Home -> Law Blog Directory -> Academic Blogs -> Legal Theory Blog

Academic

: Legal Theory Blog

Primus on Originalism (and a comment on the interpretation-construction distinction)

By Lawrence Solum

Richard A. Primus's When Should Originalism Matter? is now available in final form on the Michigan Law Review website.  Here is the abstract:

Constitutional theory lacks an account of when each of the familiar sources of authority?text, original meaning, precedent, and so on?should be given weight. The dominant tendency is to regard all sources as potentially applicable in every case. In contrast, this Article proposes that each source of authority is pertinent in some categories of cases but not in others, much as a physical tool is appropriate for some but not all kinds of household tasks. The Article then applies this approach to identify the categories of cases in which original meaning is, or is not, a valid factor in constitutional decisionmaking.

Highly recommended!  Primus's article contains a very interesting discussion of the distinction between constitutional interpretation and constitutional construction--and I have some comments. 

Frequent readers of LTB are undoubtedly already familiar with the interpretation-construction distinction.  Roughly, "interpretation" is the activity of ascertaining the semantic content (or linguistic meaning) of a text.  "Construction" is the activity of translating the semantic content of the text into legal content--e.g., the content of doctrines of constitutional law.  For more on the distinction, see the following:

Legal Theory Lexicon 063: Interpretation and Construction

Lawrence B. Solum, Semantic Originalism

Randy E. Barnett, Restoring the Lost Constitution: The Presumptions of Liberty (Princeton University Press 2003).

E. Allan Farnsworth, Contracts (4th ed., Aspen 2004).

Keith E. Whittington, Constitutional Interpretation: Textual Meaning, Original Intent, and Judicial Review (New ed. University Press of Kansas 2001).

Keith E. Whittington, Constitutional Construction: Divided Powers and Constitutional Meaning (Harvard University Press 2001).

One of the really nifty features of Primus's article is its engagement with interpretation-distinction.  I'll be quoting Primus (in ordinary black typeface) and then commenting (using bolded blue).

Here is an extended passage from Primus, which is focused on the role of the distinction in the work of Keith Whittington--whose work focused attention on this old idea:

For Whittington, . . . the claim that constitutional interpretation must be originalist is not a claim about constitutions. It is a claim about interpretation. To deny that interpretation is a matter of finding original meanings, Whittington says, is simply to misunderstand what interpretation is.

The question, then, is why it matters whether we call the activity of constitutional decisionmakers ?interpretation.? There is some temptation here to write the question off as purely nominal. ?Interpretation? is not a category, let alone a term, that exists in nature. If the point were merely about the definition of ?interpretation,? it would be an adequate response to say that we could all agree to call nonoriginalist decisionmaking ?shmerpretation? and then not to care whether judges do ?interpretation? or ?shmerpretation.? Assuming that the content of the decisionmaking process is the same no matter what label we put on it, the fact that we say ?interpretation? as opposed to ?shmerpretation? will have no impact on constitutional decisions. But the temptation to answer this way should be resisted. Whittington and some others who argue about the meaning of ?interpretation? are sophisticated thinkers, and we should presume that their concern is with something more substantive than labels.

Let's pause here for a moment.  Primus asserts that "interpretation is not a category, let alone a term, that exists in nature."  It is a bit difficult to know what he means by that assertion.  The terms "nature" and "natural" are notoriously ambiguous.  Primus might be making the point that "interpretation" and "construction" describe complex rational human activities and therefore are "nonnatural" in the sense in which the products and processes that are produced by human activities are "artificial" and hence "nonnatural."  If that is his point, then he is surely correct, but this does not imply that that the interpretation-construction distinction is purely nominal.  There is a real difference between the ascertainment of semantic content (or linguistic meaning) and the construction of legal content.

What is this real difference and why is it significant?  Determining the linguistic meaning of an utterance involves a factual inquiry: the meaning of an utterance is a fact.  But the process of constructing legal rules from the linguistic meanings of text involves the making of decisions about what the law should be--it is a normative activity that necessarily involves reference to normative considerations (such as legal norms, moral principles, or considerations of political philosophy).  This difference (between the interpretation as an inquiry into facts and construction as normatively guided development of legal doctrine) is not merely nominal--it reflects an important real difference between two kinds of human activity.

One more thing: of course, the choice of the terminology is arbitrary in the following sense.  The choice of the word "interpretation" to refer to the process of finding linguistic meaning versus the word "construction" to refer to the process of determining the content of legal doctrine was arbitrary at some point.  Other terms could have been used to mark the same distinction.  But at this point, the terminology is nonarbitrary in the following sense: the interpretation-construction distinction has a long history in Anglo-American legal history (it goes at least as far back as the19th century).  That is, the use of the words "interpretation" and "construction" to mark the distinction reflects a well-established pattern of usage.

OK, let's get back to Primus:

So why might the meaning of ?interpretation? matter? One possibility is that our constitutional culture?s use of that word provides a clue about its underlying values. American lawyers regularly use the phrase ?constitutional interpretation? to refer to the activity of making decisions in constitutional law. Other expressions like ?constitutional construction? and ?constitutional decisionmaking? are also in use, but ?constitutional interpretation? is the dominant term. So if ?interpreting? means trying to capture the meaning that some authorial actor intended to communicate, perhaps the fact that we speak of ?constitutional interpretation? indicates that our constitutional culture sets a high value on original meanings.

At this point, Primus's argument reflects a serious conceptual confusion.  The point of the interpretation-construction distinction is not to divide the world of constitutional practice into two different ways of accomplishing the same task--deciding cases--and then indicating that one mode (interpretation) is normatively superior to the other (construction).  That is, interpretation and construction are not two alternatives methods which we choose between. 

A better way of characterizing the relationship would describe them as two moments or steps in the constitutional practice.  In an idealized case, constitutional actors (judges, officials, citizens) begin with interpretation--they find the meaning of the constitutional text.  If the meaning of the text is neither vague nor ambiguous (see Legal Theory Lexicon 051: Vagueness and Ambiguity), then construction is more or less automatic--the semantic content of the constitutional text is translated into a constitutional doctrine with identical content.  But if the linguistic meaning of the constitutional text is vague (in the context of the particular constitutional problem to be resolved), then construction necessarily requires that constitutional actors go beyond the meaning of the text.  Given that the semantic content of the text is vague, some supplementary rule or procedure will be required.  Constitutional law is filled with particular doctrines (no prior restraints) and procedures (case-by-case balancing tests) that facilitate the application of vague clauses to particular cases.

I have asserted that Primus's argument reflects a serious conceptual confusion, but I want to be clear that I do not believe that this is entirely (or even mostly) Primus's fault.  The interpretation-construction distinction has a long legal pedigree, but its recent revival in constitutional theory is relatively recent and some of the explication of the distinction is ambiguous.  Primus's misunderstanding seems entirely reasonable to me--and indeed, this sort of confusion is productive precisely because it can prompt clarification and more precise theoretical elaboration of the distinction.

Now, back to Primus:

Here, Whittington and others have a point. We do use the term ?constitutional interpretation,? and at least to many people, ?interpretation? implies trying to discern the meaning that some original actor communicated. But the point should not be overstated. If we use a term that seems to imply originalism, we are probably warranted in thinking carefully about whether we are in fact committed to originalism.

Let's stop here (in the middle of a paragraph).  Primus's point can now be restated in a way that avoids the conceptual confusion discussed above.  The interpretation-construction distinction does not (by itself) have any logical implications for how constitutional practice should proceed.  This is easily demonstrated by imagining a form of constitutional practice in which the linguistic meaning of the constitutional text played no role in the construction of rules of constitutional law.  Imagine, for example, a distant future in which the Constitution of the United States remains "on the books," but is considered a historical curiousity--the actual constitutional practice has morphed into a parliamentary system and the constitution-in-action is unwritten.

Originalists are committed to a theory of constitutional interpretation.  They believe that the linguistic meaning of the constitutional text is fixed at the time each provision was framed and ratified. (I call this "the fixation thesis.")  Different versions of originalism have different theories as to how fixation works.  Original meaning originalists believe that public meaning fixes the semantic content; intentionalists believe that the intentions of the ratifiers (or framers) does the work.  This component of originalism (properly understood) is a nonnormative theory--a view about linguistics or the philosophy of language.

But originalists do not limit themselves to claims about constitutional interpretation.  They also have views about the relationship of the linguistic meaning of the constitutional text and the content of constitutional doctrine.  Different originalists have different views about this matter, but their views share a common core.  Almost all originalists are committed to what I have called "the contribution thesis."  That is, originalists believe that the semantic content of the constitutional text should constraint (in some strong way) the legal content of constitutional doctrine.

Primus's point, in the passage quoted above, is correct.  It does not follow as a matter of logic, from either the interpretation-construction distinction, or from the fixation thesis, that constitutional practice should pay any attention to results of constitutional interpretation.  Some additional argument (either an argument about legal practice or an argument of political morality) is required.

Back to Primus:

That said, our use of the term does not prove such a commitment. The term might be a misnomer. Alternatively, ?constitutional interpretation? might be a locution inherited from a time when our predecessors were committed to a kind of constitutionalism that differs from the kind we have today. We are all familiar with terms that stay the same even when the world changes, such that the terms would mislead if taken literally. Think, for example, of our use of the name ?Big Ten? to refer to an association of universities that has eleven members.67

At this point, Primus is pursuing the implications of his misunderstanding of the nature of the interpretation-construction distinction.  This argument would be sound if intepretation and construction were alterantive modalities--but it is not sound as applied to the actual distinction.

Let's turn back to Primus:

Finally, the claim that ?interpretation? necessarily refers to the process of discerning an original actor?s communication might simply be overbroad: perhaps ?interpretation? sometimes bears that sense and sometimes does not, depending on the context. Soothsayers interpret dreams, and statisticians interpret scatterplots.68 Thus, even if ?interpretation? in certain contexts is about the recovery of original meanings, the fact that people speak of ?constitutional interpretation? might not indicate a commitment to original meanings in constitutional law.

Note that at this point, Primus's argument is correct, but it might be a bit misleading.  There is a sense in which existing constitutional practice and the way that it treats "interpretation" (in the sense specified by interpretation-construction distinction) does reveal a normative commitment to some form of originalism.  Our constitutional practice does seem to be committed to the proposition that the linguistic meaning of the constitutional text constrains the content of constitutional law.  For example, the Supreme Court never claims the power to announce "amending constructions" (rules of constitutional doctrine that override the text), and it is uncontroversial that constitutional amendments can override court-created doctrines (think Chisolm v. Georgia and the 11th Amendment).

And what does Primus say next?

Given that actual constitutional practice involves many nonoriginalist techniques as well as some originalist ones, perhaps the term ?constitutional interpretation? encompasses things that go beyond the search for original meanings.

Once again, Primus's is right, but his argument misses the point actually made by originalist proponents of the interpretation-construction distictnion.  No one doubts that constitutional practice includes nonoriginalist methods.  Because the constitution includes vague provisions, it is necessarily the case that constitutional practice must include constitutional construction, and construction cannot be determined by original linguistic meanings.  And of course, we could collapse the interpretation-construction distinction, and if we did, then constitutional interpretation would necessarily include nonoriginalist elements.  The whole point of the interpretation-construction distinction is to make the relationship between the role of original meaning and the role of nonoriginalist considerations clear.  Of course, we could use another vocabulary for this purpose--but a different vocabulary could not alter the fundamental point: nonoriginalist considerations cannot change the linguistic meaning of the Constitution.  For example, considerations of political morality cannot change the semantic content of the constitutional text.

Back to Primus:

Alternatively, if the term ?interpretation? carries misleading implications about our constitutional decisionmaking practices, then perhaps we should become more comfortable using other terms. In this Article, I use the language of ?constitutional decisionmaking.?69 Ultimately, however, what we call our processes of constitutional decisionmaking is not the important thing. If we want, we can reserve the name ?interpretation? for decisionmaking based on original meaning and accept that there are more ways than interpretation to make constitutional decisions. If we prefer, we can use ?interpretation? to refer to elements of decisionmaking that do not look to original meanings as well as those that do. Either way, nothing follows about the real issue, which is how constitutional questions should be decided.

Yet again, Primus is right in a sense: "nothing follows" as a matter of logic.  The interpretation-construction distinction does not establish that constitutional decisionmaking must be constrained by the meaning of the constitutional text.  But the fact that "nothing follows as a matter of logic" does not imply that nothing follows given our beliefs about the proper relationship between the linguistic meaning of the constitutional text and the legitimate or justifiable content of doctrines of constitutional law.  If you believe that constitutional actors (judges, officials, citizens) should be constrained by the text and you believe the fixation thesis (that the linguistic meaning of the text is fixed at the time of constitutional utterance), then you are committed to some version (perhaps strong, perhaps weak) of originalism.

Read Primus!

Full post as published by Legal Theory Blog on November 13, 2008 (boomark / email).