Some Thoughts about Democracy

It seems that everyone in UCU is in favour of UCU being democratic, or more democratic. Great! We even had a democracy commission which issued a report which is not an easy read esp. with respect to the state or nature of UCU’s “democratic structures” (i.e., what are they, how are they working out, are they fit for purpose). It mostly proposals for rule changes, but many don’t have a clear (to me, yet) motivation. E.g., why reduce the General Secretary’s term length? And add term limits? What’s the addressed problem?

(I’m not saying that there isn’t one, but there wasn’t a fact finding portion of the report afaict.)

I hope to delve into the report and structures at some point (but OMG, there’s so much…which is itself a problem), but in this post I want to articulate two aspects of democracy that I see at play in a lot of discussions and which are in tension. Some of that is visible in UCU structures as well.

Breath vs. Depth

Most democratic organisations are representative. That is, “the people” don’t directly exercise power or make decisions, but instead they elect representatives who are held accountable to a body of rules and to future electorates. A generally accepted and useful metric for democratic legitimacy of a representative or set of representatives is “breath”, that is, the proportion of the electorate that supports them. One of the fundamental critiques of first past the post (FPTP) elections for geographically determined representative seats is that it tends toward functionally minority accountable rule, esp. if we add in various forms of gerrymandering.

Now, geographical determined representatives does have a rational…the representative is supposed to have a special relationship to their area, typically to represent all of their constituents even if they were elected by a bare majority of the voters (and perhaps only a small plurality of the electorate and even smaller plurality of the population). Constituents can pester their representative and attempt moral suasion as well as simple persuasion. Of course, if a group of constituents can’t muster the votes to control the election, their power is limited.

Of course, pestering a representative about an issue is a good way to demonstrate your intensity about an issue. All other things being equal, in a case where the issue is life or death for one person and a matter of perfect indifference to another, we should weight the views of the person with a hugely significant interest over the person who’s indifferent. Some fundamental rights are strongly counter-majoritarian because a majority either has no interest or no legitimate interest in that issue.

Putting aside “real” interests and their weighting, we can look at intensity of activity as something worthy of respect by a democratic representative. The mere existence of non-compulsory elections bakes some of this into most elections: Those who are eligible to vote but who don’t exhibit less intensity of activity and thus have less influence on decisions. In the UK, members of a political party (i.e., people who pay the fees) have more decision making power in the party than those who are merely reliable voters. Similarly, activists (people who show up to meetings, canvas, and the like) tend to have greater informal influence and may have greater formal influence over decision making.

Activity is generally very unevenly distributed in any polity. Politics isn’t a major focus of lots of people and there are all sorts of barriers to participation. (See all the barriers to voting the US.)

Activity requirements can be strategically increased in all sorts of ways. We’ve all experienced meetings where decisions were made out of exhaustion that ran counter to the general feeling but that one rule-mongering dedicated person who seems to have an infinitely large bladder just wears the rest out.

We can, sometimes, raise the distribution of activity. That is we can increase the number of people who are engaged in a minimal way. But as a rule of thumb, broadening engagement requires dealing down the intensity of the engagement. You’ll get more people voting by mail than attending a meeting. On the other hand, in a (good) meeting, there can be more varied input into the decision making process and more evolution toward a common decision.

If you require a lot of upfront investment of effort to even get started in engaging, then you are going to have a tough time broadening participation. Barriers to entry can be more off-putting than barriers to sustained engagement.

This is why turnout in low-salience elections tends to be much lower than in high-salience elections.

UCU Elections

Let’s compare some recent UCU elections:

Year Post Turnout (votes) Votes won (first choice/total)
2019 General Secretary 20.5% (23,638) 11515/15214
2020 UK-Elected Members HE (5 posts!) 12.9% (10,516) 703/1663.55 1369/1731.34




Total: 4916/8250

2020 Vice-President 11.1% (13,559) 4771/7145

(I want to note that these results are a bit of a PITA to extract…I guess I should have hit the vote tallies first! but multiple PDF documents per election don’t make life easy.)

These are all for UK wide positions. In each case, the turn out is small, though nearly half for Vice-President. General NEC election do somewhat better but not hugely. Grady was affirmed (at some round) by 13% of the total membership. These 5 NEC members were affirmed (in some round) by 1.03% to 1.28% of the total membership.

For me, this is the starting point of concern about UCU democracy. I firmly believe that both breath and depth need to be accommodated for robust democracy, but this seems to be a big imbalance against breath, which means that we are vulnerable to some of the characteristic pathologies of depth (e.g., over amplification of small minority views).

It’s more striking when we consider that the union, in fact, represented all workers in its remit, not just members. The union negotiates national pay scales, pension issues, and other matters and local branches do local negotiations on all sorts of things (overlapping with, supporting, and supports by the national negotiations). There are ≈400,000 HE workers in the UK and I’m having a hell of a time finding FE stats: I’m getting between 87,725 and 111,000. It’s hard to imagine that these aren’t undercounts and it’s not clear what proportion are UCU eligible. If we take the total workforce to be 500,000 (to pick a round number), then Grady was affirmed by 3% of the people she represents and the others by less than half that.

So one of our democracy problems ranges from yikes to apocalyptic.

If members (or voting members) were a representative sample of the whole polity, then this would be less of a problem, but the mere fact that members and voting members generally have more intense views speaks against that.

Now, look! It’s not all doom and gloom. Getting broad participation is a problem for all democratic organisations of any size. Unions have faced all sort of organising barriers from decades of hostility from employers and governments. But I think it’s especially important when explicitly discussing democracy in the UCU to attend to this issue.


One thing I think can improve at least intra-membership participation are factions or other groupings. One barrier to participation is information misload: Voters often have too much and to little information. Election addresses are long and not typically info-dense (mine included!!!). Hustings are high effort to attend and typically even lower in information. This post is far too long for effect broad based communication. (It’s not intended to be, of course!)

Even understanding the significance of what we’re voting for is hard. The role of the NEC certainly wasn’t known by me or my friends. The vice-president is an odd slot (it’s a one year post chained to 3 more 1 year posts!). And so on. Packaging up this information is extremely difficult and absorbing it in its current form mores. Having a common platform (or set of values or…) as a faction can reduce the info absorption requirements. Similarly, slates allow voters to exploit trust they’ve build up about one person and apply it to others.

Right now, affiliation with a grouping is hidden in election addresses (at best) or other campaign materials. Which means a voter has to scan through these to find that information. Being able to state an affiliation in the actual ballot could help! Or having such information in the supplementary booklets (e.g., organise addresses by Independent, and then by organised grouping).


Making information that would help voters make informed decisions as easy to get and absorb as feasible could raise participation rates. It’s a good thing to do regardless and one of my NEC goals.


The UCU is structurally very complex and we have far too many elections. More elections favours intensity because each election has significant marginal costs for voters, if only in the information they must mobilise. More frequent elections in principle allows for faster accountability but tends to shrink the overall amount of accountability. Having to answer more often to a smaller range of voters seems worse than having to answer less often to a broader range. Obviously there’s a tipping point somewhere in there, but we have multiple elections and ballots each year at the national or semi-national level. The decisions have be super clear and the voting mechanism super easy to make that broad based.


More, too long, why did I write 🙂 I think this is a fairly good long form piece (though it easily could have been much much longer), but it probably is worthless as a general communicative document or election document. But if you’ve gotten this far, perhaps you enjoyed it. I enjoyed thinking things through, a bit.


Ouch. The title is really bad, even if I add “union” in. I have lots and lots of thoughts about democracy 🙂

Update encore

I didn’t include any examples of a depth preferring view but then I happened today to stumble upon a pretty clear example and is 1) published in a sort of publication and 2) an anonymised op-ed so I’m not targeting anyone in particular:

This is reminiscent of a previous era – when Sally Hunt was general secretary during the 2018 USS strikes and the HEC tried to force a national ballot on members in response to a badly flawed deal – and is a shift away from participatory democracy. Having asked branches to organise meetings in their institutions to decide whether to reject or accept the current offer, it seems that the national leadership has now decided to ignore the result of these meetings and take the question again to the whole membership via an eballot. While it may appear more democratic as, in theory, it involves everyone, it only does so in a vacuum where the debate is lost, members are isolated at home and the sense of collective action also disappears. A core principle of participatory trade unionism is collectivity: where members come together to make decisions and to decide on action and so the most democratic decision-making forums are branches. These disputes have been all about members working and arguing together over the way forward. The future of the dispute shouldn’t now be reduced to an atomised electronic vote in your own house removed from everyone else.

This…isn’t good. It’s one thing to argue that intensity (specifically, direct, perhaps deliberative? participation) is a democratic value to keep in mind and another to dismiss breath concerns entirely.

I’ll note that voting by branch (and these matters typically involve votes, afaict, not consensus) just is FPTP districted voting and subject to the pathologies therein made more intense and probable by requiring attendance and deliberation in narrow time frame (and often with one’s vote being non-secret). It’s hard to imagine that turnout for branch meetings remotely approaches turnout for the lowest turnout elections.

It’s strategically poor as well. I haven’t crossed any picket lines and don’t want to, but I historically didn’t go to branch meetings (pretty much none at all before maybe two years ago). I would have been much more grumpy about participating in a strike if I had seen the contempt for my vote expressed above. It’s hard to imagine how to broaden participation with action—the key to industrial action strength—if we tell people who choose not to or cannot participate in branch meetings that they don’t get a say. Moreover that their wish to have a say anyway is not democratic!

I’m really struck by this line:

A core principle of participatory trade unionism is collectivity: where members come together to make decisions and to decide on action and so the most democratic decision-making forums are branches. These disputes have been all about members working and arguing together over the way forward.

So, 1) that’s not what “collectivity” is…secret ballot voting is a form a collective decision making, 2) when did we shift from trade unionism to participatory trade unionism?, 3) you can’t just assert that unions are fundamentally one kind of democracy and then conclude the most democratic fora are the ones that conform to your assertion, and 4) sorry, but I think the 4 Fights were “all about” the content of the dispute, i.e., the things we are fighting for.

Let’s be clear here: This is as antidemocratic a sentiment as the tyranny of the majority. If you don’t want a democratic trade union, that’s your prerogative. But you should be upfront about it and the fact that you don’t care about the bulk of the membership or the overwhelming bulk of those we purport to represent. This is a bit overstated. Technically, it’s perfectly possible to care about everyone in the sector and even to (in some sense) their voice but also to think that only certain mechanisms of expressing that voice are valid. It’s dodgy, but possible.

Yet more

I stumbled upon this article which makes similar points. It felt familiar so perhaps I read it back in Sept when it came out?

They focus a bit on the role of UCU Left and their embrace of an intensity focus. This chimes with what I’ve observed and I certainly think that the NEC’s resistance to consult was v bad.

Interestingly, one of the authors, Adam Ozanne, thinks I shouldn’t presume to run for NEC because he doesn’t think I’ve sufficient local experience. I don’t know what local experience would be sufficient in his eyes.

Of course, one should vote one’s preferences. But this preference that I’ve encountered ie that lack of “significant” (but unspecified) branch experience is a bar (rather than one, perhaps strongly weighted factor).

Most of our members aren’t active in our local branches. Perhaps having some people, who otherwise are congenial, with a similar background, would help with those members’ representation.

If you don’t think I’d be a good NEC rep then you shouldn’t vote for me! But I’m gonna side eye your reason if it’s one, which if generalised, would be highly intensity oriented gatekeeping and work heavily against eg precarious people or other folks who might be unable to participate in local meetings.

I mean, do you really want to exclude someone like Ben Pope?

Another Update

One problem with writing such things is that it turns out it’s all been written before, multiple times, and rather well. Rachel Lara Cohen, in a piece laying out the frankly mind boggling complexity of the UCU also captures what I call the depth vs. breath aspects, though she thinks of them as conceptions of democracy rather than aspects of democracy:

The division between the IBL and UCU Left has been usefully characterisedas rooted in different understandings of democracy — plebiscitary versus participatory. Thus, while neither group is monolithic, nor do they vote as a bloc on all matters, they are most polarised with respect to expectations about member passivity or potential for activism. IBL members argue that there is a large separation between members, whom they see as largely passive, and those activists who attend local and national meetings. They have therefore repeatedly argued for e-ballots and other means of gauging the opinion of inactive members, while cautioning against relying on decisions made at meetings. They are cautious about taking industrial action, or even raising subscription fees for well-paid members, on the grounds that such actions might lose member support and result in members leaving the Union. This has been visible recently in the IBL-majority NEC’s extremely tentative approach to redressing UCU’s regressive subscription rates. Currently members earning £60,000 pay just 0.48% of annual income as subs, whereas those earning £10,000 pay nearly 1.37%. In response to pressure and a 2017 Congress motion from the Anti-Casualisation Committee the NEC has submitted a motion to 2018 Congress which will reduce that difference, but only marginally, leaving the difference at 0.51 versus 1.30 percent in favour of the higher paid.

UCU Left has a more dynamic understanding of members and activists, seeing less difference between these groups and arguing that members are likely to become more active as action occurs. Therefore, in contrast to the IBL, they suggest that if the Union engages in effective action more members will join. UCU Left typically argues that meetings involving debate and active participation, and in which views can be interrogated and challenged, are a cornerstone of union activity and that e-ballots are not a substitute, especially e-ballots in which questions are determined by the leadership and members given limited closed-response choices. Critically, because of their different understandings of the role of members/activists, IBL members are typically happier than UCU Left to assign day-to-day issues, and sometimes bigger ones, to the Union’s unelected, paid officials.

It is almost certainly the case that the wider membership…not to mention the general body of workers…is far more conservative than even the broadest conception of activist. I.e., the set of people who work in FE or HE are more conservative and are not UCU members, as a whole, the subset who are UCU members but don’t vote or participate who are more conservative, as a whole, than those who only vote on disputes and offers who are more conservative than those who vote in all elections who are more conservative to those who go to meetings or otherwise engage in UCU activity more than 1 hr/week on average. All major groupings are “on the left” in a way that understates how left we are.

I had a conversation with a colleague who is a quite militant, though not activist, UCU member and generally fairly left wing wherein I was describing some HE related NEC hustings and causally said, “They don’t give a lot of new info. All the candidates mentioned wanting to get rid of student fees, but everyone wants that.” They said, “I don’t! I don’t think they should be this high but I think students should have some skin in the game.”

Now I don’t agree with my friend and we had a good chat about it, but I hope this example shows how we can become acculturated to a set of views and even think they are representative (esp if we believe them to be correct and morally significant).

I don’t quite understand Cohen’s idea that the “participatory” perspective views there being less difference between members and activists…it seems to be that members can be converted into more active roles if e.g., more action happens. But by saying “consultative ballots” are inappropriate, they seem to be saying that if you aren’t active then you have no role in decision making. That consultative ballots can be manipulated in bad was is, indeed, a problem. But so is a small, comparatively homogenous group making decisions that the larger membership fundamentally opposes.

I don’t think we can eliminate the tension between breath and depth by picking one of them. Activists can and should push boundaries and, well, lead. The union isn’t just a summation function of a snapshot of the views of its members. But if the union core is too disjoined from the membership (and the polity) as a whole then it is isn’t representing them and will lose support.