Parchment Barriers: Why Tillman and Natelson Are Wrong about the Anti-Corruption Principle

I have argued in several articles and a forthcoming book that the Anti-Corruption Principle was and is a foundational constitutional principle. Larry Lessig has argued the same in a book, several articles, and a brief before the Supreme Court. He also runs a fascinating tumblr of corruption conversations at the convention. We both show how it motivated the Constitutional convention and was the overwhelming topic of the convention. The founding drafters would judge their own success or failure in terms of the Constitution in terms of whether or not it protected against corruption.  As George Mason said as the Constitutional Convention got under way: “If we do not provide against corruption, our government will soon be at an end.” I am not going to recite the argument here–its a substantial, text and history based argument. However, it leads to treating the Anti-Corruption Principle like federalism or the separation of powers–a fundamental structural part of the Constitution.

There are two general disagreements with this view: First, that I, and or Lessig, don’t characterize the meaning of corruption at the time properly (we have slight differences), and second, that it was not and is not a Constitutional principle. Seth Tillman, and Rob Natelson citing Seth Tillman, have both recently argued that there is no such principle. Tillman sees  a kind of smoking-gun flaw with the argument–the evidence that the word “corruption” was taken out of the impeachment clause as a ground for impeachment.

Tillman writes: “But when the Framers had a chance (actually multiple chances) to give this concept prominence in the Constitution’s actual text, the Framers chose not to do so. It is not as if they forgot to use this term or, instead, used some close synonym; rather, they actively took this term out of the Constitution. So why should we today embrace the corruption concept as one having constitutional scope or dimension?”

I have three responses. First, the absence of the word “corruption”  does no more to the anti-corruption principle than the deliberate choice to exclude “separation of powers” does to the separation of powers principle. Second, the reason the word was taken out is because corruption is ill-suited to bad-intent criminal law statutes, and well-suited to prophylactic statutes or constitutional provisions that don’t reference state of mind. Tillman has a common confusion–he treats the scope of the word corruption as that which can be defined in a criminal-law like statute. Third, Tillman’s particular argument contrasts with his general support for the anti-corruption principle.

(1)First,  the founders made a similar choice with regards to separation of powers. When critics objected, Madison skewered them with the argument that they were making “parchment barriers” be more important than fundamental understandings of power. If Tillman’s argument applies to the anti-corruption principle, it applies with even greater force to the separation of powers principle. 

There is no text in the Constitution which refers to “separation of powers.” The choice not to include this separation of powers principle was deliberate, and in direct opposition to the trend in the states. Several State Constitutions had particular sections of their constitutions committing their state governments to separation of powers. In Maryland, for instance, the Constitution included this phrase: “…”. Virginia’s Constitution stated that “the legislative, executive, and judiciary departments shall be separate and distinct; so that neither exercise the powers properly belonging to the other; nor shall any person exercise the powers of more than one of them at the same time, except that the justices of county courts shall be eligible to either House of Assembly.” In North Carolina, the Constitution allowed that “the legislative, executive, and supreme judicial powers of government ought to be forever separate and distinct from each other.” Georgia’s Constitution stated that “the legislative, executive, and judiciary departments shall be separate and distinct, so that neither exercised the powers properly belonging to the other.” Many other states had similar provisions. A few years after the Constitution, the Declaration of Man in France included an express “separation of powers” provision:A society in which the observance of the law is not assured, nor the separation of powers defined, has no constitution at all.”[1]

During ratification debates, critics pointed out the absence of such a provision.

They also argued that the new Constitution’s structure violated the basic principle that separate departments was supposed to serve. The “vortex of influence,” as Lee calls it in the Antifederalist papers 77 and 78, was a central evil to be avoided, but could be avoided if the virtuous aspirations of good men were tapped into. The problem was that the new federal government was going to have few enough legislators that each one of them could be tempted by the possibility of serving in a lucrative office. The legislators, being prominent men, would likely expect that they would be selected to serve in offices after having served their constituencies. So the new Constitution did not adequately protect against this temptation by including a set amount of time in which the legislators could not later take one of these offices. Lee was concerned that the legislative function could be hurt by the “hankerings and corruption of offices.”[2]

The veto power, critics argued, would undermine the basic idea behind separated functions of government. William Penn explains the core meaning of separation of governmental powers in federalist 73. Tyranny comes, he explains, when the same group of people who have the power to raise money can use it for their own benefit. If legislators have the power to appoint, he argued, they will appoint friends to offices. Penn’s great concern was that the new constitution gave the executive branch legislative power with the veto, or the negative. He believed that with the “negative” power, the Executive could come to control the legislative branch, then having in one place the powers that are most dangerous—the ability to create and fund lucrative offices.

Critics charged the new Constitution with being a function of the elites, designed to serve elite interests. The Constitution was designed for “men of great wealth in expectation of public offices who “like greedy gudgeons” want to “satiate their voracious stomachs with the golden bait.” The golden bait is jobs.

Madison summarized these accusations as such:

One of the principal objections inculcated by the more respectable adversaries to the Constitution, is its supposed violation of the political maxim, that the legislative, executive, and judiciary departments ought to be separate and distinct. In the structure of the federal government, no regard, it is said, seems to have been paid to this essential precaution in favor of liberty. The several departments of power are distributed and blended in such a manner as at once to destroy all symmetry and beauty of form, and to expose some of the essential parts of the edifice to the danger of being crushed by the disproportionate weight of other parts.

Madison’s response to the charge was to separate out the formal statement of the principle (separation of powers) from the precise concerns related to the principle.. Madison, in Federalist 47 and 48, rejects the formal understanding as a parchment model that is unrelated to the real concerns. His argument depended on two pillars. First, he argued, there was no evidence of a clean line between the executive, legislative, and executive branches within the existing states that had, in fact, included explicit provisions for separations within their constitutions. Because of this, he said, a “parchment” commitment to separation of powers would do no practical work. The state constitutions he believed to have been “hastily” drafted, and without the benefit of experience. Second, he argued that there was necessarily some overlap between the different functions, and a strict rule could disable a functioning of the principle: power can’t be drafted in a law, while functions can, but that doesn’t mean that power isn’t important.

Instead of separation of powers as a formal commitment, Madison says, what is important is that the structure be built in a way that would lead to too much accumulated power in one branch of government. The trouble is not overlapping functions, but concentrated power, and the things that accompany that concentration of power. The complaint, therefore, is not so much that power be separated between three branches, or within the branches—Madison never accepts that argument. Instead, the trick is that there be no way for accumulation of power in “the same hands.” “Parchment barriers” are overrated.

I could go further–for now suffice it to say that I hope Tillman and Natelson are equally willing to dispose of the separation of powers principle.

(2) Second, Tillman confuses the legitimate scope of the meaning of corruption with its legitimate use as a tool for individual judgement.

This far, Tillman is correct:

“Corruption” appeared in a preliminary draft of the Constitution’s Impeachment Clause. But this language was dropped, and superseded by “treason, bribery, and maladministration,” but the “maladministration” language appeared too vague. (Meigs, The Growth of the Constitution in the Federal Convention of 1787 233-34 (1899).) The Convention did not return to the earlier “corruption” language, and instead, it chose “treason, bribery, or other high crimes and misdemeanors.”

This seems right to me, and not an indication on its face that there was an anti-corruption principle, but rather that it was not an appropriate tool for the role of determining who should be impeached. Bribery and extortion were high misdemeanors at the time, so any “high crimes and misdemeanors” would include bribery and extortion and other crimes related to abuse of office. The choice to take out the particular word, as too vague, doesn’t mean the founders didn’t believe in corruption (the most cursory review of any writing shows the opposite), but that they saw government’s tools for fighting corruption to be better placed in prophylactic rules. The anti-corruption provisions of the constitution, like emoluments clauses or the veto clause or the residency clause are designed to limit corruption but don’t reference particular corrupt actors.

Assume, for a moment, that corruption was taken out for vagueness–that doesn’t make it not a founding era principle, it simply makes it a bad candidate for being the grounds of impeachment. Arguably, the vagueness cuts the other way–it is such a fundamental concept of enormous foundational import that it would threaten to trivialize it if were litigated in terms of legitimate impeachment.

Laws that include or refer to “corrupt” or “corruptly” always  threaten to violate some of these requirements of non-vagueness, because it is not always clear and intelligible what behavior is covered by corruption statutes.

Let me give an example. A jury is told they must find that a defendant acted “corruptly” to convict under a federal bribery statute and that corruptly means, among other things, “intent to give some advantage inconsistent with official duty and the rights of others.” The scope of “official duty and rights of others” is broad. In some jurisdictions, the jury is essentially asked to make a judgment about whether the defendant thought he shouldn’t be getting the advantage he was getting. This seems to give prosecutors, and juries, enormous leeway to determine what constitutes official duty and deviance therefrom. One might argue that this is an unworkable and unconstitutional jury instruction, because it gives too much law-making ability to fact-finders, and leaves citizens without clear guideposts about what constitutes illegal vs. illegal behavior. Rick Pildes and Sam Issacharoff argued as much in an amici brief on behalf of former Alabama Governor Don Siegelman, convicted of bribery (among other things) for trading official acts for campaign donations. They argued that the vagueness of the jury instruction in Siegelman’s case is unconstitutionally vague. Here’s the instruction: “A Defendant does not commit a crime by giving something of value to a government official unless the Defendant and official agree that the official will take specific action in exchange for the thing of value.” Instead, they argue, the Court should instruct the jury that it must find evidence of an explicit quid pro quo.

However, whatever legal determination is made about whether it is a legal jury instruction should not then be used as evidence of the scope of what “corruption” means more generally. It is simply a statement about what is appropriate to criminalize. The scope of permissible enacted corruption concepts must be disentangled from the question of the meaning of the concept. The limits of possible law do not serve as the limits of conceptual understanding. If they are not untangled, the conceptual analysis suffers from an unspoken and under-theorized positivist limitation on the corruption concept. The sensible thesis, which is simply that: “Laws that reference corruption must do so in a way that will not violate the Due Process clause” should not bleed into this less defensible thesis: “corruption means only that which can be written in a statute that does not violate the Due Process clause.”

The sensible conclusion that corruption is too vague as a basis for impeachment has no fundamental relationship to the argument that anti-corruption was a foundational constitutional principle. To tie these two first arguments together, the founders could have taken out an impeachment clause for “bribery, treason, and violation of the separation of powers” on grounds that the latter was too vague for impeachment; that wouldn’t  make it too vague to have constitutional meaning.

(3) Tillman’s initial historical read was more supportive.

Tillman, up to now, has generally supported the argument for an anti-corruption principle . In a colloquy between us at Northwestern he wrote: “There is a structural anti-corruption principle of constitutional dimension. That principle, the ACP, vindicates substantial congressional control over statutory officers, i.e., officers under the United States. It puts federal statutory officers in a fiduciary relationship under the government, in respect to the officers’ elected masters and all of the nation’s citizens.” However, he and I disagree about the scope–he thinks it applies only to appointed, not elected officers. Tillman’s general interest, for those being introduced to the debate, is the meaning of the word “officers”–his argument is in the prior article, my response on officers is here. We disagree about offices, but I think he has strong and interesting arguments. I hope he returns to this basic principle, and engages the question of what corruption means inside the historical context of the acceptance, instead of rejection of the principle.

You may also like...

10 Responses

  1. brad says:

    Bringing the anti-corruption principle in line with separation of powers seems like a weak victory. Although it is certainly given lip service, with the non-delegation doctrine in virtual abeyance, the unitary executive widely rejected, and article I courts flourishing it seems a principle more honored in the breach than the observance.

  2. NLS says:

    Correct me if I’m missing something here, but your argument is that (1) neither the terms “separation of powers”, nor “corruption” appear explicitly in the constitution, (2) the separation of powers doctrine shouldn’t be limited by parchment barriers simply because it does not appear explicitly in the constitution, and (3) the anti-corruption principle should, likewise, be afforded the same deference.

    My problem is this: the separation of powers doctrine can be seen in the basic form of the Constitution, and it is derived from the accumulation of many different sections and interpretations of the document. Heck, it was even defended by Madison. Corruption was once thought about during the drafting of impeachment proceedings, and the anti-corruption doctrine can really only be traced back to early 20th century election administration laws. These are not alike. They are far from alike. I am not trying to make a qualitative argument about either side of the campaign finance debate here. But this particular argument just seems erroneous.

    You’re arguing about an orange when we’re talking about an apple in the book of Genesis. Neither is mentioned in the book, sure, but one has a rich history of religious/art/etc tradition based around it. The other is also a fruit.

  3. Jon Roland says:

    A few mistakes in the text:

    The anti-federalist author “William Penn” should be in quotes to avoid confusing him with the founder of Pennsylvania, and his paper on this was assigned “No. 2” by Storing. “Federalist No. 73” should be capitalized, written by Hamilton. In (2) the word should be spelled “judgment”. Anti-federalist papers are only given numbers by Storing and a few other editors, but each uses a different numbering system.

    The passage “what is important is that the structure be built in a way that would [not] lead to too much accumulated power” needed the inserted “not”.

    Missing is the point that the Constitution does not confer authority on Congress to make “bribery and high crimes and misdemeanors” a crime, as it does for “treason”, although it arguably does for discipline of the military and militia in federal service. However, that does not cover civil officers. I discuss this omission in “Defect in the Constitution” at

  4. Jon Roland says:

    A few mistakes in the text:

    The anti-federalist author “William Penn” should be in quotes to avoid confusing him with the founder of Pennsylvania, and his paper on this was assigned “No. 2” by Storing. “Federalist No. 73” should be capitalized, written by Hamilton. In (2) the word should be spelled “judgment”. Anti-federalist papers are only given numbers by Storing and a few other editors, but each uses a different numbering system.

    The passage “what is important is that the structure be built in a way that would [not] lead to too much accumulated power” needed the inserted “not”.

    Missing is the point that the Constitution does not confer authority on Congress to make “bribery and high crimes and misdemeanors” a crime, as it does for “treason”, although it arguably does for discipline of the military and militia in federal service. However, that does not cover civil officers. I discuss this omission in “Defect in the Constitution” at my Constitution blog, linked from my website

  5. Pathfinder says:

    Lessig and Teachout are engaged in the worst type of “scholarship” – attempting to conjure up a theory that will turn their policy preferences into a constitutional mandate. In doing so they ignore all other possible readings, including many that have more direct support in the Constitutional text and structure (which they pretty much have to admit).

    Obviously the Founders were concerned about virtue and preventing corruption in government. (That is essentially their point of agreement with Tillman). But this does not and cannot mean that there is some free-floating “anti-corruption” principle embedded in the Constitution, a sort of vague “legislative intent” that overrides the actual language and structure, to be pulled out whenever one needs to support one’s favored government policy (for Lessig and Teachout, broad restraints on political speech and participation). It would make as much sense to comb the page of late 18th century political tracts to create a basic “liberty” principle that trumps the actual language and structure of the Constitution, causing most of the current edifice of government to fall; or to find an “equality” principle dictating much more redistribution of income. The framers were concerned with many things. They enacted a Constitution to address those concerns. To suggest that one area of concern, one broad principle of government, somehow trumps specific provisions of the Constitution or overrides its structural elements is absurd, particularly when they further interpret that broad principle, rather tendentiously, as being achievable only through the policies they favor.

    Their approach is to ignore opposing evidence or to simply interpret all evidence as supporting their preferred policy outcome. Thus, for example, they interpret the removal of “corruption” from the text as supporting their view – and just as surely, they would have interpreted leaving it in as supporting their view.

    They take broad problems and concepts articulated by Madison and others and then ignore the actual solutions/provisions of the constitution aimed at addressing those problems/concepts, to instead argue that the “real” constitutional solution is an “anti-corruption principle,” which only they truly understand and can interpret for the rest of us. And – surprise – when they interpret it for us, why, it turns out that the true, hidden Constitution and their policy preferences 225 years later line up in perfect harmony.

    This isn’t even “nonsense on stilts.” It’s more like nonsense with shoe lifts.

  6. Zephyr Teachout says:

    Brad: I agree that the SOP is not that strong. But it has some weight, and is used, as I think the ACP should be used, to understand other parts of the Constitution, to give a gloss on texts, and to be a freestanding principle from time to time.

    Jon: Its a blog! Yes, there are mistakes.

    NLS: Actually, the ACP is just as deep and strong and a fundamental constitutional commitment that shows up throughout the 19th century. I hope you read my forthcoming book (it will be several months), and, in the meantime,

    Pathfinder: I’m sorry you find it unpersuasive. But my research preceded any particular pending case–it came from a genuine confusion about the thin-ness of the concept as expressed in Buckley, and anyone reading the papers of the era will find that corruption kind of hits you over the head–as Wood puts it, its the common grammar of the era. Unless one has a view of the constitutional that allows for no structural principles–in which case I understand your argument–the ACP is as fundamental a principle as any. And even if you believe in no structural principles, the ACP should help define the scope of the other provisions.

  7. NLS says:

    Thank you, I’ll definitely read.

  8. Rob Natelson says:

    Unfortunately, I’m being mischaracterized here. I never said there was no anti-corruption principle. I DID say that the Founders believed republican government and officeholders in republican government had obligations of “trust”—a term more generic than that used today to include fiduciary duties generally. Among these were a duty to avoid conflicts of interest—a concept included in the contemporaneous meaning of the term “corruption.”

    Certainly fiduciary duties should inform the reading the Constitution. The issue is the extent to which they should trump the text, as the Founders understood that text. For example, it now seems clear that the contemporaneous meaning of the term “freedom of the press” included a right to anonymity when contributing to political debate through the media. I don’t think, therefore, that a generalized fiduciary principle should be used to override the meaning of that phrase to uphold disclosure statutes inconsistent with “freedom of the press” (which includes political advertising).

    Finally, I made another point not mentioned here: The Founders’ concerns about conflict of interest actually cuts AGAINST the view that the Times, Places and Manner Clause authorizes congressional regulation of federal election campaigns. The matter was discussed at excruciating length during the ratification debates, and the only way the Federalists were able to obtain acquiescence to the Clause was to represent that “Manner of holding Elections” would be construed narrowly. As then understood, that phrase wouldn’t have extended to regulation of campaigns anyway, but the ratification-era discussion makes it very clear that most people—including some strong Federalists such as Noah Webster and James McClurg (himself a framer)—did not want Congress regulating federal elections unless there was some kind of disaster, such as an invasion, that prevented one or more states from doing the job. The sort of detailed congressional campaign control we see today entailed precisely was the kind of conflict of interest they would have called “corruption.”

  9. Brett Bellmore says:

    I think part of the problem here is the notion that corruption is something which proceeds from the public, to the government, that what we need to do is keep the people from corrupting the government. That corruption may be originating with the government itself, (Campaign funds being extortion, not bribes, for instance, seems not to be a consideration.

    So the conflict of interest involved in officeholders regulating how people may try to unseat them just doesn’t register.

  10. Zephyr says:

    Rob–thanks for chiming in. As I read your post, you said that Lessig and I asserted that there was an anti-corruption principle, and Tillman’s response “devastated” those assertions. I read that to be a rejection of the ACP. I’m glad to know that you are at least open to looking at things through the lens of the ACP.

    The early TPM clause work you’ve done is really interesting and I look forward to reading it more closely. Two questions, serious one: do you think the states should have the capacity to have campaign finance limits? And is your view that bribery is outside the normal conduct of elections mean that if we see BCRA as a bribery statute, we think it is within federal authority? I may elevate this discussion to a full post–look forward to your thoughts.

    Brett: I agree with you about conflict of interest. In fact, one of the problems with the current Kennedy vision of corruption is that it does not include conflict of interest, or if there is a “conflict of interest” corruption he can’t thematically link it to “quid pro quo” corruption. It points out, I think, the thinness of the quid pro quo frame. I don’t think that it is silly to be concerned about self-interested governmental behavior at all–and that the point is a serious one. However, I think bright line rules, like the conflict of interest statutes and provisions and bright line contribution limits, do a better job of regulating governmental and private abuse of public power than, say, “bad motive” bribery laws which give enormous governmental discretion. (Allowing for partisan prosecutions, etc)