Now that I've told about the rally, I may as well finish the story. The next day [Saturday, Feb 9] was the caucus. It was scheduled for 1:00 pm at an elementary school about a mile from here (0.99 miles, according to Mapquest). I had originally intended to walk, but I was slow getting ready so at about quarter till I decided to save some time and drive after all.

I didn't save much time. I spent about five minutes looking for somewhere to park, and I ended up about a third of a mile away. It's not a crowded area with scarce parking by any means, but the turnout was enormous. The Republican caucuses were at a large church a block away, so after parking lots filled at each site the overflow overlapped. (Though the imbalance isn't as extreme here as in Seattle, Democrats still outnumber Republicans by quite a bit here in Shoreline, too. I'm sure the Republican caucuses represented people drawn from a larger area.)

For the Democrats, the local elementary school was one of 18 sites for the district (ie, Washington state legislative district, one of 49 in the state). It hosted 24 precinct caucuses; 12, including mine, were in the cafeteria, and the other 12 were in two other rooms elsewhere in the school. The cafeteria was packed. People outnumbered chairs by at least three to one, so most people were standing with a few sitting on the floor. There was a large cafeteria table for each precinct. My precinct's happened to be near the side of the cafeteria where there was a large recessed alcove, forming a sort of stage. Various large items (covered piano, etc) were packed up into this area, so the stage space wasn't usable. What interested me was that it was three steps higher than the main floor, so it provided a wide stretch of steps to sit on. Several people had taken advantage of that, and I joined them, at the end near my precinct table.

At the last caucus I attended, in 2004, I was struck by two things. One is that the people I met were far less bitterly partisan and negative about other candidates than one sees in almost any public forum (eg, blog comments, talk radio call-ins). This was true again this year. I believe this is because the public in general is not very partisan, but the minority who are dominate public forums by being so much more vocal. This is one of the reasons I prefer caucuses to primaries. It's easy to believe that the rancor we see on television represents the general public, and voting in a primary does nothing to dispel that illusion. When you participate in a caucus, you see that the person supporting the other candidate is just a normal person with a different view, and not a faceless enemy who hates you.

The other thing I noticed last year was how horribly chaotic and confused the whole affair was. At my 2004 precinct caucus, no one anywhere near me knew what was going on, and the PCO who was supposed to lead us did little to help that. I don't know if it's the difference of being four years later or six miles further north, but this year was a marked contrast. At both the area and precinct level there were competent and informed people keeping things going smoothly. Not that there wasn't some room for improvement in terms of signage and crowd control, but on the whole it was an orderly procedure with people near at hand to answer any questions.

Promptly at 1:00 the area manager began to speak. She is the person who runs the portion of the meeting applying to all 12 precincts together, which mostly consists of reading through the script explaining what's going on plus asking for (voluntary) money contributions. The area manager had an assistant, who seemed to be even more experienced, moving around the room dealing with questions and occasionally popping up to whisper reminders in the ear of the area manager. This assistant was also a woman. In some blog forum or another I read a commenter's complaint that all the caucuses she sees on TV are run by men who are opposed to Hillary Clinton. My experience has been the opposite. In both of the caucuses I've attended here in Washington, women predominated among the leadership. This year I noticed that Clinton supporters are more strongly represented among the leadership (a bit more than half) than they are in the membership as a whole (roughly 3:1 for Obama in my area).

I assume there was an area manager at the caucus I attended in 2004 (also in a school cafeteria with about 12 precincts represented), but he or she didn't have a microphone and was not on my side of the room, so I neither heard nor saw him or her. This year, the area manager was prepared with a microphone, which was set up at one corner of the stage. By coincidence this was about three feet from where I had parked myself, so I spent this portion of the meeting essentially sitting at the area manager's feet. She explained the process for the many who didn't know what was going on, helped guide people to where they were supposed to be, and let them know what was going to happen and how they would participate.

Much of this was following a script prepared by the party, and I assume it was part of the script that the meeting begin with a pledge of allegiance. One of the things I observed last year is that the Republican caucus featured the pledge of allegiance while the Democratic one didn't. (But come to think of it, perhaps there was a pledge at the Democratic caucus in 2004 and I simply never heard it.) I have no particular love for the pledge of allegiance to the flag (as opposed to, say, swearing to uphold the constitution, which inspires me much more than the flag), but I have a distaste for the brand of liberalism that balks at it, so I was pleased to see that this room full of Democrats happily recited the pledge with no visible sign of rebellion or apprehension.

Since our area manager was efficient, she was done with everything required by the script by about 1:15, but according to the rules actual caucusing couldn't begin until 1:30, so the rest of the time was spent fielding questions from the floor, and since there weren't a whole lot of questions there was some general discussion about the local party activity generally.

Before this, near the end of the scripted part, she called for a volunteer to serve as secretary, adding parenthetically that the secretary doesn't really have to do anything except sign "this page," which she waved about, so it could be anyone with a pen. I was right there and I had a pen (with which I was taking the notes that I'm referring to in writing this now), so after waiting about 10 seconds during which no one else stepped forward, I said OK, I'll do it. And thus I became the secretary for the 12 precincts at that cafeteria. It turns out the area manager exaggerated a little. The page she had waved was a form for the minutes of the area part of the meeting. The secretary was supposed to record what happened at the meeting, and these minutes would be signed by both the secretary and the area manager. I suppose I could have just signed and it would have passed muster, but I took my duty seriously, so I scribbled out a few sentences stating that the area manager explained all the rules and took questions, with a few times and names for specificity.

Somewhere in the middle of the area meeting, my PCO came to me at the steps and asked if I would keep the tallies for our precinct and I said yes. I had made contact with the PCO -- the precinct committee officer, the person in charge of running the caucus for the precint -- a few days earlier. Or more precisely, I made contact with him a year and a half ago but I identified him as PCO a few days earlier. I suppose I know my current neighbors slightly better than I have my neighbors anywhere else I've lived in since I've been an adult, but that's not saying much. Among my neighbors the only ones I know even remotely well are Larry and Jeanne, who live twp houses down. Jeanne is the captain of the neighborhood blockwatch and she's moderately involved in local government. She knows our council members, local police chief, and that sort of thing; she keeps up with what's going on in the neighborhood and knows everyone on our two blocks. Come to think of it, I have to give her credit for striking the perfect balance. She's just nosy enough to know what's going on and keep you informed, but she has a good sense of leaving you alone and respecting your privacy.

Jeanne is an email person like me, so we occasionally trade emails about neighborhood stuff. Earlier in the week I had found a list of PCOs for my district online, and though the precincts weren't identified, when I spotted Larry's name I was not at all surprised. So I sent an email to Jeanne letting her know I'd be at the caucus and that, after the 2004 fiasco, I intended to come informed and prepared.

From 2004 I remember only two precinct officers, the PCO and a secretary -- and that time we elected our secretary right there at the caucus. I don't know if the rules changed or if we did it wrong in 2004 or what, but this year Jeanne had already been designated secretary, and she told me there's actually a third position for the person in charge of tallying the precincts votes. When Larry came to me on the steps he was asking me to come be that tally person.

So as soon as I was done with the area minutes, I came back to the precinct table. Someone offered me a seat (ah, the perks of office!), and I started gathering up the sign-in sheets. The way it works is that each participant in the caucus is supposed to write their name and address on a special page. The page has two spots for declaring a preferred candidate: the top one is for your choice when you come in, and the bottom is where you write your new choice if you change your mind during the caucus. If you don't have a choice (either before or after) you can sign as uncommitted.

This was not entirely understood by everyone. A few thought the second spot was supposed to be for their second choice and filled it in accordingly (with the result that their vote would be registered for that second choice). Others didn't fill in either spot, or made other various clerical errors in the signing in. I spent most of my time, both before and after the official caucusing began at 1:30, chasing people down and clarifying their sign-ins. Meanwhile, new people were continuing to trickle in. We got a proper ruling on this (from the area manager's assistant) that anyone who was there before 1:30 could sign in and vote. Anyone arriving after 1:30 should still sign the sheet and could participate in the discussions, but was not allowed to vote.

They were eager to be started, so I managed to get a tentative tally (which turned out to be correct) by 1:30. We had 53 voters in our precinct, with 15 for Clinton, 35 for Obama, and 3 uncommitted. No one voted for Edwards or any of the other candidates who had left the race, so the viability threshold was never an issue. The math was pretty simple this time. Our precinct was allotted five delegates to the district convention. Barring a significant number of declared voters changing their minds, our vote tally would give one delegate to Clinton, three to Obama, and the fifth to which ever side could get two out of three of the uncommitteds to join them.

In fact, one of the three who was officially uncommitted had written uncommitted on the top line and Obama on the bottom, which was going to count as an Obama vote if she didn't clarify with me. She told me to go ahead and count her as officially uncommitted for now but she was sure she'd go for Obama. I also knew that the formula was such that a split (ie, if one of the uncommitteds remained uncommitted) would also go to Obama, so the only way Clinton would get a second delegate is if both the other uncommitteds went to her. But the others didn't need to hear this level of detail, they were content in the knowledge that four delegates were determined and the fifth would be settled by their ability to sway those last three voters. I think many were rather excited about the idea of engaging in participatory democracy trying to win that last vote.

That set the agenda for the next 15 minutes. Larry had separated the Obama people from the Clinton people -- something that never happened at my caucus in 2004, but which I knew the rules specified. He also asked each group to discuss their arguments and strategy among themselves and select a spokesperson to present their case. I missed most of this, since I was still double-checking my tally, dealing with the late people, and generally getting my pages and forms in order. From what little I heard, I gather it was mostly all the individuals sharing their thoughts on their favored candidate -- more of a social function than a deliberative one, but that's a good thing. The two uncommitteds wandered back and forth between the groups, as did one other woman who was officially tallied for one side (I forget which) but was determined to listen to both and keep an open mind.

After a bit, the two representatives made their arguments. The Clinton person spoke first. They had chosen someone who was well informed but did not have a loud voice -- probably a tactical error, since it required us listeners to try to shush everyone and crowd closer in order to hear her. Then the Obama representative spoke. To my dismay, both of them made their candidate's "electability" a prominent argument. (I had heard "electability" mentioned among the Obama people earlier, but I kept my mouth shut about it. I was, and am, of the belief that Obama is no more or less electable than Clinton, but there was nothing to be gained by launching that debate there.)

In the end, all three of the uncommitted voters went to Obama, and the one unofficially uncommitted voter stayed in whichever camp she was originally in. I had already clarified the votes of those who had filled out the sheets wrong, so my tallies were now complete. We would send four delegates for Obama and one for Clinton.

The next task was to actually pick those delegates. The Clinton people were to elect one delegate and one alternate from among their group, and the Obama people were to elect four delegates and four alternates from among theirs. As has now been widely reported with regard to other state caucuses, these delegates aren't actually pledged to their candidate. If you are, for example, a Clinton supporter, part of your job in choosing the delegate to represent you at the district convention is to pick one who will stay true to Clinton and not change her mind. (Or, I suppose, if you yourself have some criteria for appropriate grounds for changing your mind, you should try to find a delegate who would do likewise.)

The printed instructions included the Party's official encouragement that the delegates chosen be diverse in terms of gender and race. (I don't recall if they mentioned other categorizations.) We were given the same instruction in 2004. It's a wishy-washy sort of instruction. There's no actual requirement or quota, and it's clear that you're supposed to vote for whoever you want, but the Party gives its push for diversity. Perhaps the language is the result of some compromise among those writing the rules. I saw this on the rules page that I had available to me. I don't recall hearing it said to the voters, but it's possible Larry read it to them when I wasn't listening. In any case, if diversity was a factor in anyone's voting I never noticed, but our end result was reasonably diverse anyway. In terms of gender, the delegates and alternates we chose were evenly split at every level. In terms of race we had one guy who was of some indeterminate multiracial color and another half-Asian, which I think is about the same proportion as our 85% white neighborhood.

Larry and Jeanne were in the Clinton group, so it fell to me to supervise the selection of the Obama delegates and alternates. Ours was the harder task, since we had four times the work to do. Larry had already specified that anyone who was interested in serving as a delegate should say so and then each candidate could give a one-minute presentation to the group. In our group six people, including me, volunteered themselves as possible delegates. A seventh said she'd be interested in being an alternate but not a delegate.

In the interest of hastening things along, I spoke first. I didn't say a whole lot. Even at the time I was aware that, by virtue of my relative visibility as tally person, I could most likely have gotten myself chosen as a delegate if I would just make a reasonably forceful case, which I could have easily done. But I was also well aware that the district convention was scheduled for the heart of tax season, and even though attending the convention sounded interesting to me, I knew it would be hard to spare the time from work. That left me ambivalent, so I made a brief and half-hearted statement expressing willingness to serve but no real enthusiasm. Of the others, two women gave very eloquent presentations; one guy was the one who had served as our spokesperson in trying to sway the uncommitteds, so he was already well known; and the last two, like me, didn't say much. (For those of you keeping score, the spokesperson was the person of color, and I was the half-Asian.)

Presentations made, I took a stack of Post-It notes and had everyone take a page, and I instructed them to write down four names from among the candidates (and we went through the group again with each of us stating our first name clearly). I'm still not 100% sure this was the correct procedure, to ask them to vote for four names each, as opposed to just one, but that's what made sense to me, so that's what we did. Afterward I asked Larry and Jeanne about it, and their response was, "Oh, I never even thought about that. We only had to pick one on our side."

I should note here that I was not dealing with all 38 who chose Obama. About half the voters left once the vote for the candidates was complete. The area manager and the PCOs (well, mine anyway) had done a good job of explaining what was the significance of each step in the process, and they made it clear that you don't have to stick around for the delegate selection if you don't want to. Since people had already been there a bit over an hour, many were ready to go home.

It was pretty obvious to all who would be three of our delegates. As expected, almost every ballot listed the three eloquent candidates, splitting votes on the fourth. I thought I was most likely to be the fourth, but I was wrong; one of the other guys edged me out by two votes. I still have my tally here. The vote was 17-17-16-10-8-3. I'm the 8, and for those who have done the arithmetic, yes, one voter chose only three candidates. As I was counting up names on the Post-It notes, some other guy materialized out of nowhere and was doing his own tally, which was helpful. Afterward he confirmed that his count matched mine.

I knew that we still needed to pick alternates, so before I was even done counting the votes, I announced over my shoulder that, assuming that the two unsuccessful delegate candidates would be interested in serving as alternates, and adding that other person who volunteered only as an alternate, we were still one candidate short of our four alternates. I further observed that if we were to have one more candidate and only one, that would hasten the voting procedure. They took the hint, and by the time I was done tallying the votes for delegate, a fourth candidate for alternate had presented herself. We now had exactly four candidates for the four alternate positions, so we chose to forgo the formality of voting on them and elected the four by acclamation.

The others dispersed and I was left with Larry and Jeanne to sign in all our delegates and alternates, and then wrap up the paperwork.

My position as alternate gains me admittance to the district convention on April 5 and also the county convention on April 12, I later discovered. Alas, I gave out false information at the caucus. At the time I thought the event scheduled for April 12 was the next step in the caucus procedure, and I specifically told one of the delegate candidates that she needn't attend on April 12 unless she was selected to move on at the April 5 caucus. A few days later I figured out that I was confusing the county convention with the caucuses in May which is based on U.S. legislative districts. For most of the state the county caucus is the second step, but here in the big counties near Seattle we use state legislative districts instead, and the county convention is a completely separate path. The delegates chosen here go on to the district caucus and the county convention. The county convention deals with other business of the Democratic Party but plays no part in choosing the presidential candidate.

I haven't completely ruled it out, but I doubt I'll attend. I'm told that it's not likely to take more than a couple of hours, but we're so busy at work that even a couple of hours is more than I can easily spare.

I'd have to check the rules, but I'm pretty sure that if we were not able to come up with a fourth alternate from the Obama group, the Clinton group would be entitled to fill in with an alternate from their side instead. Jeanne and I had a bit of debate by email about how alternate selection works, which we never entirely resolved.

I noticed that on the results page filled out for the Party, Larry listed the Clinton delegates on the top line (because they finished choosing first), and added the Obama delegates beneath. Another page of instructions had a formula for determining the order of listing. I forget how it worked -- something to do with the fractions left over when assigning number delegates to candidates, I think -- but the result was that we were supposed to list an Obama delegate first, then the Clinton delegate on the second line, and the remaining Obama delegates beneath that.

I think the reason this makes a difference is for priority of seating of alternates. If a chosen delegate is unable to participate in the next caucus, an alternate takes his or her place, but it's not automatically an alternate supporting the same candidate. Instead, priority is given to the first alternate on the list. If our list were filled out like I think it should have been, that would be a small advantage for Obama. With the way the list was actually filled out, it would be a small advantage for Clinton, assuming anyone tried to enforce it. (This sort of thing is one of those details they have in mind when they say the "results" announced from the caucus in Iowa or Texas or wherever isn't really a final result.)

A few days after the caucus I checked the rules on this, and I think I'm figuring it right. Jeanne thinks I'm mistaken. She emailed me a link to something that I never got around to reading, so maybe she's right and I'm wrong. She thinks that if the Clinton delegate is unavailable a Clinton alternate fills in, and if an Obama delegate is unavailable an Obama delegate fills in. Now I fully agree that this makes sense and it's how it ought to be, but by my reading of the rules that's not how it actually is.

Jeanne and I agree, though, that it makes little practical difference. So long as all the alternates agree who ought to be seated, they can make it so by simply having the other four make themselves unavailable, in which case the official seating priority doesn't matter. Not only that, but even if a "wrong" alternate is seated, he or she can still vote for the candidate of the missing delegate. For example, suppose that the Clinton delegate becomes unavailable, and for whatever reason I am the alternate chosen to replace her. Even though I was originally an Obama supporter, I could cast my vote for Clinton anyway if I felt that was fair.

As a matter of fact, in such a situation, I would do exactly that. If anyone asked, I would say that although I personally support Obama, I am at the district convention representing the 15 out of 53 voters in our precinct who chose Clinton; they shouldn't be disenfranchised just because their delegate had an unexpected family emergency (or whatever), so as their representative I'd cast my vote for Clinton. I would be perfectly comfortable explaining that to anyone at the convention, any of my Obama-loving friends, or any of the voters in my neighborhood. If, on the other hand, I made clever use of a technicality in the rules have our precinct cast five votes for Obama instead of the four that he earned, I would not enjoy explaining that to anyone. Maybe some people like that sort of "strategy", but I don't. Jeanne tells me that if the situation were reversed, Larry (who is the Clinton alternate) would do the same. Bottom line is that even though I do like Obama quite a bit more than I like Clinton, whatever tiny advantage I could squeeze for my candidate is worth less to me than fairly representing my neighbors.

I've seen some discussions about how easily caucus results can be cheated by clever people in influential positions. Having been in an influential position, albeit at a very small scale, I've given some thought to that. If I had wanted to cheat a little in favor of my candidate, how easily might I have done that? As the person filling out the papers, I may have had better opportunity than anyone in my precinct, possibly even more than the PCO.

I remember offering the pages to Jeanne to confirm my count. I think she did, but I'm not certain. It would have been pretty easy not to offer, in which case I suppose I might have nudged the numbers slightly in one direction. Particularly given that some of the votes were confused as they were originally entered, I might have selectively sought to correct the ones that I didn't like and leave be the ones I did like, even if I thought it wasn't the voter's intent. Beyond that, I suppose I could have blatantly falsified a vote, by just writing in names in the empty second box.

There's a practical limit to how far one can go with this when the supporters gather up together (literally, when they caucus). Supposing I were a Clinton supporter, I might have announced to the group that we had 25 for Obama and 25 for Clinton, rather than the actual 35-15, but then when they collected into groups, everyone would notice that the actual Obama group was twice the size and I'd be immediately caught out. But if I misrepresent the totals by two or three maybe I could get away with it.

Another route would be to give an honest answer during the caucus, but then falsify the numbers after everyone had left. The main obstacle to that would be the PCO and secretary, who would be hard to fool, but say for the sake of example that they're on the same side as me and we're all colluding. The problem with this approach is that the actual delegates were chosen during the process, and those individuals are expecting to go to the next caucus, so even if we cook the books afterward, we can't make those delegates disappear. Sure, we could trick the Party's official count long enough for the news to report it wrong, but when the district convention comes, those delegates are going to expect to be seated and if someone presents our falsified pages showing they aren't really delegates, it's going to be obvious that something is crooked.

I could explore this in more detail, but what it all comes down to is that by various subtle means someone in charge of a precinct caucus could pretty easily push the result by one delegate and one delegate only. But that could happen only if the actual result is reasonably close to begin with, and the only way you can push it is in the direction of the candidate who would otherwise come up just short.

I really don't see this as a flaw in the system. The difference of one delegate at the precinct level is tiny. It's likely to get evened out by similar differences in the other direction elsewhere. To whatever extent a net sum can be amassed in a single direction, it's still just statistical noise in the larger picture of the election.

This is not a popular idea. People want to cling to the fantasy of a perfect count, but anyone with experience in any sort of polling or auditing realizes there's no such thing. Any non-trivial election has a margin of error that is greater than zero. A better election process can reduce that margin but it won't make it go away. Attempts to reduce it yield diminishing returns, and at some point the pursuit of perfection isn't worth it.

As a practical matter, any election where the margin is less than one vote out of 10,000 is a statistical tie. In such a case, either result might be correct, and it's going to come down to random chance plus whatever small nudgings can be accomplished by officials in positions of influence. People hate that, but the answer is simple. If you don't want your election "stolen", as Florida in 2000 arguably was, simply win by more than 600 votes out of an electorate of six million. The only stealable election is one that is so close that there is no clear answer anyway.

10:54:46 PM  [permalink]  comment []