When the Constitution was written and ratified there was no free-floating principle of sovereign immunity, but now the Supreme Court has recognized one.
In my most recent several posts in this series, I set forth the nearly overwhelming case for a broad understanding of “high Crime and Misdemeanors” as a matter of the original meaning of the Constitution. Those posts addressed, in turn, the English backdrop understanding and usage of the term; the Constitutional Convention’s conscious choice to adopt that well-known term-of-art standard in preference to other formulations; the sweeping understanding of the impeachment power set forth in The Federalist; and other ratification-era evidence fully supporting such a broad understanding.
I ended the last post with discussion of serious and thoughtful debates in the First Congress in 1789 displaying a roughly contemporaneous, early post-enactment understanding of the impeachment power (in the context of addressing a different constitutional question), followed by discussion of James Wilson’s early learned commentary (1790-92) on the scope of the impeachment power in his Lectures on Law and the much later learned commentary of Joseph Story in his Commentaries on the Constitution (1833). All of these sources support the conclusion that the scope of the constitutional power of impeachment is extraordinarily broad, embracing “political” offenses against the Constitution and the system of government. Even so, I was somewhat more skeptical about the weight of post-ratification statements as evidence of original meaning – and especially of Story, because his views are “second generation” understandings.
What About Early Practice?
What, then, of early practice – congressional “precedent” as it were – regarding impeachment? Is that competent evidence of constitutional meaning? Is it reliable evidence, in the same way as other early sources, entitled to the same (or at least some) weight? Or is it entitled to less (or no) force in consideration of the Constitution’s original meaning?
My starting premise is this: As we begin to drift further away from contemporaneous statements and other evidence of the probable objective public meaning of the Constitution at the time of its adoption – evidence of generally understood backdrop understandings that formed part of the linguistic currency of the day; the records of the Constitutional Convention evidencing deliberation over specific word choices; The Federalist’s excellent explanation and exposition; the consensus of ratification debate usage and understanding – and move to later (and later yet) post-ratification evidence of understanding, such as the debates in the First Congress over the proposed Department of Foreign Affairs, Wilson’s Lectures, and especially Story’s Commentaries, there is a steady diminution in both the “admissibility” of and the “weight” to be accorded such evidence. 
So too is the early practice of impeachment under the Constitution, beginning in 1797 – some twenty years after the close Constitutional Convention – a less reliable source of the original, objective meaning of the Constitution’s standard of “high Crimes and Misdemeanors.” This is so not only for reasons of temporal proximity to the language’s adoption as part of the Constitution, but for several further reasons peculiar to congressional constitutional interpretation and especially peculiar to interpretations made in the course of actual impeachment proceedings.
First, while the experience of early practice can be interesting, informative, and instructive – sometimes even insightful – such experience consists of after-the-fact, on-the-ground application of the document in concrete situations, not (so to speak) “above-ground,” context-free, more-likely-to-be-objective, contemporaneous-with-adoption abstract exposition of the document. As a matter of first principles, the objective meaning of the written Constitution is the standard for judging the actions of Congress in applying it, not the other way around. As Vasan Kesavan and I have put it in an academic article some years ago, “the Constitution is not best read by reference to subsequent practice.” Subsequent precedent or practice interpreting the Constitution is highly dubious evidence of the original, objective public meaning of the Constitution.
Second, there are sometimes strong reasons not to accord serious weight to congressional “precedent” interpreting of the Constitution, in particular, when it occurs in the course of legislative practice. These are intuitive and obvious: sometimes Congress carefully considers questions of constitutional meaning (as it should); sometimes, however, it does not; and sometimes, Congress considers a question carefully and botches the analysis anyway (just as courts and presidents do). The historical evidence suggests that Congress in earlier days, and especially the First Congress, took its obligation to consider constitutional questions much more seriously than does Congress today. Still, even early Congresses were not always thorough and careful. And even when they were, it does not mean they necessarily got the answers right. 
Third, the unreliability of congressional interpretations of the Constitution seems especially likely with respect to impeachment matters, where, precisely as Hamilton anticipated in Federalist 65, members of Congress might well be corrupted by political motives and partisan loyalties. Impeachment proceedings “will seldom fail to agitate the passions of the whole community, and to divide it into parties more or less friendly or inimical to the accused,” and are likely to “enlist all their animosities, partialities, influence, and interest on one side or on the other,” Hamilton wrote. This creates a “danger” that impeachment decisions in major contested cases will be “regulated more by the comparative strength of parties than by the real demonstrations of innocence or guilt.” We have seen this more than a few times, as I hope to demonstrate in some later posts discussing some important impeachment proceedings in our nation’s history.
Fourth, and also specific to “interpretations” of the Constitution occurring within the context of contested impeachment proceedings, it is hard to know what should count as the relevant “interpretation” or “understanding”: Is it the decision of the House to impeach for offenses it believes constitute “high Crimes and Misdemeanors”? Or is it the action of the Senate in convicting or failing to convict – especially given the two-thirds majority requirement in the Senate and the possibility of acquittal or dismissal resting on grounds other than the question of whether a described act fits within the meaning of “high Crimes and Misdemeanors.” For example, an acquittal might rest on the ground that the offence charged was not satisfactorily proven to have occurred; it might also rest on leniency, on doubt, or – as Hamilton feared – on considerations of party loyalty or low politics.
Relatedly, if the power of impeachment is properly understood as one conferring a fairly broad range of discretion and judgment as to its application, it is difficult to assign much interpretive weight to any particular application of the power in any particular instance. In a certain sense, a decision not to impeach, or not to convict, is much like Congress deciding not to use a legislative power in any particular instance: is a decision not to enact a given commercial regulatory measure an authoritative interpretation of the meaning of the Commerce Clause or the Necessary and Proper Clause? Thus, even if early impeachment “precedent” counts in figuring out the original meaning of the Constitution’s impeachment standard, it is hard to know what rightly counts as such precedent.
The asymmetry of impeachment – the fact that an offence might be impeachable as a constitutional matter but yet not result in impeachment or conviction, for a variety of reasons – makes it especially difficult to accord much interpretive or precedential force to impeachment acquittals. The two-thirds supermajority Senate vote requirement magnifies and amplifies this asymmetry: a minority bloc of one-third-plus-one-member can vote to acquit for any of a number of disparate or undetermined reasons. By contrast, convictions necessarily involve a judgment by a majority of the House and a supermajority of the Senate that the act charged is constitutionally an impeachable offence.
While I am thus generally quite skeptical about the reliability of impeachment practice – even early practice – as saying anything authoritative about the original meaning of the impeachment power, I hazard a hypothesis flowing from the asymmetry of impeachment and the two-thirds supermajority requirement: impeachment convictions generally should carry far more weight than acquittals as “precedents” or “constitutional interpretations” of any sort. Indeed, one might go further: decisions by the House to impeach generally might count for more, in terms of whether an offence falls within the range of impeachable misconduct, than judgments of acquittal on unspecified grounds by the Senate. Indeed, one could go further yet: as it concerns the meaning and scope of what constitutionally might constitute “high Crimes and Misdemeanors,” a Senate simple majority vote to convict – but that falls short of the two-thirds supermajority required – is a stronger reflection of congressional constitutional judgment as to the meaning of the impeachment standard than is the fact of an acquittal by virtue of narrowly failing to attain the two-thirds threshold.
I hope to take up, in a later post – out of interest and for the sake of illustration – several early historical instances of impeachment in practice, in the first several decades under the Constitution. Some of the results, I will argue, are right; and some of them are simply wrong. Some trials resulted in rightful convictions; some trials resulted in wrongful acquittals – including at least one that probably constitutes more of a precedent for a broad impeachment power than a narrower one; no actual impeachment trial, in my view, has produced a wrongful conviction. And some “non-impeachments” – failures to pursue impeachment (as in the case of Aaron Burr) – are hard to take seriously as meaningful precedents of any kind. They surely should not be taken as narrowly construing the scope of the impeachment power in any authoritative sense.
A Broader Understanding
The right answer is that, as a matter of interpretive methodology, these early applications of the impeachment power should not be given much weight as evidence of the objective, original public textual meaning.
The evidence that is entitled to such weight – as my preceding several posts have shown – is almost overwhelming and points in a single direction. The evidence of 400 years of English practice and usage of the term “high crimes and misdemeanors” as a standard for impeachment; the evidence that the framers drew on that understanding, borrowed the institution of impeachment from English practice, and specifically chose the known term “high Crimes and Misdemeanors” with reference to English precedents and in preference to other terms; the evidence of such a widely understood and agreed meaning, as reflected in Alexander Hamilton’s exposition of the impeachment power in The Federalist and as confirmed by statements in the ratification debates, the First Congress, and James Wilson’s Lectures in Law, all points to an extraordinarily broad understanding of the impeachment power.
In my next group of posts, I will attempt to distill broad categories of impeachable offences that can be gleaned from this evidence of the original meaning of “high Crimes and Misdemeanors,” consider the most important objections to such a broad understanding, and then take up some concrete instances of presidential and judicial impeachments.
 For a full treatment, consider the academic article Vasan Kesavan and I wrote several years ago treating the relative reliability of various second-best extrinsic sources, at pages 1164-1183.
 In the above-linked article at pages 1170-72, Vasan Kesavan and I set forth some fairly major constitutional questions that the First Congress probably got wrong.