word “cruel,” but the leading originalist scholar, John Stinneford has made a convincing argument that the key to understanding the text is actually “unusual,” which he shows meant something like “contrary to longstanding practice and tradition” or “contrary to long usage.”  His word on cruel suggests that the eighteenth century meaning was much like the contemporary meaning, and hence that the death penalty in 1791 might well have been considered cruel but not unusual.  Stinneford’s research suggests that the contemporary death penalty might well have become cruel when it fell out of use for an extended period of time and hence became unusual (in the original sense of that term).  Stinnerford’s two most important articles are John Stinneford, “The Original Meaning of ‘Cruel’” and “The Original Meaning of ‘Unusual.’”  If Stinneford is right about the original meaning of the Eighth Amendment, then the application beliefs of 1791 may not be a reliable guide to the correct application of the constitutional text today, because what was usual in 1791 might be unusual today.  To take, an example made famous by Justice Scalia himself, punishment by stocks or pillory (arguendo, usual punishments in 1791) would be unusual today.  Assuming they were and still are cruel, these punishments would not have violated the Eighth Amendment in 1791 but would violate it today. 

In addition to the conceptual distinction between original meaning and original expected applications, there is also an important normative difference.  Every provision in the constitutional text has been ratified, either by state conventions or by state legislatures.  The democratic legitimacy of the ratification process has varied over time, but even the 1791 provisions were ratified by a process that was accepted as legitimate at the time.  By way of contrast, expectations about application are not ratified by anyone.  There is no evidence that anyone thought that it was someone’s application beliefs and not the constitutional text that was being ratified in 1789, 1791, or 1868. 

And there is another problem with original expected applications as the source of binding constitutional law.  Application beliefs are a form of original intentions, psychological states of individuals.  As such, they suffer from the same defects that do other forms of original intentions.  The mental states of individual drafters, framers, ratifiers, and the public are likely to differ.  Many of the participants in the process likely had no application beliefs with respect to most constitutional issues.  This leads to the well-known summing problem.  There is no agreed upon method for combining the various application beliefs (or lack thereof).  So, original expected applications cannot be a binding source of constitutional law. 

Nonetheless, judges are tempted to treat application beliefs as either decisive evidence or as an independent modality of constitutional argument.  That temptation may be particularly hard to resist when there is an argument from application beliefs to a result that a judge, scholar, or lawyer favors as a matter of constitutional preference.  The temptation to use application beliefs as a shortcut will be especially strong when it would require considerable effort to discover the actual communicative content of the constitutional text.  When getting the result you want from application beliefs is easy, but determining the actual original public meaning is hard, it is no surprise that many judges succumb to temptation. 

The temptation to substitute application beliefs for original meaning is related to another development in constitutional theory.  Some critics have used original expected applications as the basis for a repackaged version of what is called the “Indeterminacy Thesis,” the claim that the legal texts cannot constraint legal practice because they are indeterminate, permitting any result in any case.  The Indeterminacy Thesis was the subject of intense debate in the 1980s.  My take on the 1980s iteration of the indeterminacy debate was presented in “On the Indeterminacy Crisis.”  One of the keys to understanding that debate is the concept of “underdeterminacy.”  The communicative content of legal texts is very rarely (almost never) completely indeterminate.  Rather, we can think of degrees of underdeterminacy. 

The new version of the Indeterminacy Thesis is based on the idea that while original expected applications can sometimes provide a relatively high degree of determinacy, the original meaning of the constitutional text cannot.  Why not?  Usually, the reason is based on the idea that the text consists of broad and abstract provisions that are vague or open textured.  This argument takes on many forms: one particularly popular version is based on Dworkin’s deployment of the concept-conception distinction.  Thus, it is argued that the Equal Protection of the Laws Clause deploys the general concept of equality, and it is up to judges to determine what the best conception of equality is.  Likewise, the Due Process of Law Clause is said to require fair procedures, and it is up to judges to determine what conception of procedural fairness is best. 

As a general objection to originalism with respect to the whole constitution, this argument clearly fails, because only a few constitutional provisions even arguably have original meanings that are captured by the idea of abstract moral concepts of which they are many competing conceptions.  Indeed, the most important structural provisions of the Constitution, e.g., those establishing a bicameral Congress, the President, and the Supreme Court and specifying the system of elections and qualifications for office, are quite determinate.  Similarly, the Recess Appointments Clause, the Birthright Citizenship Clause, and many of the enumerated powers of Congress have highly determinate original meanings.  There may be some zone of underdeterminacy for these provisions, but it is relatively small. 

More importantly, the indeterminacy objection is usually based on what might be called “armchair originalism,” speculation about original meaning based on contemporary linguistic intuitions without consideration of the actual evidence of original meaning.  Once we take the evidence into account, things look very different.  Take the Fifth Amendment Due Process of Law Clause as an example.  An armchair originalist might conclude that “due process” means fair procedures; that would be a tenable hypothesis based on twenty-first century linguistic intuitions.  But the original meaning of the clause was quite different.  As Max Crema and I show in “The Original Public Meaning of the Fifth Amendment Due Process of Law Clause,” the evidence overwhelmingly supports a very narrow interpretation: the word “process” was understood in a technical legal sense that is very close to the meaning of that word in the phrase “service of process.”  Given the actual meaning of the clause, most of contemporary Fifth Amendment due process clause doctrine is demonstrably incorrect.  Importantly, it makes no difference that original expected applications are not binding for public meaning originalism.  The relative determinacy of the clause is produced by its communicative content. 

The Equal Protection of the Laws Clause is another provision used to illustrate the claim that original public meaning is indeterminate if original expected applications are not binding.  Again, the claim is based on armchair originalism.  The word “equal” is plucked out of context and used to construct an argument that the Equal Protection of the Laws Clause is a general guarantee of equality, delegating to judges the task of determining what moral theory of equality is correct.  But once the evidence is considered, this thesis about the original public meaning of the phrase “equal protection of the laws” begins to look very dubious indeed.  In fact, the best originalist interpretation suggests that the key to the meaning of the clause is the phrase “protection of the laws,” which referred to an affirmative obligation to protect persons against violence, theft, fraud, and other invasions of their legal rights.  Christopher Green’s articles, “The Original Sense of the (Equal) Protection Clause: Pre-Enactment History” and “The Original Sense of the (Equal) Protection Clause: Subsequent Interpretation and Application” were pioneering efforts in recovering the original meaning of the clause.  The word “equal” modifies the phrase “protection of the laws,” creating an obligation for the states to protect all persons equally.  In this case, the text is not precise.  Implementing doctrines are required to distinguish equal from unequal protection.  But the relative underdeterminacy of the text with respect to the content of the implementing rules should not obscure the more fundamental point: the Equal Protection of the Laws Clause is not a general guarantee of equality.  The consequence is that the edifice of “tiers of scrutiny” cannot be sustained from an originalist perspective.  And once again, the nonbinding nature of original expected applications does not alter the conclusion.  It is the meaning of “equal protection of the laws” and not application beliefs that undermines the modern tiers-of-scrutiny approach. 

Sometimes constitutional theory makes progress.  Sometimes it chases its own tail.  Jack Balkin’s work on original expected applications is widely known, but some of its lessons have yet to be learned.

Lawrence B. Solum is William L. Matheson and Robert M. Morgenthau Distinguished Professor of Law and Douglas D. Drysdale Research Professor of Law at the University of Virginia School of Law. You can reach him by e-mail at lsolum@law.virginia.edu