CUPE 3903 and the Failure of Top-Down Mobilization

Picket sign with slogan “Strike To Win!” (Photo by author)

As many of you may know, CUPE 3903, the union representing Teaching Assistants, Graduate Assistants, and Contract Faculty at York University (a union of which I am a member) recently concluded a successful strike. Our month-long strike ran from March 3rd until March 30th, and ended with the ratification of a contract that met virtually all of our major demands.

The strike was a tense and emotional event, and we all went through our share of ups and downs. With the strike over, I wanted to take advantage of this opportunity to reflect on what I feel were important moments in the months leading up to and during the strike. In particular, I want to discuss some of the divisions that have been present in 3903, and that I believe played an important role in shaping events surrounding the strike.

Although some of the disagreements and debates that have occurred between members on both sides of this divide over the past year have been personal and unproductive, the nature of these divisions are not personal. Rather, what we have are two fundamentally differing visions of how the union should function. On one hand, we have active components of the rank-and-file that believe in, and advocate for, a union that operates from the bottom-up and is driven by its membership. On the other hand, there are those who last year ran in the union’s elections on a slate called “Union Renewal,” and that have served on the executive during the past year, that seem to favour a more top-down approach to union organizing in which the leadership exists not to follow directions from the membership, but to direct them instead.

I want to talk a bit about these competing divisions and some of the ways in which the top-down approach of the slate members of the executive and their supporters (many of which are currently running for the Better Union slate) has manifested itself over the past year. I’ll do this by referring to three events that I feel best reflect their mode of organizing. The first are events that occurred during a December 2014 meeting held to decide on a timeline on the strike vote and potential strike, a meeting that was stacked by the executive in order to ensure that the vote went their way. I will then discuss how the slate-dominated executive’s mode of organizing failed during the strike, allowing membership-driven unionism to prevail and win the strike. Much more could be written about the internal politics of CUPE 3903, but I will leave the rest to be written by someone with a lot more time on their hands.

This analysis is largely based on my own experiences. My hope is that it will make a worthwhile contribution to the post-strike analysis of the events leading up to and during the strike of March 2015. Contributions and/or disagreements are more than welcome!

Strike Vote Timeline Debate and Top-Down Mobilization

The instance that I think best illustrates the 2014-15 slate executive’s top-down approach occurred on December 4, 2014 at a Special General Membership Meeting (SGMM). The meeting had been called with the specific purpose of discussing potential timelines for the upcoming strike mandate vote and potential subsequent strike dates. The executive had proposed one timeline, while rank-and-file members in the Bargaining Mobilization Committee (BMC) had proposed two other. The differences between the three were negligible, a matter of having the strike vote a week or two earlier, and the tone of the debate at the meeting reflected that. Proponents of each timeline presented their case and a debate was had. At several occasions during the meeting, proponents of the timelines that differed from the executive’s spoke at the mic, specifying that the differences were minor and that we should all be able to work with whatever timeline the members chose. Needless to say, although some members were passionate about their respective preferred timelines, this was not a particularly contentious meeting.

This began to change about 90 minutes into the meeting, right around the time that we were scheduled to vote on the motions. Almost all at once, 30 to 40 members entered the room. They registered, grabbed a slice of pizza, and when it came time to vote, they all voted (from what I could tell) the same way. That is to say that, without hearing any of the debate or discussion of the previous 90 minutes, they voted to support the timeline proposed by the executive committee. As a result, their timeline was accepted.

The tone of the meeting seemed to change as soon as the vote went through. Members began to become visibly upset and frustrated, not at the result of the vote, but at the manner in which it had occurred. Dozens of members entered virtually at the same time and then all voted the same way, only to leave en masse as soon as the vote was done (and long before the meeting was over). What had happened was painfully obvious to everyone in the room.

Members quickly approached the microphones to denounce was had clearly been a successful attempt by some of the executive members to stack the meeting with supporters in order to win the vote. The frustration in the room was not over the result of the vote, but at the manner in which it had been won. As was mentioned in discussions leading up to the vote, each timeline had its pros and cons, but the mood in the room seemed to be that each were generally workable. It was blatantly obvious that some members of the executive had rallied their friends to show up to the meeting at a specific time specified on an agenda that members were powerless to amend, and vote a particular way without hearing any of the discussion or debate. The party (or slate) machine had done its work.

It’s important to specify that none of the executive members denied what had happened. Instead, their reaction was to criticize rank-and-file members who denounced their strategy. One of the slate members of the executive told the room that the executive had simply done a better job of mobilizing members, and that if members were unhappy with it, they should just do a better job of mobilizing next time. Another member of the executive chastised those at the microphones for questioning the intentions and motivations of other members.

I didn’t speak at this meeting as there were plenty of other people to do so, but I did agree with the executive member who argued we should not question the intentions or motivations of individual members. Members of CUPE 3903 have every right to attend a meeting simply to vote, just as much as we have the right to attend a meeting to listen to debate and discussion for three hours. Rather, the focus should have been on the members of the executive who invited rank-and-file members to come to the meeting solely for the purpose of voting a particular way on a particular motion, without hearing any debate or discussion whatsoever on the issue.

One slate member of the executive referred to this as mobilization, but this was anything but. Stacking a meeting with supporters who arrive at a specified time in order to vote a specific way does not empower the membership to continue to work with the union, nor does it strengthen us in the long run. It is successful in winning votes and nothing more. It’s a top-down approach to mobilizing and organizing that in fact does nothing to mobilize rank-and-file members or encourage their initiative. In this context, it was a strategy meant to ensure that the executive’s timeline for the strike mandate vote, and strategy for mobilization remain in their hands, rather than in those of the BMC, where rank-and-file members had a say and a vote.

The March 9th Ratification Vote

The March 2nd vote on the employer’s “final offer” saw the Bargaining Team (BT), executive, and membership come together in a rare moment of unity. The BT and executive recommended that the university’s offer be rejected, and the vote by the membership echoed the same sentiment. Approximately 70 percent of members rejected it, meaning that we were on strike as of midnight on March 3rd. As the picket lines went up, the BT continued their work to obtain a deal that could be approved by the membership. Progress was made as the employer made some significant movement, so much so that, by Friday March 6th, the BT had announced that they had agreed to bring an offer to a ratification vote the following Monday, March 9th.

As details of the offer spread, the unity that had been achieved on March 2nd quickly dissolved. We had made some gains in the offer being brought to a vote, but it still fell far short of a number of the so-called “red lines” that the membership had agreed not to cross and return to work without. Specifically, members at the February 3rd SGMM on bargaining priorities had voted that we would not return to work without significant movement on four issues: tuition indexation language that would protect us from wage cuts implemented through tuition hikes, greater funding for Graduate Assistants (GAs) in Unit 3, LGBTQ equity language, and greater job security for contract faculty (Unit 2).

Unfortunately, the offer brought to a vote contained very little language on the first three, and very much on the fourth. Unit 3 members, who already lived well below the poverty line, were offered a small increase representing between $300 and $400 a year. On LGBTQ equity language, the employer presented us with a strange offer stating that that we would be given employment equity language, but only if it was given to another on-campus union first. Unit 2 was offered significant gains in terms of job security, namely an increase in the number of positions converted from contract to full-time, but this was most beneficial to a small number of high-seniority members. Finally, the issue of tuition indexation was addressed through a promise of a three-year tuition freeze during which time no current or incoming members would have their tuition increased. In other words, it offered us temporary protection from wage cuts through tuition increases, and increased the likelihood that we would have to continue the fight over these tuition increases in another three years.

In addition to falling short of our demands, the offer was concessionary. It was particularly so on the issue of tuition indexation. CUPE 3903 had secured tuition indexation language for its members during a strike in 2000-2001. It stated that, for every dollar that our tuition increased, our funding would increase by the same amount, thus protecting our wages from being cut through tuition increases. The language capped our tuition at 2005 levels. This protection was maintained in our contract until 2013 when York University reinterpreted it in a manner that allowed them to increase tuition for incoming students. Without warning, they went after international students, raising their tuition by $7,000 a year. This represented a huge loss for international students who already paid higher tuition fees than domestic students. The union filed a grievance and went into arbitration, but also decided to fight to regain this protection at the bargaining table.

Although our tuition indexation language was still in the collective agreement, and that we were offered a three-year tuition freeze, the offer was still concessionary. Although the previous language was still technically there, the protection it offered us was gone. We were proposing language that would unambiguously reinstate our interpretation of the tuition indexation clause. The absence of this sort of protection from the offer was what made it concessionary. If we were to accept this offer, we would in essence be abandoning our claim to the same level of protection we had benefited from until 2013. We would be replacing it with a temporary freeze that, when expired, would have in all likelihood either resulted in tuition increases/wage cuts for future members, or constant battles to extend the freeze every three years.

It was also clear to many members of the local that the offer was going to be divisive. It leaned so heavily in favour of Unit 2 that it was difficult to imagine that they would not accept it and, in the process, split the local. One concern was that, because it offered Unit 2 so much more than the other Units, it would create a situation in which contract faculty would return to work. The fact that York University had suspended all classes at the beginning of the strike gave the union enormous leverage. If Unit 2 members returned to work, there was a chance that York could attempt to reopen classes. The same thing had happened a week earlier at the University of Toronto, where the sessional lecturers in CUPE 3902 were offered, and accepted, significant gains, allowing the university to resume classes and invite TAs to scab. It seemed that York was attempting to borrow from UofT’s playbook.

Unfortunately, several members of the BT and executive failed to recognize this and recommended that the membership accept the concessionary offer. It is important, however, to note that the Unit 3 representatives on the BT recommended rejecting it, as did the non-slate members of the executive. Those who favoured the offer were by and far either slate members of the executive or allies of the slate on the BT. As the March 9th ratification vote meeting went ahead, emotions were extremely high. A large portion of the discussion had that evening consisted of members expressing their frustration with the fact that a concessionary offer had been brought to a vote. Their frustration that night was made clear and expressed very loudly. For better or for worse, the executive and BT were not let off easily.

In the end, the predictions of some of those who opposed the offer came true. Units 1 (teaching assistants, or TAs) and Unit 3 voted to reject the offer, while Unit 2 voted to accept it and return to work. Our local had been split, but the strike was not over.

The Collapse of Top-Down Mobilization

Despite having been told that the offer was the best that we would get from the employer, and that rejecting it could lead to a strike that could last for months, Units 1 and 3 displayed a willingness to continue to fight. Some of the more active segments of the rank-and-file, as well as members who had been motivated and inspired by the strike, also began to organize to mobilize for the continued strike. One of the ways in which they did so was by petitioning for weekly SGMMs in order to allow members to discuss and strategize around the strike. Despite having been directed to do so by the membership at the March 2nd meeting, the executive had failed to display any intention of organizing a meeting. Therefore members themselves, as stated in the by-laws, gathered enough signatures to hold an SGMM to discuss the strike. The first of these strike SGMMs was held on March 18th. It was at this meeting that the differences in strategies and visions for the union not only became most apparent, but that the top-down methods of mobilizing used by the slate members of the executive began to collapse.

The collapse of the executive’s top-down approach became most evident through the discussion and debate surrounding two motions presented from the floor. The first of these was a motion to overturn the executive’s decision to “pro-tem” (temporarily assign) two members to the BT. During the strike, two members, a Unit 3 representative and the Recording Secretary, resigned from the BT and had to be replaced (I’m not sure of the reasons for their resignations, and therefore can’t speak to them). Following the instructions of CUPE National, the executive decided to find two replacements and assign them to the BT on a temporary basis. They did so by sending, by email, a call for nominations for the two vacant positions at 7pm on March 16th, two days before the first strike SGMM. The next morning, at 8:30 am, two new members were selected by the executive and assigned to the BT, all of this barely 12 hours after the call out had been sent, and long before many people were even able to see it.

Several members took issue with the quick and un-transparent manner in which the two candidates were selected. Almost as soon as the meeting began, a motion was made to overturn the executive’s decision to pro-tem these two members in order to allow members at the meeting to nominate and vote for the candidates of their choice. The Chair of the meeting, a representative brought in from CUPE National specifically to preside over the meeting, ruled the motion out of order. He was challenged by members from the floor who won the vote, allowing the nomination and election of these two BT members to move forward.

Although there was some debate about the constitutionality of this vote in relation to our by-laws, the debate leading up to the vote was revealing of how the executive saw itself in relation to the membership. In particular, one slate member of the executive who defended the decision to assign the two members to the BT in the manner that it has occurred, stated that the executive was entitled to do so because they were elected by the membership and were therefore were better suited to represent the will of the membership. This statement was made in response to a claim by a member from the floor that allowing members at a meeting to nominate and elect the two new members of the BT was more transparent and representative of the will of the membership.

This exchange, I believe, is an important demonstration of the clash between two fundamentally opposing views of the union. The claim that decisions made by the executive better represent the will of the membership than the members assembled in a meeting is false in at least two important ways. First, numerically speaking, only a few hundred people participated in the elections that elected the executive. Although I’m not sure of the number of participants at that meeting, judging from the people in the room, I believe that it’s safe to say that they numbered in the hundreds.

Second, the claim that the executive could better represent the will of the membership than the membership itself is not only in direct contradiction of the spirit and letter of our by-laws (and of the principles of democracy), but also reveals quite a bit about the manner in which this particular slate member of the executive viewed the executive’s position in the local. As stated in the by-laws, the executive exists to act as the governing body of the local in-between General Membership Meetings, during which time the membership assembled in a meeting become the highest-decision making body in the union. The executive committee’s job is to serve the will of the membership and, according to the by-laws, “fully carry out the decisions and instructions formulated at membership meetings of the local,” not dictate the will of the membership to the membership itself.

The second contentious vote at that meeting was a motion from the floor directing the Bargaining Team to not alter the wording of our tuition indexation proposal without first bringing it to a vote at a General Membership Meeting. Although the Chair of the meeting initially ruled this motion out of order on the grounds that it tied the BT’s hands, preventing them from bargaining, he quickly rescinded his ruling once it was explained to him that this practice had been used in the past by the local, and that rather than tie the BT’s hands, it simply instructed them to bring any new proposals on this particular issue to the membership. The motion went forward and was approved by the membership.

Although there was some opposition to this motion from the floor, the greatest criticism came after the fact. Discussions on email listserves by close allies of the slate (and in some cases, current candidates on the 2015 “Better Union” slate, which is closely linked to the 2014 “Union Renewal” slate) characterized the motion as having tied the bargaining team’s hands. Others spoke of a “coup” by radicals and so-called “red liners” in the union who undermined the will of the membership. Attempts were also made to delegitimize both of the motions made at the March 18th SGMM by claiming that they only represented the will of a small segment of members rather than the will of the membership as a whole, a claim that was, to say the least, very problematic.

Although the motions passed at the March 18th SGMM did reverse the executive’s decision to pro-tem two particular members of the BT, they also allowed the membership to have a direct say in the composition of the group of people who would be speaking for them at the bargaining table. As for the motion to direct the BT to not change the language of the indexation proposal, contrary to what slate members and supporters argued, it did not tie the their hands in any way. Rather, it provided them with explicit instructions on what the membership wished to see in the next offer that would be brought to a vote. Instead of tying the BT’s hands, this motion empowered them to hold the line on tuition indexation against an employer that we were told would be unwilling to move on this issue. As a result, the membership provided the BT with a clearer idea of the type of offer that the they would accept, allowing everyone to avoid a third failed ratification vote. In other words, the will and desire of the membership was made clear.

These attempts by supporters of the slate to delegitimize decisions taken by members assembled in a SGMM were troubling to say the least. Picking and choosing which decisions are legitimate and which ones aren’t based on the number of people who participated in the vote is antithetical to the spirit of democracy. Needless to say that a greater participation rate is always preferable, but absence of full voter participation should not delegitimize any decision made in a membership meeting that has met quorum. The fact that this strategy was resorted to, in my opinion, is reflective of a level desperation and frustration at seeing the balance of power tip away from the executive and towards the membership. Although some people characterized this as a “coup,” it was anything but. Rather than steal power from the executive, what the March 18th SGMM did was simply return it to its proper place: in the hands of the membership. The greatest decision-making power in CUPE 3903 has always belonged to its members, and what we witnessed on March 18th was the membership reaffirming this fact.


The bargaining team, now fully composed of representatives elected by the members, was able to resume negotiations with the university while following the guidelines established by the membership at the March 18th SGMM. Mirroring the membership’s continued commitment to tuition indexation, the BT was able to hold the line against the employer and secure an offer that met virtually all of our so-called red lines. Our interpretation of the tuition indexation language was maintained and the $7,000 tuition increase/wage cut to international students was reversed. A promise was made to form a committee to explore and implement LGBTQ as an employment equity category in six months, and Unit 3 members secured a modest, but nonetheless significant, increase to their funding package. This was an unambiguous victory for the union and its rank-and-file.

The March 18th SGMM signalled the downfall of the slate’s top-down approach to mobilization and running the union. Although several of the local’s dedicated members had been accused by the slate and its supporters of being ‘radicals’ and ‘red liners’ for advocating a more membership-driven local, their strategies had in the end paid off. The so-called “final offer” rejected by Units 1 and 3 on March 9th had proven to be anything but final. The “coup” that they had orchestrated and the binds they had allegedly wrapped around the BT’s hands proved to be instruments that were ultimately crucial to winning the strike. These helped the BT focus on winning the strike rather than ending it as quickly as possible, as the Unit 3 member that had been pro-temmed by the executive on March 17th had suggested the union do (the same member, currently a candidate on the Better Union slate, was not elected to the BT at that SGMM). The fact that other decisions taken by slate members of the executive, such as their decision to withhold strike pay from certain members of the local, were overruled by rank-and-file members at the following week’s strike SGMM further demonstrated that control over the direction of the strike had been reclaimed by CUPE 3903’s rank-and-file membership.

The weakness of the organizing strategy employed by the slate members on the executive failed because it did very little to encourage long-term involvement of the membership in the affairs of the union. Although they had previously been able to organize members to come out to vote for them, as well as pack meetings with people who would vote a certain way, they failed to recognize that these strategies do not build a healthy union. They might help win elections and certain votes, but as one member of the Bargaining Mobilization Committee pointed out during the December 4th 2014 SGMM on the strike timeline, organizing members in this way does little to ensure their future involvement in the union. In other words, it’s a short-term strategy with little promise of long-term gain.

These flaws became evident after the March 9th ratification vote. At that point, the portions of the membership that the slate executive had relied on for support were nowhere to be seen. Rather than encourage them to attend meetings in order to increase rank-and-file involvement in the union, they were instead encouraged to vote in elections and attend meetings only when their vote was needed. Rather than working to build membership involvement in the union as a whole, whatever mobilization was done was done to meet the executive’s immediate goals. Those who argued for a more membership-driven union, however, did not have to rally people or stack meetings. Rather, they simply showed up as they had always done in the past. Added to their numbers were quite a few unfamiliar faces who may have previously not attended unions meetings but decided to do so, no doubt in large part due to the mobilizing nature of a strike. They were not rallied by any group of people to attend the meeting in order to vote a certain way. Rather, judging from the number of people who rejected the offers on March 2nd and March 9th, and who voted to overturn the executive’s decisions at the March 18th SGMM, they were there because they had taken an interest in their union and, to varying extents, shared the belief that the power to steer their union belonged in their hands, as members of CUPE 3903.

Picture of picket sign attached to barricade. Sign reads “No indexation = No agreement.” Photo courtesy of Twitter user @MurrayCooke10)

5 thoughts on “CUPE 3903 and the Failure of Top-Down Mobilization

  1. Just a minor correction: the BT that went back to the table after March 18 wasn’t fully elected by Members. At least one Member of that team had been acclaimed to a Unit 1 position, and this after someone who tried to run against them had their nomination ruled ineligible. This same acclaimed BT Member was still attempting to lower our tuition indexation demand on the day that we finally won it. Only minutes before, in fact, such a suggestion was made. They were, as you can guess, tightly aligned with the “Union Renewal” team….

  2. Given this, doesn’t it make sense for 3906 to leave CUPE and stop sending dues to these hacks at National. An independent militant and democratic local in the sector would be an example for others in the sector to follow. A multi-local union in the sector of this nature would be an example to workers in Ontario more generally. This seems doable in this sector especially where most members don’t give two shits about CUPE or any other of bureaucratic unions.

    1. I don’t know how independent or militant it was, but 3903 and most other University CUPE locals in Ontario weren’t always part of CUPE. A lot of them used to be part of the Canadian Union of Educational Workers (CUEW), but merged with CUPE in 1994. I don’t know the details of the merge though.

      But the way I see it, CUPE National is only part of the problem. For example, if 3903 left CUPE, we’d definitely be free to do more without having to deal with all of the headaches, especially related to things like the 8th Line business they imposed on us. But the thing is that our value and strength to ourselves and to the rest of the sector comes in large part from the fact that we are a membership-driven union that isn’t afraid to fight when it needs to. Our tradition of bottom-up mobilization of members is what makes us so strong. And right now, we have an executive and a slate of candidates that doesn’t seem to appreciate this, or opposes it outright. So even if we left CUPE, our tradition of being a membership-driven local that isn’t afraid to fight could still come under threat from our own members, by people who seem to prefer a more bureaucratic approach to running the union, as is happening now.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s