I started to draft a comment in response to Kagro X's diary, but this is too long for a comment, and frankly, responding to Kagro's diary is too important (I say self-importantly) to bury in a comment.
Please bear with me through the the discussion of legal theory I begin with. I promise it will all come back to Kagro's point about Ickes and why it is terribly misguided (and needs to be clearly refuted).
With all respect -- and I am a huge fan of Kagro X's marvelous work on this blog -- his post reflects a law student's understanding of law and equity, not a lawyer's. Harold Ickes' argument is wrong and does not pass muster as legal analysis, and don't let anyone tell you otherwise. More after the jump...
OK, first the necessary legal theory part:
How "the Law" Works, and Equity Too
To start with Kagro's last point first: It is true that originally, courts of equity had the discretion to apply the Crown's power to avoid any result at law that the Chancellor deemed unjust. However, over the centuries, equity has become more and more lawlike, and has largely been merged into law in the US. With few exceptions (e.g., the Delaware Court of Chancery), there are no longer separate equity courts in the US -- the same courts apply law and equity. More importantly, when courts apply equitable doctrines, they do so by following other cases that have applied those doctrines before and have developed a set of principles for applying them. So a typical equitable doctrine now has a recognized set of elements that must be shown, just like a legal doctrine.
[Here's an example: The legal principle of the statute of limitations (e.g., you can't sue on your contract claim after 3 years have passed) has a counterpart in the equitable principle of "laches." "Laches" means you waited so long that it's unfair to allow you to sue, even if 3 years haven't passed. "3 years" is a rule. "Unfair" is a standard. That may sound vague, but in actual practice, judges will determine what "unfair" means by looking to earlier laches cases and seeing what factors courts have recognized as showing this sort of unfairness. For instance, did you, who wants to sue after a long delay, act in a way that suggested you would not sue? And did the other guy rely on that to his detriment? And so on. Equity is not a free-for-all, and hasn't been for centuries.]
So, "equity" does not mean a judge can do whatever he thinks is generally fair. Rather, principles of equity are similar to doctrines of law, but just a bit more flexible and fuzzy around the edges.
Meanwhile, over on the law side: Through hundreds of years of American jurisprudence and law reform, the law has become far less form-bound than it used to be, back when equity was created to correct the injustices that resulted from rigid application of legal rules. By the late twentieth century, law had evolved from a rules-centric to a standards-centric regime. It's not that law doesn't contain hard and fast rules -- it has many. However, the paradigm now is that in most areas of law, the core principles are applied through flexible standards rather than rigid rules. The reason for this is that using standards instead of rules actually makes it easier to prevent people from subverting the rationale behind the rule.
To explain this point -- which is the key to understanding why Kagro and Ickes are wrong -- I'll give an example from my own area of practice, intellectual property law. To prove copyright infringement, the owner of a copyright must show (among other things) that the work s/he accuses of infringing is "substantially similar" to his copyrighted work. It should be obvious that a phrase like "substantially similar" is not clear enough by itself to be a rule. It is a vague standard. Two different people could compare Harry Potter and the Sorcerer's Stone to some crappy knockoff novel, and come to opposite conclusions about whether the knockoff is "substantially similar" to Harry Potter. Fortunately, there is a massive body of case law applying the "substantially similar" standard -- and that case law has developed more specific sub-standards, and worked out how to apply the standard and sub-standards to many different types of copyrighted works, in various factual settings. So, in a specific instance, if I am considering publishing a story that resembles someone else's story, I can study the case law and get a pretty good idea of whether my story would be found to infringe the other fellow's copyright.
Lawyers and (especially) law professors do disagree quite a bit about whether the standards world we are in now is really better, i.e., leads to more consistent and fairer outcomes, than some other ways we could do things (including the old rules-based approach). But that is the world we live in. That is how our legal system works.
Why "the Law" Refutes Ickes’ Argument, and Kagro X’s Too
Now, at last, we get to Ickes and Kagro. (Thanks for your patience!) Here’s what the relevant DNC rule says:
"Delegates shall be allocated in a fashion that fairly reflects the expressed presidential preference or uncommitted status of the primary voters or, if there is no binding primary, the convention and/or caucus participants."
Harold Ickes based his objection to the Michigan compromise on the principle of "fair reflection" of voter preference. He argued that allocating delegates differently from the election results would violate this principle. Interestingly, Ickes compared "fair reflection" to the First Amendment as a fundamental principle that should guide the B&RC, and supersede petty concerns expressed in other rules. The comparison is a good one, but it doesn't support Ickes/Kagro. The First Amendment is full of broad standards whose precise application is not obvious from the words of the Amendment, but which have been fleshed out over two centuries of case law. For example, "Congress shall make no law restricting freedom of speech." Well, actually, Congress has made lots of laws that restrict freedom of speech. Copyright law, for one. I am not free to publish words that infringe your copyright. I am also not free to speak in ways that defame other people... unless what I say is true... or the person I am defaming is a public figure and I did not know my statement was false, or I did not act with "reckless disregard" for the truth. See how complicated it gets to figure out what "freedom of speech" means? (And that's a very very small sample -- entire law school courses are devoted to that one clause of the First Amendment alone).
It should be obvious by now that "fair reflection" of voter preference, like "freedom of speech" (and "substantially similar") is a broad standard, not a rule with a clear, self-evident meaning.
The fact that "fair reflection" is in the DNC rules at all assumes that there is more than one way to interpret the same election results. Any way of interpreting them that is a "fair reflection" is permitted. Equally obvious is that the DNC did not have much (or any?) precedent for determining what is a "fair reflection" -- or someone would have brought it up during the debate. Several members expressed concern about setting the wrong kind of precedent, so clearly the committee considered precedent to have weight. Without precedent, they had to fly solo in trying to apply the "fair reflection" standard in this context. And it looks like they did a fine job of it...
Applying "Fair Reflection" In This Case Shows the Michigan Democratic Party Did the Right Thing
Here are some things we all know about the Michigan primary:
*Voters were told the primary wouldn’t count, including by Hillary Clinton herself. Therefore it is reasonable to assume that some number of registered voters, greater than zero, did not vote who would have voted if they had been led to believe the election would count. This is supported by the fact that voter turnout in Michigan was significantly lower than in demographically similar states whose voters believed that their primary would count. The primary results cannot possibly reflect the preference of Michigan voters in this category.
*The number of voters who wished to vote for Barack Obama was indisputably greater than zero. Even if one assumes exit polls are somewhat inaccurate, it is absurd to contend that the Michigan exit poll’s 36% showing for Obama actually reflects zero voters. Even the Clinton people conceded in the R&BC meeting that 75% of "uncommitted" voters were Obama supporters.
*It was literally impossible to vote for Barack Obama by any means. Barack Obama was not on the ballot. As we learned yesterday from Michigan Democratic Party Chair Mark Brewer, 30,000 write-in votes were cast but, by rule, not counted. Therefore, it was impossible to vote for Barack Obama by writing in his name. Even a voter who went to the polls determined to vote for Barack Obama could not express his or her preference for Sen. Obama.
*Given that there were more than zero voters who wanted to vote for Barack Obama, but the ballot did not allow them to do so in any way, the primary results cannot possibly reflect the preference of this category of voters.
*If one is tempted to argue that it was Obama’s own fault for taking himself off the ballot, I would point out two things: (i) Barack, Hillary and John Edwards all signed a pledge "not to campaign or participate" in the Michigan primary. "Participate" must mean something other than "campaign" or there’d be no need to add that word. A fair reading is that Obama and Edwards complied with the pledge not to "participate" in the primary by taking their names off the ballot while Hillary broke her pledge; (ii) In any case, whether Obama was right or wrong to take his name off the ballot, the standard here is not whether the candidate acted fairly, but rather, "fair reflection" of the voter’s preference. (If a candidate’s unfair behavior were enough to overcome the "fair reflection" principle, then Hillary would have to be penalized for suppressing turnout by misleading voters about whether the election would count.)
Thus, application of the "fair reflection" principle leads inescapably to the conclusion that the Michigan primary results did not "fairly reflect" the preference of the voters. If the election was not to be thrown out altogether – as the DNC had previously decided must happen, under the rules – then data beyond the official results had to be consulted to figure out what the preference of the voters was. There’s no question that the specific means used to do that were bound to be imperfect. But the evidence that was used was reasonable.
The Nitty Gritty – The Numbers
To show why, I’ll do a rough calculation, ignoring district allocation. Here’s why it makes sense to do that in this case: The actual numbers were 328,309 for Clinton, and 238,168 for Uncommitted, giving Clinton 57.956% of the vote that was above the 15% threshold. That percentage, multiplied by Michigan’s 128 delegates, comes out to 74 delegates, or just 1 more than the 73 Clinton would have received using the actual primary results. So, delegate allocation closely tracked statewide percentage in this instance.
The 30,000 write-in votes, which a Clinton supporter (Ickes, I think) admitted during the meeting were likely mostly for Obama, were over 5% of the total votes cast. They were not counted. Just adding those actually cast votes to Uncommitted (rather than throwing them out, as happened), dilutes Clinton’s percentage of the total enough to knock her down from 74 delegates to 70.45 of 128. Knocking off the extra vote we started with (74 instead of 73), Hillary’s 70.45 translates to 69.45, or about exactly what the R&BC awarded her. Hmmmm.
Perhaps this is not fair to Clinton, however, because not all the Uncommitteds were for Obama. As I mentioned, the Clinton people (Ickes again, I think) agreed with an estimate that 75% of the Uncommitted voters wanted to vote for Obama, and 20% for Edwards. For the sake of argument, I’ll accept that number. Since Edwards’ 20% of the 42% Uncommitted falls below the 15% threshold for receiving any delegates, I’ll drop that from Uncommitted, along with the 5% that was for minor candidates. Adding the 30,000 Obama write-in votes to the remaining Uncommitted, the Obama total is 208,626.
However, Michigan Democratic Party Chair Mark Brewer also reported to the R&BC that 13% of Clinton voters said in the exit poll that Obama was their first choice and Hillary was their second. Remember, back on Jan. 15 there were lots of candidates still in the race – it was not clearly going to be Clinton vs. Obama, and therefore, Obama supporters would not have had a strong motivation to specifically refrain from voting for Hillary, especially as they had been told the election would not count. They could not vote for their first choice, so there was no harm in voting for their second choice in a beauty contest (another example of how being misled about whether the election would count distorts the results as an expression of voter preference).
Even if we assume the exit poll overstated the number of Hillary voters who preferred Obama by 30%, and it was really only 10% of Clinton voters instead of 13%, her popular vote total would be 295,478, and Obama’s would be 241,457. This would give Hillary about exactly 55% of the popular vote and Obama 45%. This again works out to 70.44 delegates, and again, if you knock off the one delegate that the original percentage method added to her total, you end up with 69.44 delegates for Hillary or, about exactly what they gave her.
So already, without even taking into account all the registered voters who stayed away because they were told the election would not count (probably hundreds of thousands of people if other demographically similar states are any guide), and making only slight use of exit polling data, Hillary has dropped to the 69 delegates the R&BC awarded her.
No doubt people will jump on my calculations as based on faulty methodology, but my point is not to be precise about the final results. I agree much more detailed and subtle calculations could be done, and no doubt were done, by the committee of Michigan Democratic heavyweights would came up with the 69-59 allocation.
Conclusion
I have simply tried to show that Kagro X and Harold Ickes are wrong that Hillary’s 73-delegate number is the only one that’s really consistent with "the law" or "the rules." Once you recognize that the principle of "fair reflection" is a standard and not a simple rule, and that the Michigan primary clearly did not meet that standard, then you must go beyond the primary totals to other evidence of voter preference. And a fair consideration of the data is very likely to put you in the ballpark of what the Michigan Democratic Party proposed.