The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
The Use and Misuse of Section Three's "Legislative History": Part II
[Note: This is the sixth in a series of essays responding to objections that have been made to enforcing Section Three of the Constitution. The first five essays can be found here, here, here, here, and here.]
Earlier today we began discussing the use and misuse of constitutional legislative history in interpreting Section Three of the Fourteenth Amendment. We first set out some important general principles about the proper role, and limitations, of the use of legislative history as potentially probative, second-best evidence of constitutional textual meaning. Next, we discussed at length some serious mischaracterizations of portions of the legislative history of Section Three set forth in recent draft scholarship and essays by Professor Kurt Lash. In particular, Lash misportrays statements by Representative Thaddeus Stevens and by Senator Lyman Trumbull as if they denied that Section Three of the Fourteenth Amendment is a self-executing constitutional command, when they did not. We also despaired that this mistaken account has been adopted by others. This highlights one major concern about the use of constitutional legislative history: its capacity, if not done faithfully and carefully, to mislead and distort. This should be a general caution about its use, and one that we think has proven unfortunately true in Lash's discussion of the self-execution issue.
Today we take up some problems with Lash's use of constitutional legislative history, even where accurately recounted, in the course of legal analysis of the Constitution's text. Here, our point is not that the excavation and presentation of the legislative history itself is flawed. Rather, the legal analysis and conclusions thought to flow from that legislative history are unsound. Lash in our view makes several types of missteps in this regard, and the errors tend to compound one another.
First. Lash at several junctures treats evidence of the drafters' (believed) purposes, motivations, or immediate desired political consequences as if they necessarily constitute limitations on the range of the text's original linguistic meaning. As we wrote earlier, this mistakes the proper role of constitutional legislative history: to explicate the meaning of the text's words. As we have discussed, the legislative debates culminate in the enactment of an authoritative legal text, which sometimes is narrower and sometimes is broader than the "purposes" articulated by some of its drafters. The Constitution is a binding, authoritative legal text, not a collection of specific historical intentions or political goals. Where the rule specified in the text applies, it applies. It is not limited by evidence of subjective intentions of expectations.
Lash's treatment of the legislative history of Section Three departs from these premises at a number of points. He imputes – sometimes reasonably, as an historical matter – certain purposes to certain drafters of Section Three. This is not itself wrong. But Lash then assumes that any reading of the text that might extend its application beyond these supposedly limited historical purposes must therefore be considered "unclear" or "ambiguous." On this assumption, it is very difficult to understand any clear and general statements of law that apply beyond the immediate politics of the day.
For example: "The only thing that is clear about the text of Section Three is that it accomplished the only purposes Reconstruction-era Republicans cared about," Lash writes. These purposes ostensibly were limited to a concern to "prevent leading rebels from returning to Congress" or from being presidential electors or "receiving appointment to federal or state office." "All else," Lash concludes, must be considered "historically unclear or textually ambiguous." (P. 8). At other points, Lash ascribes to Section Three a narrow, historically-specific political purpose of some proponents to affix a "brand of treason" on rebels who supported secession and implies that this might suggest that Section Three is limited in its application to the Civil War situation. (Id. at 25, 44). From this Lash infers that it shaped both how "Trumbull viewed Section Three" and how "The ratifying public understood" Section Three as "targeting thousands of still-living leaders of the recent rebellion." (Id. at 57-58). "Beyond this, little else is clear," concludes Lash. (Id. at 57-58).
At one level, it is no big surprise that draftsmen of the Fourteenth Amendment had in view, and focused their public statements on, what was happening right then and there, in the immediate political situation of Reconstruction. But it is subtly misleading to suggest that this limits the meaning of the text, when the text does not itself contain a limitation to the Civil War but applies to "insurrection or rebellion" generally and to persons who have "engaged in" it in violation of a previously sworn oath to the Constitution. For instance, it has been said that the purpose of the Takings Clause of the Fifth Amendment was to ensure compensation when the military commandeered horses or other supplies from civilians during wartime. But that hardly makes it ambiguous whether the Takings Clause applies to non-horse property, or non-military takings.
Second. Lash at several points uses the absence of legislative history on a particular point – legislative silence, rather than statements – as if the absence of such legislative history somehow established that the text of Section Three must therefore be regarded as "ambiguous." For examples, Lash suggests (at 6) that the office of President of the United States might not be included in Section Three because "no scholar has identified a single example of a ratifier describing Section Three as including the office of the President," making similar inferences-from-silence about coverage of the Presidency throughout the piece (at 21, 31, 41). He makes a similar inference (at 61) from the silence of most people on the issue of self-execution.
These inferences are fundamentally unsound. Just as the Constitution consists of its text and not its legislative history, how much more is it the case that the Constitution does not consist of silence in its legislative history. Negative inferences from silences in legislative history are therefore especially unreliable as a rule. An exception might exist where a proposed reading of the text is so odd or extreme that, if that were actually the correct reading of the text, someone almost certainly would have mentioned it (a "dog-that-didn't-bark" inference). But unless silence in a particular setting shouts loudly and clearly, the absence of discussion in the legislative history, without more, establishes nothing. And interpreters must resist the temptation to assume that their own views are so obviously correct that all opposing views are in this category – thinking, I am so surely right, that if I were wrong, somebody would have said so! Sometimes nobody says anything because there is nothing to say and a point is regarded as obvious – it literally goes without saying; other times they may have limited time, or other concerns, or simply be thinking about something else. Most of the time, we know basically nothing about why nothing happened.
Third. Lash sometimes relies on the statements or positions of opponents of Section Three as persuasive evidence of the text's true meaning, or as at least demonstrating the text's supposed "ambiguity." Indeed, Lash uses dissents to establish textual ambiguity, even where the dissenting views were unrepresentative of the actions of the drafting or ratifying body and are demonstrably unsound readings of the text.
For example, Lash, in discussing the question of ratifiers' views on "whether Section Three applied to future insurrections," states (at 45) that "[v]ery few ratifiers specifically addressed" the question, but those who did "came to different conclusions" on this point. But Lash's chief evidence of supposed disagreement comes from the expressed views of opponents of ratification. Lash gives considerable attention to a "Minority Report authored by members of the Indiana Assembly" which "criticized Section Three because it applied only to the past rebellion." (Id. at 45.) Lash quotes the "Minority Report" as follows:
[Section Three] disfranchises all of that class of persons therein named, who "shall have engaged in insurrection or rebellion against the United States, or given aid or comfort to the enemies thereof," but denounces no penalties against those who may hereafter commit the same act. . . . It would be difficult, in our opinion, to frame a law more thoroughly the offspring of passion, and less in accordance with sound policy and statesmanship.
But to place such a provision as this in the Constitution – the organic law which is designed to last for ages, affecting, as it does, past offenses and offenders only, and containing no guarantees for the future, and that must become obsolete at the end of the present generation, is an act of folly that vengeance and not statesmanship could sanction.
(Id. at 45, quoting Indiana committee minority report.) Lash adds the further comment that it "is not clear" whether the majority of the Indiana Assembly "disagreed with the minority's understanding or whether the majority had no objection to a clause addressing only the past rebellion (or whether they found the matter not worth defeating the entire proposal regardless of the meaning of Section Three)." Ms. at 45-46.
As purported evidence of the meaning of Section Three's language, this is highly unreliable stuff. It is not the view of the Indiana Assembly as a body, which voted for ratification, nor of the committee majority, but of its minority opponents. It is also just unpersuasive interpretation of the text of Section Three by those opponents. As we set forth in our article (Ms. 7-11), the text of Section Three sets forth a general rule not limited to the specific historical circumstances that prompted its enactment. To offer, as evidence of possible ambiguity on this score, notwithstanding the seemingly quite plain text, the non-textual analysis of a lone minority report, is about as feeble support for the "ambiguity" proposition as one can imagine. (Indeed, if we were to speculate, it seems just as likely that the minority was repeating a shopworn democratic talking point that had been leveled against the disenfranchisement-until-1870 provision that we discussed in the last post. But such legislative history is unreliable precisely because it requires such speculation.)
Lash then compounds this error by implying that the Indiana majority, by not saying anything in response on this point, might possibly have agreed with the minority on this point but not wanted to mention the fact! This is the mistaken inference from silence again. Silence in the face of nonsense by the minority does not imply acquiescence by the majority – this is a point that should be familiar to nearly every reader of judicial opinions, and also clear from common experience. (We wish to be clear: We do not implicitly agree with every argument we do not refute in this short series of essays! Alas, some arguments simply do not merit attention.)
Lash also discusses at considerable length "The Concerns of Thomas Chalfant," a ratification opponent in the Pennsylvania legislature who believed Section Three would have self-executing legal force as a constitutional rule of immediate application but who also thought that there ought to be a further requirement of prior criminal prosecution and conviction for treason or other crime as a prerequisite to disqualification – but that Section Three contained no such requirement. In essence, it appears that Mr. Chalfant's statement agrees with our analysis of Section Three's meaning on these points, but that Chalfant opposed Section Three as a result. Lash states that "Chalfant presumed that every ratifier in the room agreed with him" that (in Lash's words:) "the only possible solution to the raft of problems" was "Congressional enabling legislation." (Ms. at 53.) But it is mysterious why Lash is justified in drawing this inference. It seems at least as plausible that Chalfant and his audience shared a view that Section Three was self-executing, and he was trying to convince them that this was bad and dangerous.
The core problem with relying on such dissenting views, especially relatively isolated examples, is that it is hard to make responsible inferences when the goal of the dissenters is to defeat the proposal. Sometimes the dissenters correctly state what the majority's proposal is doing, and simply disagree about whether it is good or bad. But other times they incorrectly describe the proposal in order to make it easier to oppose. When there is direct exchange, a meeting of the minds, or some explicit common ground between the majority and the dissent, we can sometimes draw plausible conclusions about their understandings of the text. (We will come to such an example in the next point.) But the claims of an isolated dissenter give us little to work with – unless the user of the legislative history is simply looking for citations to cover for his own pre-existing views.
Fourth. A further problem with Lash's use of legislative history – and an example of his highly selective deployment of skepticism – concerns his bizarre treatment of the exchange between Senator Reverdy Johnson and Senator Lot Morrill over whether Section Three's language covers the offices of President and Vice President. As set forth in our article (at 109), Reverdy Johnson raised an objection to the text as he understood it – its failure (he thought) to include the Presidency and Vice Presidency as covered offices. The proposed amendment "does not go far enough," Johnson said. "I do not see but that any one of these gentlemen may be elected President or Vice President of the United States, and why did you omit to exclude them?" (Cong. Globe, 39th Cong., 1st Sess. at 2899 (1966)). To which Morrill replied: "Let me call the Senator's attention to the words 'or hold any office, civil or military, under the United States.'" Johnson accepted this as a complete answer to his concern. "Perhaps I am wrong as to the exclusion from the presidency; no doubt I am; but I was misled by noticing the special exclusion in the case of Senators and Representatives."
This is an unusually clear and helpful piece of legislative history, because the two speakers did find common ground after their initial disagreement. Sometimes legislators' statements merely muddy the interpretive waters. But here, on a point where the legislative history of the proposal of Section Three is illuminating – where it tends to clarify a possibly uncertain point, serves to answer a question honestly framed, and is a reasonably reliable indicator of the meaning of constitutional terms as they were being used at the time by proponents of Section Three – Lash oddly treats such legislative history not as probative of likely textual meaning but, just the reverse, as proof of ambiguity! In a nutshell, Lash's argument is that because a question was raised in the legislative debate, the text must therefore be ambiguous on that question, regardless of the answer given to the question and apparently accepted as satisfactory.
Lash asserts that the Johnson-Morrill exchange was not widely reported in newspapers, citing one newspaper that did not report this exchange and saying that he had found no others that reported it. But even assuming that Lash is correct about this, it hardly means that the text is therefore "ambiguous" because other persons, Lash speculates, might have misread the text as Johnson initially did, and not had the benefit of specific knowledge of the corrective exchange between Johnson and Morrill.
This is very strange use of legislative history. It is not used in service of ascertaining the objective meaning of the words employed in the text in the sense they were understood by the persons using them. Rather, Lash uses it to suggest that a legislator's initial confusion, even if clarified and agreed to, means that the initial confusion was therefore justified and the agreed-to meaning is actually evidence that the text's meaning is ambiguous. With respect, the proposition practically refutes itself. As Michael Stern, an expert on Congress, puts the point in an extensive criticism of Lash's use of legislative history:
Somewhere Justice Scalia is spinning furiously in his grave. He famously opposed the use of legislative history to resolve textual ambiguity, but this goes far beyond that. It is using the fact that a senator was confused upon casual reading of a legislative text as evidence of its ambiguity, even though he quickly agreed that the text was not ambiguous. And it is suggesting that because some members of the public might also have been confused by the text (which pretty much applies to every legislative text) that itself establishes ambiguity. . . . At the end of the day this is an attempt to create ambiguity out of thin air.
Fifth. Professor Lash draws unwarranted inferences from differences in language choices between one earlier unenacted proposal and the eventual language employed in Section Three. Notoriously, there were many different competing proposals for constitutional amendments suggested at early stages in the process that eventually led to proposal of the Fourteenth Amendment. These various proposals had different objectives (sometimes at odds with one another), different points of concern or emphasis, and different language. It is fair to describe the process as fluid, especially at earlier stages.
Nonetheless, Professor Lash draws an inference from the existence of language in one such early proposal that specifically mentioned the offices of President and Vice President and the fact that Section Three instead used the general category of "offices" under and "officers of" the United States. Lash is more careful here not to overstate his conclusion. He does not argue (at least not quite) that the use of "President" and "Vice President" in the early proposal, and the absence of such language in Section Three, warrants the firm conclusion that Section Three does not cover those offices and officers (in its more general language). But he comes close, suggesting that this is a possible valid inference and renders Section Three ambiguous on this point:
In sum, comparing the final draft of Section Three to the full set of prior drafts renders it unclear whether the drafters intended the final language to include the office of the President of the United States, or barred persons from seeking to "qualify" for that office. Future ratifiers following the framing debates in the daily newspapers could reasonably conclude that the framers had intentionally omitted the language of prior drafts. (Ms. at 4, emphasis added).
Put to one side questions of what "future ratifiers" might or might not have read in newspaper reports of early proposals. The true problem with all such uses of constitutional legislative history – making textual comparisons between unenacted draft proposals and final versions ultimately enacted, and drawing inferences from these differences – is that it is often difficult to tell what inference is the proper one to draw. The Supreme Court has recognized this problem, observing in District of Columbia v. Heller that it is "always perilous to derive the meaning of an adopted provision from another provision deleted in the drafting process."
The short explanation for this hesitation is that one never really knows – at least not without a contemporaneous explanation being given at the time that explains, reliably, the reason for the difference between different drafts – exactly why different language was chosen. One perhaps can make intelligent guesses as to why the change was made, and draw possible inferences as to what differences in meaning are thereby signaled. But this is always to some extent guesswork, absent a reasonably explicit and authoritative explanation. One possible inference is that the content of what was omitted was meant to be rejected. But another, quite opposite, possible inference is that the omitted content was viewed as redundant, fully embraced by other retained or substituted language. Both possibilities occur frequently. There is no universally correct answer, and no clear rule for attributing meaning to changes between or among successive or multiple unenacted drafts. The answer depends to some degree, we submit, on context and ultimately on the objective meaning or more natural understanding of the language ultimately employed, standing on its own, in the context of the enactment as a whole.
This brings us back to the point that ultimately it is the language of the Constitution that is the most probative, not the language of the legislative history. If the language ultimately adopted itself reads as rejecting the language of an earlier draft, that is the likely inference. And, by the same token, if the language ultimately adopted reads as entailing or embracing the prior language, then that is the likely inference.
It is precisely for this reason – a focus on the words finally employed – that, on the question of whether Section Three applies to insurrections and rebellions other than the Civil War, we believe the correct answer is that it does. The final language adopted is cast in general, prospective terms. The fact that earlier versions of some proposals would have explicitly limited that version of section three to the "late rebellion" is instructive, by way of contrast. The better inference, if any is to be drawn, is that any limitation to the Civil War was rejected. But ultimately what is most important is that the language ultimately adopted contains no such limitation and that it is extremely difficult to infer from that language an intention to incorporate the limiting language of an earlier, unenacted draft.
Lash implies that the difference between an earlier proposal's specific listing of President and Vice President as covered offices and Section Three's use of the general categories of "offices" under and "officers of" might warrant a similar inference that coverage of the President and Vice President was rejected. This seems to us wrong, for two reasons. First, the language ultimately chosen does not plainly reject inclusion of the President and Vice President. The general language of "officers" is equally (if not more) susceptible to the reading that it includes, by the general category, what a different proposal had listed specifically. Second, on this issue, as noted above, the legislative history does clarify which reading – rejection or incorporation – is the better inference. Reverdy Johnson asked; Lot Morrill answered: the language chosen was understood by its proponents – and accepted by those questioning it – as including the President and Vice President.
As noted, Lash takes some care not to overstate the strength of the inference from the difference between an unenacted early draft and the final language used. But advocates and amici in the Trump v. Anderson litigation have not been so restrained; in some cases they have wildly overstated the inference that Lash says might be drawn from the legislative history in this respect – even to the point of outright misrepresentation. The Meese/Mukasey/Barr/Calabresi/Lawson brief makes an egregious misrepresentation on this score, asserting in a point heading (at 13, point I.B.) that "Prior drafts of Section 3 confirm that excluding the President was deliberate" and arguing, well beyond anything Lash says, that "the words "President or vice president' were deliberately edited out of the final version of Section 3." But the words were not actually "edited out." The earlier draft was merely one of several early proposals that had been offered, and which was no longer under discussion. The two versions differed in the language used, but that is all. And whether the differences were substantive, let alone deliberate, is also unproven. The brief's inference that these changes "make[] clear that the Framers of Section 3 did not intend for it to apply to those running for President or Vice President" is ultimately a flat misrepresentation of what the legislative history supports.
* * *
Constitutional legislative history can be a slippery thing. While it can provide potentially probative, second-best evidence of constitutional textual meaning, it can also be misused and abused to mislead and misinform. One must not confuse a limited purpose with a limited (or ambiguously-maybe-limited) text. One should not treat legislative silence as acquiescence or ambiguity. One must be wary of inaccurate statements by isolated dissenters to a law. One should not selectively ignore clear evidence against one's thesis. And one should be careful in drawing inferences from changed or abandoned proposals. Unfortunately, Lash's analysis of Section Three—and even more so some of the commentary and briefing that has relied upon Lash – does not respect these limitations and simply cannot be trusted as legal analysis.
