Law Review
Northwestern Law
Northwestern University Law Review : Colloquy : 2006 : Solum & Bennett

A Dialogue on Originalism Occasioned by Bennett's Electoral College Reform Ain't Easy

By Lawrence B. Solum & Robert W. Bennett

[Editor's Note:  The following piece is an informal dialogue between Professors Solum and Bennett which grew out of Bennett's prior piece on the Colloquy, Electoral College Reform Ain't Easy (click here to read the original piece).  Other readers wishing to participate in this dialogue are invited to submit a piece to the Colloquy Editor, or to leave a comment below.]


I recently posted about Bennett's very interesting piece on Legal Theory Blog: Bennett on Electoral College Reform.  My post quoted a long passage, but in this comment I want to focus on the following:

If we took the argument for constitutional protection [of an elector's right to vote contrary to instructions] seriously, we would also have to take seriously other aspects of the original conception of the presidential selection process.  For instance, neither political party designation nor the names of presidential and vice presidential "candidates" could appear on ballots, because that is a way of signaling pre-commitment of electors, rather than a process of debate and discussion that was the reason for creating the electoral college.

My post then commented:

Bennett's argument simply does not follow.  From the fact that the Constitution contemplates that electors can vote free of instruction, it simply does not follow that ballots cannot list candidate names or party affiliations.  There is a tension at the level of rationale, but such tensions are ubiquitous in the law.

Professor Bennett's reply has been posted on Legal Theory Blog (Bennett Replies on Electoral College Reform), and our exchange is cross-posted here on the Colloquy.


In a piece on electoral college reform in the inaugural issue of the Northwestern Law Review's new blog called Colloquy, I provided a short version of an argument that "faithlessness" of electors is not constitutionally protected, despite the original conception of electors as exercising discretion as they debated and decided on who would be our President.  I suggested that if one took such constitutional protection seriously, radical reconstruction of our presidential selection process would seem to be required.  I mentioned specifically that "neither political party designation nor the names of presidential and vice presidential 'candidates' could appear on the ballot, because that is a way of signaling pre-commitment of electors, rather than a process of debate and discussion that was the reason for creating the electoral college."

Professor Lawrence Solum quoted the argument in his Legal Theory Blog, and then asserted that "[f]rom the fact that the Constitution contemplates [that] the electors can vote free of instruction, it simply does not follow that ballots cannot list candidate names or party affiliations."  The reason he gives is that "[t]here is a tension at the level of rationale, but such tensions are ubiquitous in the law."  I asked if I could respond on the Legal Theory Blog, and Professor Solum graciously said that I could.  I take up that invitation, because I think that further discussion may shed some light on current debates about the approach to constitutional interpretation commonly called "originalism."

For present purposes, I'll deal with the version(s) of originalism that refer contemporary questions to original intentions or understandings (or, for that matter, meanings—it matters not which).  Put aside the difficult problem of "summing" states of minds of those responsible for some constitutional provision, and also the reach of explicit constitutional language.  One could argue that the constitutional provisions governing presidential electors imply discretion, but the language does not say so directly.  At the level of intentions (or understandings, or meanings), the originalist argument for discretion is that the constitutional "intenders" wanted electors to exercise discretion—indeed would have insisted upon it—rather than having them rotely parrot someone else's choice for president.

Once past the summing problem mentioned above, that is basically an accurate historical observation.  But I would argue that it provides no serious basis—even in originalist terms—for solving the contemporary question of elector discretion, for the contemporary question bears no real resemblance to the one those intenders thought about.  Thus if we could transport an interrogator back to the room where the intenders were deliberating and have him describe the contemporary problem to them in all its detail—including the history that led to it—in all likelihood they would say that the "discretion" they had in mind had nothing to do with the contemporary question of discretion.  The mere fact that both can be phrased in terms of "discretion" does not make them the same question.  And if the intenders insisted on discretion in the contemporary context as explained to them, they would in all likelihood simultaneously insist on reconstructing the contemporary presidential selection process to make that discretion out in the open and meaningful.  They would, in other words, insist on the types of additional moves that I suggested.

I don't know what Professor Solum means when he characterizes the tensions I mention as "at the level of rationale," and then dismisses them as "ubiquitous in the law."  I am inclined, however, to demur.  He may quarrel with my hypothetical reconstruction of what the intenders would have thought about the contemporary question, but if his quarrel is on some other ground, I would like to understand more about it.  I fully grant that tensions of this sort I discuss are, if not ubiquitous, at least quite common in the law, but that is one very good reason why, taken seriously, originalism—of the state of mind variety—will seldom yield answers to contemporary constitutional questions.  If two questions differ in important normative dimensions, then regardless of "level[s] of rationale" there is no way that one can plausibly be referred to anybody's intentions with regard to the other.


I have a few thoughts in reply to Professor Bennett's comments.  I am absolutely in agreement with Bennett that the kind of argument he made in his original piece and elaborated in his reply does have purchase if deployed against what is sometimes called "original-intentions originalism."  But I would note that Bennett's point (along with other criticisms) has led sophisticated originalists away from "original-intentions originalism" and to what is sometimes called "original-meaning originalism"—which focuses on the public meaning of the constitutional text at time of adoption.  It is what an ordinary reader (putting aside some complexities about specialized audiences) would have understood the text to say, and not the intentions or purposes of the framers or ratifiers, that is authoritative.  Or to be more specific, from the fact that a constitutional text T has purpose P and from the fact that realizing P would require result R, it does not follow that T requires R.  Rationales are not rules, and to the extent that original-intentions originalism implies that they are, then that form of originalism fails as a theory of constitutional interpretation.

For more on the distinction between original meaning and original intentions, see:

Legal Theory Lexicon 019: Originalism

Randy Barnett, An Originalism for Nonoriginalists

Jack Balkin, Abortion and Original Meaning

On a personal note, it was a thrill hearing from Dean Bennett, who holds the Nathaniel L. Nathanson Professorship at Northwestern.  Bennett is of the of the most distinguished members of our profession, and it is an honor having his reply appear on Legal Theory Blog and to participate in this version of the exchange on the Colloquy.


Professor Solum's additional "thoughts" in our exchange about originalism draw on the notion, increasingly common in originalist literature these days, that constitutional requirements are given body and nuance through original "meanings," rather than through original intentions or understandings.  As I suggested in my earlier comments, however, I fail to see how that solves the basic problems with the originalist enterprise that I discussed.  Constitutional provisions have no "meanings" apart from what people ascribe to them, as Professor Solum recognizes when he invokes the "ordinary reader."  So some person, think of him as real or as hypothetical, it matters not (and I will leave aside just how "ordinary" he is to be), will have to tease out constitutional "meaning" and then apply it to some problem of the here and now.  Finding that that "meaning" implied discretion for electors in the setting of presidential (s)elections as "ordinary readers" in the eighteenth century would have understood those (s)elections, tells us virtually nothing about how those "ordinary readers" (yes, those back then) would think about the "meaning" of those provisions for the problem of elector discretion today were they educated about today's totally different (s)election environment.

Moreover, if those "ordinary readers" found that constitutional "meaning" implied discretion today, I can't imagine that they wouldn't simultaneously find that that "meaning" implied real discretion, so that the other moves I suggested—removing presidential candidate names and political party designations from the ballots—would not similarly be implied by that constitutional "meaning."  If "meaning" doesn't mean that, it has become meaningless.

Perhaps even more fundamentally, I do not agree with Professor Solum that the recent retreat of originalists to "meaning" rather than "intentions" is a mark of "sophistication."  Rather it is an attempt to salvage something from a misbegotten enterprise, an attempt to find stability in a world that does not stand still.  Part of the appeal of originalism is that, where possible, lawmakers, not judges, are in charge of what is to be done with the laws they make.  But finding that the governing "purposes" must somehow be divorced from those lawmakers is to flee from the very soil that nurtures the enterprise.  To be sure, in other contexts in the law, the understanding or objective "meaning" of language is sometimes preferred to the author's meaning.  But that is because some consideration other than the authority to promulgate in the first place has intervened, most importantly the possibility of reliance by third parties.  If some constitutional provision did mislead people, because the "intention" behind it was not reasonably well revealed by the language used, then there would be a argument for excuse, or perhaps even instantiation of the reasonable understanding or meaning (beware, however, as that might shift over time!).  In the absence of some possibility of reliance, however, I find no "sophistication" in the most recent shifts of originalist rationales.

(None of this, by the way, is to deny the very real problems with figuring out "intentions," or "purposes" animating a provision, very much including the "summing" problem that I mentioned in an earlier phase of this exchange.)


Once again, Professor Bennett has provided some very interesting and provocative observations.  I should like to comment on what I see as the core of his response to original-meaning originalism:

I fail to see how that solves the basic problems with the originalist enterprise that I discussed.  Constitutional provisions have no "meanings" apart from what people ascribe to them, as Professor Solum recognizes when he invokes the "ordinary reader."

This is a remarkable passage.  Notice that although Professor Bennett's assertion is framed in terms of the meaning of constitutional provisions, his claim must be a larger claim in the philosophy of language.  That is, he is asserting that sentences in a natural language such as English have no meaning apart from the meaning ascribed to the sentences by persons (or perhaps by "human persons").  Of course, Professor Bennett is writing in the context of a post on Colloquy, and given the occasion, he has, quite naturally, expressed his point briefly and informally.  But at this point, many readers will be, I think, quite curious as to what general view in the philosophy of language would provide the theoretical backing for Bennett's claim.  I would like to suggest that Bennett's claim is not consistent with our best understanding of language and meaning.  One way to get at this issue is via Paul Grice's distinction between speaker's meaning and sentence meaning.  Grice's idea of speaker's meaning is actually quite familiar.  We get at the idea of speakers meaning all the time in ordinary conversations:  "What did he mean by that?"  In the context of legal texts, we ask questions like:  "What did the legislature mean by the provision?"  "What did the judge mean by that sentence in the opinion?"  "What did the Framers mean by that clause in the Constitution?"

Grice contended that speaker's meaning, in turn, can be analyzed in terms of a speaker's (or author's) intentions.  His point is illustrated by the following thought experiment:

Imagine that you have stopped at an intersection at night.  The driver of another car flashes his lights at you, and you make the inference the reason for her doing this is that she wants to cause you to believe that you lights are not on.  And based on this inference, you now do, in fact, realize that your lights are not on.

In this example, the meaning of the flashing lights is the product of the following complex intention, as explicated by Richard Grandy and Richard Warner:

  1. (1) The driver flashes his lights intending
  2. (2) that you believe that your lights are not on;
  3. (3) that you recognize her intention (1);
  4. (4) that this recognition be part of your reason for believing that your lights are not on.[1]

In the case of imperatives, the intention is that the audience (or reader) perform a certain act on the basis of the reader's recognition of the author's intention that the reader perform the act.

What about sentence meaning?  In its simplest (and perhaps simplified) form, the idea is that words and expressions have standard meanings—the meaning that is conventional given relevant linguistic practices.  As my colleague, Dean Heidi Hurd, puts it:  "In other words, the sentence meaning of a particular utterance can be understood not by reference to the illocutionary intentions of the speaker, but rather by reference to the illocutionary intentions that speakers in general have when employing such an utterance."[2]

There is an obvious parallelism between Grice's discussion of speakers meaning and sentence meaning and contemporary debates in constitutional theory.  In the case of a constitution, speaker's meaning (or author's meaning) can be redescribed as "framer's meaning."  Framer's meaning depends on what the framer's intended, given what they knew about contemporary ratifiers and interpreter's knowledge of their intentions.  Likewise, "sentence meaning" can be redescribed as "clause meaning."  Clause meaning is the meaning that would be assigned to a clause, on the assumption that the clause was written with the knowledge that it would be ratified and interpreted by readers who would have very limited access to information about the framing and who would be under normative pressure to disregard any information wasn't universally accessible.

The meaning of the Constitution is best understood as the clause meaning of its provisions.  But this does not entail that history and evidence about original meaning is not relevant to the process of constitutional interpretation.  Clause meaning is not ahistorical or acontextual.  Linguistic conventions change over time.  Words and phrases that once had one public meaning may come, over time, to acquire another.  Of course, in the case of the Constitution, the Constitution itself acts as a check on this process.  That's because the Constitution is itself public, widely available, and central to our legal culture.  Constitutional usages are likely to be preserved, simply because they are repeated, studied, quoted, and interpreted.  But in those cases in which the original public meaning of the Constitution has been swept away by a shift in the linguistic winds, the clause meaning is the "sentence meaning" that would have been assigned at the time the constitution was ratified and not the sentence meaning that we would assign based on contemporary linguistic practices.

With the distinction between framer's meaning and clause meaning in place, we are in a position to evaluate Bennett's assertion that "Constitutional provisions have no "meanings" apart from what people ascribe to them."  How does this claim relate to the idea of  clause meaning?  Clause meaning does assume that there are speakers of the natural language English.  It further assumes that these speakers use words and phrases in standard ways, and that the conventions that are constituted by patterns of usage are known to ordinary speakers of the language.  This knowledge and those conventions mean that I can write a sentence or paragraph and convey meaning to an audience that doesn't know who I am or in what context the sentence was written.  But this fact does not entail the further conclusion that particular sentences have no meaning other than the meanings that particular readers assign to them on particular occasionsQuite the contrary, Grice's work on sentence meaning establishes that legal provisions, such as clauses of a constitution, can have meanings that are independent of particular speakers, particular readers, and particular occasions.  It is precisely because clause meanings are independent of original intentions and original anticipated applications, that original-meaning originalism offers a distinctive theory of constitutional interpretation.


I think we're in danger of getting sidetracked by an interesting, but irrelevant, excursion into the philosophy of language.  I don't doubt that in particular linguistic, cultural and historical settings language can have meaning that is independent of particular people who utter the language or who constitute its audience.  I was, however, making two arguments that hold regardless of the reality of such "sentence meaning."  The first was a normative argument that if forced to a choice originalists should generally prefer speaker's meaning for constitutional provisions over any more objectified "sentence" meaning, at least in the absence of some substantial reliance on that sentence meaning.  Thus even if a possible meaning of the phrase "domestic violence" in Article IV (at the time and place of the making of the Constitution) was battering of women and children by men in the house, originalists should prefer a meaning of civil conflict out of doors, upon discovering that that is what the lawmaker had in mind.  The simple reason is that authors are the ones who were making law, and no reason appears for detaching their words from what they were trying to accomplish in exercising the authority that originalists (and all the rest of us) insist upon.  The second was basically an argument that no plausible ascription of meaning relieves an interpreter of asking questions of application to problems that arise.  And any sensible interpreter will shun "meanings" that somehow insist on applications that are (normative) worlds apart from applications that would have come to mind at the time and place when the language was initially deployed.  Thus it is hardly surprising that we have an "absurd results" exception to the most literalist of approaches to interpretation of authoritative language in the law.


Unlike Professor Bennett, I believe that theories of constitutional meaning deeply implicate the philosophy of language—or to put it more prosaically, interpretation and meaning, like love and marriage, go together like a horse and carriage.  How can we answer the question, "What does the constitution mean?," without a theory of meaning?  Such a theory provides the criteria by which the rightness or wrongness of answers to questions of constitutional meaning can be judged.  And even when these questions are answered without an explicit theory, it will always be the case that some implicit theory will be doing work behind the scenes.  And once the implicit theory is made explicit, it may turn out to be wrong, confused, or even self-contradictory.

Professor Bennett then argues that originalists should prefer speaker's meaning (framer's meaning) to sentence meaning (clause meaning).  But I find this position to be odd—given Professor Bennett's own devastating critique of original-intentions originalism.  And Bennett is not the only critic of intentionalism:  others include Paul Brest (in his famous "The Misconceived Quest for the Original Understanding"), Ronald Dworkin, and even Antonin Scalia.  Fundamentally, intentionalism fails because a given clause will reflect many different intentions, at different levels of generality, reflecting the many different framers and ratifiers.  There is nothing that would qualify as "the intention" of any given constitutional provisions; there are many intentions, and in a variety of situations, these intentions will support inconsistent results.

I agree with Professor Bennett that determining the meaning of the text does not always determine application.  In this regard, the distinction between "interpretation" and "construction"—emphasized by Keith Whittington and Randy Barnett—is especially helpful.  Ambiguity (the problem of bi- or multivalent meanings) can usually be resolved by reference to clause meaning, the original public meaning of the Constitution.  But vagueness (the phenomena of borderline cases) requires construction—vagueness-resolving glosses on constitutional meaning.  I also agree with Professor Bennett that law can and should be supplemented by a practice of equity—in the Aristotelian sense of the exercise of the virtue of practical wisdom (phronesis) guided by an internalization of the values of the law and deeply held, widely shared social norms (the nomoi).

But this agreement with some of Professor Bennett's premises does not lead me to his conclusion that "any sensible interpreter will shun 'meanings' that somehow insist on applications that are (normative) worlds apart from applications that would have come to mind at the time and place when the language was initially deployed."  Actually, I must confess that I don't fully understand what this means.  If he simply means that we should avoid absurd results, then I concur.  But if Professor Bennett means that judges should override the constitution when they conclude—on the basis of their own first-order, private judgments about political morality—that the rule should be other than that which the Constitution requires, then I disagree.  Indeed, a primary purpose of law is settlement—to provide the rule of law virtues of certainty, stability, and predictability—by transforming first-order disagreements about what ought to be done into legal disputes governed by rules that are laid out in advance.  Perhaps then, the disagreement between Professor Bennett and me is ultimately about instrumentalism and formalism.  On that question, I come down firmly on the side of the formalists—as is increasingly the case in contemporary normative legal theory.


As both Professor Solum and I recognize, the "summing" difficulties with authorial intention for the Constitution are substantial.  I do think there are ways to deal with those difficulties, but I won't get into that large subject at the present time.  (Inevitably part of the real-world answer is judicial "judgment" or choice that so exercises many originalists.)  I would only comment here that recourse to "sentence meaning" provides only the appearance of wrestling those problems down.  An objective observer trying to tease out sentence meaning in a given setting would presumably be familiar with the way real speakers use language, and he would construct his "sentence meaning" against that background.  If real speakers use similar sentences with, in Professor Solum's words, "many different intentions—at different levels of generality," then our objective constructor of meaning will have to wrestle with the self-same problems of "summing" that would be involved in a search for "authorial meaning."


One of the real differences that has emerged from this exchange is about method and the relationship between constitutional theory and other disciplines, particularly the philosophy of language.  Professor Bennett has been operating on the premise that constitutional theorists can safely set general theories of meaning to the side; my approach emphasizes the need for our theories of constitutional meaning to cohere with the best theory of meaning.  Professor Bennett's most recent response illustrates the need to square views about constitutional interpretation with the philosophy of language.  Professor Bennett suggests that Gricean sentence meaning cannot get off the ground, because, "If real speakers use similar sentences with . . . 'many different intentions—at different levels of generality,' then our objective constructor of meaning will have to wrestle with the self-same problems of 'summing' that would be involved in a search for 'authorial meaning.'"

There are several problems with Professor Bennett's argument.  First, if this argument were correct, then it would be impossible for a natural language like English to have conventional meanings.  The implications of this consequence are enormous—Professor Bennett would have demonstrated that communication through language is simply not possible.  One lesson to be learned from this point is that it is always important to check one's views of constitutional meaning by generalizing them as views about what language means in ordinary non-legal contexts.

Second, as a technical matter, Professor Bennett's argument is not sound.  The summing problem to which Professor Bennett refers is a problem for an intentionalist theory of meaning (and hence, for original-intentions originalism).  That theory equates the meaning of a sentence (or clause) with the intention of the speaker (or framer).  Given multiple speakers with inconsistent intentions, the theory fails to assign a meaning to group utterances that lack unanimity of intention.  But this problem does not affect Grice's theory of sentence meaning.  In fact, problems like this are what motivated Grice to develop his theory.  Gricean sentence meaning derives from conventions about meaning.  For example, there is a convention that the phrase, "Constitution of the United States" refers to the Constitution that is officially reported in the United States Code and that is enforced by American courts.  Of course, a given speaker could use that phrase for a different purpose, and if the audience were aware of his intention, then the speaker's meaning of the phrase would differ from the convention.  Take a simple and perhaps silly example.  I say, "By 'Constitution of the United States,' I mean to refer to Lewis Carroll's poem, 'Jabberwocky.'"  I know that the readers of this post know that is my intention.  I then say, "The first line of the Constitution of the United States is, "Twas brillig and slithy toves did gyre and gimble in the wabe."  My meaning is clear.  What I meant is that the first line of "Jabberwocky" is the line with brillig and slithy toves.  That meaning is what Grice calls "speaker's meaning."  The important point is this:  the derivation of the conventional meaning of the phrase, "The Constitution of the United States" does not require a summation of inconsistent intentions.  Instead, it requires identification of conventional usages.  In other words, Professor Bennett's argument against sentence meaning (or clause meaning) and original-meaning originalism simply fails.  It is based on a serious misunderstanding of Grice's views in the philosophy of language.

There is a larger point to be made about this exchange.  The formulators of early versions of originalism made a mistake that is very much like the mistake that Professor Bennett has made in this exchange.  They formulated a theory of constitutional meaning that could not work, because it was inconsistent with fairly elementary truths about meaning in general:  the early originalists made serious philosophical errors.  Original-meaning originalism—which focuses on the public meaning of the text (which I have called "clause meaning")—avoids those mistakes.  Of course, original-meaning originalism does not have the same "normative punch" that early originalists mistakenly attributed to intentionalism.  Jack Balkin has made that point clearly and forcefully in his recent work.  In my experience, many constitutional theorists haven't yet grasped that there has been a sort of Copernican revolution in constitutional theory.  Many constitutional scholars assume that "originalism" means "intentionalism," and that talk about "original public meaning" is just a minor and insignificant tweak in originalist theory.  Moreover, there is an implicit assumption that the "new originalism" is intended to serve the same political agenda as the "old originalism."  These assumptions are simplistic and, for the most part, false.  Original-meaning originalism is not a close cousin of intentionalism—it is a completely different theory.  And the political agendas of the new originalists are as diverse as the political views of Randy Barnett and Jack Balkin—hardly identical to those of Robert Bork or Raoul Berger.

But the debate over theories of constitutional meaning is, at bottom, not about politics.  It is about what the Constitution means.  Of course, we may ultimately decide that judges should be free to disregard the meaning of the Constitution and decide on the some other basis.  That is another can of worms.  But let's take one can at a time.


I am actually catching a glimmer of some common ground through the mists of this exchange between Professor Solum and me.  I gladly acknowledge that in stipulated language communities of time and place utterances will have "conventional" meanings apart from what the authors may have had in mind.  And I further agree that such conventional meanings are the grounding for a great deal of human communication.  I don't agree that departing from those conventional meanings in the case of constitutional phrases will necessarily be, in Professor Solum's words, "overrid[ing] . . . the Constitution," or ignoring something "that the Constitution requires."  Nor in fact does he, as when he acknowledges space for avoiding "absurd results," and no doubt would for adjusting for scrivener's error as well.  But much more fundamentally, this is not the subject that I thought we were discussing.

This all started because Professor Solum took issue with my assertion that presidential elector discretion today is not necessarily protected from state regulation, simply because it was understood at the time the Constitution was crafted that those electors would have discretion.  But I was careful to put aside "the reach of explicit constitutional language."  I acknowledged that the constitutional language dealing with electors might "imply discretion," and I'll return to that possibility below.  But the subject I thought we were dealing with was when recourse to something other than bare language was necessary:  "intentions (or understandings, or meanings)."  In light of the turn that this exchange has taken, I probably should not have lumped "meanings" with intentions and understandings, but it should have been clear that in the exchange I was dealing with constitutional problems where the language does not have language meaning free of lots of interpretational problems, usually depicted as ambiguity and vagueness.  This is, of course, the kind of constitutional language that gives rise to most of the interpretational problems that have exercised originalists and others—phrases like "equal protection of the laws," "commerce," and "the executive power."  ("Due process of law" is often advanced as an exception as applied to the problem of "substantive due process," but that is a large subject I will leave to the side for the moment.)

With this background it may well be that Professor Solum and I have large areas of agreement.  I don't think there was any "conventional" meaning of "equal protection of the laws" when that phrase was put in the Constitution that allows the interpreter to avoid navigating among levels of generality and specificity with no clear guidance from the language.  It was in this sense that I argued that the "summing" problem for authors' intention(s) in the case of the Constitution is present for "language meaning" enthusiasts as well.

Nor should any of this be surprising.  Now I'm just a Midwestern boy who hasn't read a word that Paul Grice wrote, but I think it is clear that "conventional" meanings of language are closely related to the way speakers use that same language.  After all, those speakers are drawing on the same conventions.  So in the case of pretty determinate language meaning (like "two" and "state" in "The Senate of the United States shall be composed of two Senators from each State") it is likely that there wouldn't be much of a summing problem if we used authors' meaning ("intention(s)"), and there wouldn't be much if we used "sentence meaning" either.  Conversely, in the case where the Constitution's framers entertained different thoughts about the import of the language they were enacting, it wouldn't be surprising to find that "language meaning" of the language they used similarly presented a problem of a range of possible meanings as applied to particular problems that arise.  This is not to say that the summing problems are identical in the two situations.  The Constitution's authors may have had some technical understandings (legal, for instance) or even eccentricities in the use of language that would not leap to mind in the search for conventional meaning.  I continue to think that in the case of such dissonance that might make a real difference, and absent substantial detrimental reliance on the conventional meaning, originalists should prefer the authorial intention, and I suppose that is a real point of difference between Professor Solum and me.  (Would Professor Solum really contend that if every single author and every single ratifier of the Constitution attached a technical legal meaning to the phrase "Letters of Marque and Reprisal" in Article I, but that it had a different "conventional meaning," the conventional meaning should be preferred by originalists?)

Still, there may be wide areas of agreement.  Professor Solum recognizes that the constitutional language will often leave substantial space where "construction" will be necessary.  To put it less politely, this is space where judicial discretion—"judgment"—will be required.  And he recognizes that this space—which has recently been emphasized in the work of modern day originalists like Keith Whittington[3] and Randy Barnett—severely compromises the basic animating purpose (doing battle with some judicial decisions) of the contemporary originalist revival.  I count this recognition a very important contribution to the ongoing debates about originalism.

Now what has all this got to do with elector discretion?  Just to get our language cards on the table, here is the guts of the governing Twelfth Amendment provision (not for present purposes materially different from the original provisions in Article II) that might be thought to bear on elector discretion:

The Electors shall meet in their respective states and vote by ballot . . . ; and they shall make distinct lists of all persons voted for . . . and of the number of votes for each, which lists they shall . . . transmit . . . to the President of the Senate;--The President of the Senate shall, in the presence of the Senate and the House of Representatives, open all the [lists] . . . and the votes shall then be counted . . . . [4]

As will often be the case in questions of constitutional interpretation, this language does not frontally address the question posed—in this case elector discretion.  But I think that the reason it doesn't is simply that it was assumed by essentially all concerned that electors would have, and would exercise, such discretion.  Further I assume that upon reading that language, informed readers of the day would likewise conclude that electors were to exercise discretion.  My simple assertion that started this exchange going was that if we were to afford constitutional protection to elector discretion today, that would also call into question other aspects of today's presidential (s)election process.  In particular, I suggested that "neither political party designation nor the names of presidential and vice presidential 'candidates' could appear on ballots because that is a way of signaling pre-commitment of electors, rather than a process of debate and discussion that was the reason for creating the office."  I remain at a loss to understand why Professor Solum disagrees with my conclusion on this point.  To be sure, I was trying to reconstruct what constitutional "authors" would have said in response to questioning about the practices I mentioned, but I think that constitutional readers of the day would have come up with the self-same answers about "language meaning" if pressed by my hypothetical interrogator.


1. Richard E. Grandy & Richard Warner, Paul Grice, on Stanford Encyclopedia of Philosophy, available at (link).

2. Heidi Hurd, Sovereignty in Silence, 99 Yale L. J. 945, 964 (1990).

3. Whittington, by the way, agrees with me that it is authorial "meaning" that should be preferred to any "language meaning".  See Keith E. Whittington, Constitutional Interpretation: Textual Meaning, Original Intent, and Judicial Review 40 (1999) (link).

4. U.S. Const. amend. XII (link).


Copyright 2006 Northwestern University

Cite As: 101 Nw. U. L. Rev. Colloquy 31 (2006),

Persistent URL:

(Download PDF)


(back to top)