wbai.net Pacifica/WBAI history   events   links   archive   bylaws etc   bylaws revision
PNB   LSB   elections   contact info   opinion   search

Critique of the Constituency Model

From: Larry Romsted
Date: Tue Oct 8, 2002 1:42 pm
Subject: Critique of the Constituency Model by Eric Romsted


My son, Eric Romsted, has been thinking about the Constituency Model and its flaws and has drafted a critique (see below). I like the Concluding Remarks in particular. May be relevant to some of discussion at the WBAI Bylaws Subcommittee Meeting tonight.

You will need a copy of the constituency model in hand to follow some of the details which you can find by going to the Web link given below.

Larry Romsted

A point by point critique of the "Constituency Model / Model of Inclusion" as presented in the "WBAI Unity Caucus Proposal; Pre-Bylaws Draft (9/18/02)" by Eric Romsted

You can find the proposal at:
My comments will not make complete sense without first reading the proposal.

I would first like to congratulate the authors on the work that they put in to writing this proposal. Although you will find me harshly critical, I appreciate the discussion which this clearly written proposal allows.

Membership and Voter Registration:

1 I won't weigh in on the long debated question of membership eligibility as I see little harm coming from either a totally open membership (as proposed here) or a sponsorship requirement with an easy-to-use hardship waver option.

2 The requirement to re-register to vote at every election would add significantly to the cost in labor to hold the elections. It would also likely lower voter participation by adding another hoop to jump through. There is no reason why any member in good standing should not simply receive a ballot at each election. Although the listener survey information would be valuable, it does not outweigh the extra work required. In addition the ability to switch one's constituencies each election would make for some complex electioneering as different constituency seats come up in the staggering (See Staggering below)

3 "The constituency categories" These are relatively reasonable selections as oppressed groups, and it is clear that a board which did not include any people from these groups would be a fatally flawed as a Pacifica station board. However, any attempt to label these groups ahead of time is to assume the voters' priorities for them instead of letting them decide. I will conclude with a few thought experiments to make this point clear. National Board: I am not going to deal here with the national board proposals as the composition and election of the local board is the main focus of this document. Local Board Size & Composition: 1 Candidates for representation and voters have some odd choices here. If candidates from the various "category one" constituencies all run in one of the "category one" slots they would leave all of the "category two" slots half of listener elected slots open for the only people remaining, straight-white-older-middle-class-men. If they choose to run in the "category two" slots they may face more opponents and they will make the category one elections less meaningful as those voters have fewer choices. Voters have a similar problem in choosing in which constituency to cast their votes. And things only get worse if candidates and voters from a particular constituency make different bets on where it is best to run/place their votes. More on this in the conclusion.

2 the size of the proposed board, at 36 members is unwieldy for a body that will mainly be involved in matters of practical station oversight.

Pre-Election "In-Reach"

1- Why this is called "in-reach" and not outreach is beyond me. There is no real problem here and it is important that the stations commit real resources to encourage as much and as thoughtful participation as possible. The most important resource being station air-time. One note of caution station resources should be equally available to all candidates and should not be used in such a way as to bias the election.

Nomination of Station Board Representatives:

1- Petition requirements make some sense in general elections where eligible voters are fairly easy to find and there is some reason to test the candidate's viability to stand for election. However, in the case of Pacifica it will be much harder to independently find 50 registered voters to sign a petition. Would prospective candidates be giving lists of the people in their constituencies to contact for petition gathering? If not how are they to find them? Already formed listener groups would have a disproportionate advantage here, as there seems to be no prohibition on signing multiple petitions.

2 There is also the question of who will do the work of maintaining the registration logs and checking all of the petitions against those records. This would be a significant amount of work and expense.

3- I understand the desire to make sure candidates are serious and viable before they are put on the ballot. However, this bar seems to me too high. I would suggest a lower number of required signatures and a clear way to give all prospective candidates the tools needed to find registered voters.

4 The "Big Tent" idea is generally wonderful. Direct democracy is such a rare thing in today's world. However, the more direct the democracy, the more work it requires of the organizers and the participants. The only problem I have with the "Big Tent" meeting is with the official endorsements (see below).

Election of Station Board Representatives:

1 As I understand the proposal, the "Big Tent" meeting is part of the official election process. As such it would be organized by the official Election Board and its expenses would be covered by the station. We don't really know how many people will attend this meeting, but we can be fairly sure that it will be a relatively small part of the electorate. However, any form of official endorsement will carry disproportionate weight arriving in peoples' election packets. This would be an unfair bias, and I fear would put too much power into the hands of the Election Board as the organizers of the meeting. What we need is significant airtime discussion of all of the candidates to reach as many listeners as possible, not a small discussion at a meeting used to officially endorse candidates. Of course it is perfectly legitimate for independent organizations to hold meetings and endorse candidates.

2- The registration and nomination process ads a peculiar twist to the election As I understand it registration will be ongoing through the nominations process and for one month afterward, during the campaigning. But the character of one's registration choices changes over this period. People who register early will do so without knowledge of the candidates but will be able to sign nomination petitions. As nomination progresses, and after they are complete, people will be able to choose their constituencies based on the nominated candidates. Someone who registers late could choose not to vote in a constituency available to her because that race is uncontested or because she has no preference over the available candidates, preferring to save her votes for contested races. A voting process which so changes the meaning of one's vote based on date of registration seems to me seriously flawed.

3 Requiring that people vote in person is an undue burden that will only serve to lower participation. I can think of several possible rationales for in-person voting, none of them compelling: A) Voting in person brings a greater attachment to the democratic process and a greater commitment to the election and the station. B) Under the constituency model it is actually logistically easier to distribute the various color-coded ballots and keep track of who voted in person than to organize it all into a mailing. C) To police peoples' constituency choices as they come in to vote.

"A" a may be true, but with an electorate spread across three states, setting up accessible polling places will be nearly impossible. The burden will be greatest on poor and working class voters outside of NYC, where public transportation becomes far more inefficient and the time required to go and vote will be much greater.

"B" may also be true. I hate to think of the work required in stuffing all of the envelopes with all of the right color-coded ballots and then dealing with the inevitable mistakes when some people receive the wrong ones. But in person voting has its own raft of problems; Are people going to be assigned to certain polling places? If not how do we keep people from voting twice, in different locations? If so who is going to make the assignments and what do we do if people show up in the wrong place? Who is going staff these polling for three weeks and at hours that allow working people to go out and vote? And who is going to train all of the poll monitors to make sure this complicated voting process actually works?

I truly hope that "C" isn't part of the rationale, because it is, in the end, impossible. Although membership in some constituencies is visible to the naked eye, for most it is not. And even in racial categories there is so much variation that the disputes raised by poll watchers trying to do this police work would be endless.

Election Board/Voting Methods/Vote Tallying Methods/Term Limits No real problems here that are not addressed elsewhere.


1-This proposal would entirely undermine to the process of democratization. If accepted, it could mean a majority unelected board for the first and possibly several years. And if accepted by all five station boards, it could mean an entirely unelected national board for even longer (if the local boards continue to nominate unelected members to the national board even as more local seats come up for election). This is completely unacceptable. Although there is some use to continuity and the past service of the current board is recognized, the goal at hand is to create an elected board. The one concession I would make would be to exempt current members from the term limits provisions; i.e. current members could be elected to two new terms regardless of past service.


1-the multiplicity of categories in the constituency model turns the rather simple notion of staggered elections into a real mess. One of the strangest effect is that people could strategically change their constituency identifications as the different seats come up for election. This would allow people to vote in more elections, but only if it occurred to them to do it.

Bylaws Convention:

1- This is a very good and necessary idea. However, 1 year (i.e. one election cycle) is not enough time to assess the process. This would be especially true if grandfathering were allowed, as many seats would remain unelected. 3 to 5 years seems more appropriate to me.

Concluding Remarks:

In reading earlier works describing the constituency model I thought I saw two main ideas motivating the work. First was a proposition that there are well defined groups in society (based on shared oppression and marginalization) and that it is vital that the governance board of a progressive, dare I say, radical radio station include representatives from those groups. Second was a proposition that to be a representative of a community one must be in a vital way a part of that community. That the community must come together to nominate candidates and then must amongst themselves vote for their own representatives. This need for this direct procedural link between representative and community is why attaching quotas to an open 1 person-1 vote election is insufficient to satisfy the first proposition.

I think the first proposition is correct in that there are definitely groups with shared interests and ideas based on shared oppression. However, I do not believe those groups are well defined. The edges are always fuzzy and individuals felt attachments to various groups will vary over time, place and the issue under discussion. Therefore, although it is clear that a board which consistently lacked members of one or many oppressed groups would be flawed, to attempt to firmly lay those groups out in advance is dangerous.

The second proposition is more difficult. I am attracted to the direct democracy principal within it, but given my reservations on trying to define communities ahead of time I could never make up my mind. However, this vision of a direct connection between constituency and candidate has been so blurred by the complex process in this latest version of the model that it is no longer at issue. Voters are forced to pick only two constituencies to vote with. This ensures that the constituency representatives are elected by only a fraction of their possible constituencies. And the candidates are forced to play an odd strategic game in choosing which seat to run for. I would argue that the single-transferable vote model actually provides a more direct link between voter groups and their representatives.

I'll conclude with two thought experiments to illustrate the above points.

First, imagine an election of a 25 member board (I did say the board should be smaller and it makes the math easier). And lets imagine that the voting pool is 40% African-American. And finally lets assume that race is the number one concern of the African-American voters.

Under Single-Transferable vote if all of the African-American voters place all of the African-American candidates at the top of their voter preference lists they are guaranteed to elect 10 African-American candidates (40% of the board). Of course there must be 10 African-American candidates running (fielding candidates is a separate problem). Also, not all African-Americans might pick race as their top priority. But if that is the case, why should we try to pre-define them into an African-American constituency.

Second, lets again imagine a 25 person board. And lets imagine that the voting pool is 20% Latino. This would mean about 10% of the voters would be Latina women. This would be more than enough (only 8% is required) to elect 2 Latina woman representatives under singe-transferable vote.

However, that same group of Latina women voter would be hard pressed to ensure the election of those representatives under the constituency model, even with the much larger proposed board. First the candidates would have to decide in which constituencies to run for office. If they choose the Latino constituency, there are only two seats total and they might find it difficult to win both and could very well lose both. If one chooses the Women constituency she will likely face a majority of non-Latina voters. And in any other constituency the Latina vote will only be more diluted. Then the Latina voters will face the difficult decision of which constituency to vote in. If they register early they may choose wrong and find that the Latina candidates have chosen to run elsewhere. If they register late, they give up the ability to nominate candidates and are still not guaranteed that their preference will be counted.

top of page | bylaws revisions process info page | governance proposals | bylaws etc | home