Irrational politics: is it about signalling?

There is a puzzle regarding the Labour Party at present. The vast majority of new members, and of Momentum members, appear pleasant and altruistic in person, and their support for the current leader deeply-felt and principled. A political science friend experienced in Left politics recently said: ‘I have experienced both Militant and Momentum and I have to say they are very, very different beasts. Militant were incredibly well organised and Momentum just aren’t, nor do they possess any dogmatic ideology[;] in fact they are a little bit fluffy really’ [personal communication].

The moderate Labour activist Luke Akehurst seems to agree regarding the real-life civility of most new members:

‘There seem to be two Labour Parties at the moment… There’s the rather pleasant one I am a member of locally. Fundamentally it is the same Labour Party as before May, but with a load more members… The new members seem much like the old ones — friendly, idealistic, well-intentioned, good people. There is certainly none of the horribleness one sees on social media, just a lot of sincere debate about the kind of world and country people want to live in and a lot of hope invested in Jeremy Corbyn to bring change… the fundamental rhythms of the local party roll on. Good natured debate. Fun social events featuring a bit too much wine and a lot of good company. The annual quest to find candidates to fight the local elections. A lot of canvassing, with some dry jokes about national events between canvassers, but nothing sharp enough to annoy the people you are enjoying your Sunday morning stroll with a Labour sticker on with.

‘And then there’s the national party. Which is a bit like watching a horrible car crash, in very slow motion’.

The quasi-religious turn

There has undeniably been a surge in conspiracy thinking, more polarised thinking, and incivility at rallies and online, severely so in the latter case. It isn’t entirely self-evident why. We could put it down to ideology, although a substantial section of Corbyn supporters appear to be soft Left rather than illiberal Left. Many will disagree; Nick Cohen and others have written well about the sources of illiberalism in the Party.

What is also of note is how often the Party is discussed in religious terms, and so I want to look at the quasi-religious turn in terms of signalling, using models from the social science of religion. The historian Glen O’Hara recently attended a meeting of his CLP to nominate Corbyn or Smith, and referred in his report to a millenarian mood, to ‘deeply held faith’, to fervour, to ‘a deep attachment to one man as a transformative figure’, and the contribution of one woman who said ‘I’ve got hope. I’ve got something for my sons’.

My own observation of politically-inclined friends has been similar: that those involved with or sympathetic to Momentum are good, moral people. Among those less involved but generally sympathetic to Corbyn, there seems to be a concern with loyalty and authenticity — perhaps at times less out of affinity to ‘true Labour’ and more to deep loyalty to public service professions, particularly teaching and academia. My own position is, unexceptionally, as follows: first, Corbynism won’t work, not because it is too left-wing, but because it is wrong, to paraphrase Hopi Sen. Second, it is unelectable, as a plethora of polling data and canvassing reports demonstrate. And third, the Corbyn leadership was in a minor sense, but significantly so, responsible for Brexit. Even if we find campaign effects were very limited, the decision to run a homeopathic campaign signified misalignment with Labour policy as well as sympathy to what has been widely described as a catastrophic act of national self-harm.

I understand why comrades might disagree regarding the first, but intrigued by denial of the second and third — and particularly the third by pro-EU Corbynites, which doesn’t make sense (there are also the anti-marketeer Corbynites who are at least consistent). Indeed, ‘denialism’ is increasingly used to describe Corbynites, while religious language has often been used to describe Labour in the past: as ‘a broad church’; one which ‘owes more to Methodism than Marx’. Indeed, in an interview published yesterday in The Times, Alan Johnson described supporters for Corbyn in cult-like terms:

‘… like a religion, they follow Marx, Engels and Trotsky as if they are the Bible, but they were creatures of their time. If Marx was alive now he would have a completely different view, but they crystallise it into a faith. You have to be a true believer’.

And one of the most interesting summer comment pieces on Labour’s situation was made by the sociologist of religion Gordon Lynch, who wrote of his CLP meeting that support for Corbyn was

‘impervious to any critical reflection… I have never experienced an atmosphere like that Party meeting anywhere before outside of conservative religious groups who are deeply convinced of the truth of their way of seeing the world. The fact of the growing Party membership was, as in any committed Evangelical group, taken as confirmation of the moral rectitude of the movement, with no interest shown in whether connections with the wider electorate were being made…

‘Corbyn is a model puritan leader. It is a movement of moral certainty, largely devoid of policy, fuelled by symbolic struggles against evil, treachery and compromise’.

Why? First and simplest, they may ‘simply believe’. We could point to psychological explanations, and magical thinking as a response to stress — and we live in stressful times (Stavrova and Meckel 2016). There is evidence of soldiers in World War 1 carrying rabbits’ feet and otherwise displaying many examples of folk religion. We see it in professional sports where athletes exhibit strange rituals, or the levels of superstition shown by actors and other performers. Perhaps, but this feels an insufficient account of why the climate of the party has taken such a marked turn.

The club good theory of religions — and other social movements

Rational choice theory might also be useful to explain some of the peculiarities of what can be described as divergence from the mainstream. These features emerge to solve free rider problems. Some years ago, Laurence Iannaccone developed a club good theory of religion with wide applicability to other social movements. He sets aside the question of what religious activity is ‘for’, assuming instead that ‘religious activities provide utility in proportion to the scarce resources devoted to them. Religion is modeled as a club good that displays positive returns to “participatory crowding.” This approach yields insights relevant to both religious and secular institutions: churches, communes, fraternities, political parties, business firms, and possibly even families’ (Iannaccone 1992: 272).

He used this to explain why strict religions are strong: the more distinctive a church, the higher the attendance rate, with the less stringent requirements of Unitarianism and the Episcopal Church contrasted with Mormonism and the Jehovah’s Witnesses. In a later paper written with Eli Berman, he cited Charles Gide’s observation of early settler colonies in New England and elsewhere:

‘Another peril, and perhaps the gravest of all, lies in the fact that these [communal] colonies are threatened as much by success as by failure. For if they do not succeed it means misery, ruin, dispersal, and a general rush for safety. If, on the other hand they attain prosperity they attract a crowd of members who lack the enthusiasm and faith of the earlier ones and are attracted only by self-interest. Then there is a conflict between the older element and the new, and ultimately a demand is made for a sharing out, and each member goes his own way. We should not be surprised that these societies die young’ (Gide 1930, cited in Iannaccone and Berman 2006: 14 [preprint version]).

Drawing on the earlier paper, they summarise the apparently gratuitous costs demanded by many religions, involving sacrifice and inviting stigma: burnt offerings destroying valued resources; displaying a distinctive appearance that invites ridicule or scorn; focus on dietary and sexual prohibitions; restrictions on medicine and technology. Taken to further extreme, these render religions sectarian, with strict behavioural standards; dramatic conversions; high participation rates; resistance to social change; and lower-status or minority appeal: ‘These emerge as formal consequences of a high-cost/high-commitment strategy’ (Iannaccone and Berman 2006: 14).

Accordingly, odd proscriptions and stringent standards arise because the participatory benefits of group membership are also high for other members. Ideally, a group or movement would take account of those by internalising the benefits of participation, perhaps via subsidy; but this would require a great deal of monitoring including of unobservable behaviours. Alternatively, the less committed can be screened out via proscriptions and stigmatising behaviours, which achieves the same result.

Therefore, to take part in the movement and enjoy all the benefits the group provides, adherents must pay a ‘price’. Adherents are prepared to pay this given the level of participation engaged in by others, which enriches the activity. This framework accounts for religious extremism in particular, and may well be applicable to subcultures, including political subcultures, more broadly.

What they disavow is the idea that most members of ‘deviant “cults” and fundamentalist “sects” are victims of indoctrination’ (Iannaccone and Berman 2006: 9). Berman, also an economist of religion, suggests that we can interpret sects as ‘faith-based communities, dedicated to the production of communal goods… Leaning heavily on words like “angry” and “intolerant,” the standard accounts portray sect members as militant fanatics… The truth is far different and far less frightening… Sectarian movements flourish because they provide their members hope for the future, benefits for the present, and insurance against misfortune’ (Berman 2010: 17–18, italics added).

The economic approach is not highly intuitive. Many social scientists reject it because it doesn’t ‘feel’ right: people don’t pretend to believe to access group benefits; they truly believe. But intuition can be a bad guide. Economist Robin Hanson suggests that ultimately people care status and prestige, and with regard to social signalling, ‘a lot of our behavior is at root down to baser motives that we don’t like to acknowledge… part of what happens when we signal is that we self-deceive, and that’s part of the essence of the signal in the same sense is our ability to fool ourselves… [it’s] part of our ability to be noble and rise to the ideal levels that we sometimes hope we do is to not acknowledge how much we do this for reasons of glory and appearance and to be admired’ (Hanson 2008: online). Social interaction is thereby interpreted as costly signalling, often of loyalty.

As Iannaccone suggests, this framework could apply to many cases outside religion. He notes how demanding boot camp is, for example, as well as the exclusivity requirements of employers, those marrying, and political parties: indeed, the demand for exclusivity has confused some new Labour members who are sympathetic to the Greens.

Many of the particular characteristics of Labour members can therefore be interpreted as signals of commitment in addition to revealing information about the member’s position on the issue itself (for example regarding Palestine or Trident). Many of the cultural characteristics associated with different parties might well exist to provide a signal as to political type and extent of commitment. More obviously serving as purely signalling devices are encyclopaedic knowledge of Labour history, use of the term ‘comrade’, a self-descriptor as ‘democratic socialist’ by people who are ideological social democrats, and knowing all the words to the Red Flag and Internationale. In organisations with fragile leaderships might be added the expectation of public displays of loyalty, disavowal of scapegoats, and so on.

This model seems to come close to the Labour case, though it ultimately depends on what the party is for: to satisfy the members (as in the club good model, which doesn’t state further), or to win elections. For some it may well be all about the participatory benefits: ‘hope for the future, benefits for the present’. For others, social events raise funds, and help people feel bonded ahead of coming campaigns or serve to celebrate or wake old ones. It could also be that parties offer menus of different types of activity to sort members: those who turn up for socials on a Friday and to discuss the ‘new economy’ on a Saturday afternoon are distinguishable from those who turn up on weekend mornings at the Quadrant to go canvassing. Parties can encompass and indeed value both, but need to know which is which. And the social events can be used to make an environment more or less welcoming. A respondent to a qualitative study of civic engagement described her experience as follows:

Interviewer: Was there a noticeable change when you joined in the ’70s in the ward meetings to in the ’80s, like there is this debate going on with the local council — was it noticeable in ward meetings as well?

Respondent: I think they were getting a bit turned off by it to be honest with you — I mean I think they weren’t aware what the arguments really meant — I don’t think people like nasty fights anyway. When everybody got on well together and we all socially mixed [?missing] but when there is divisions and of course that was also the time when the Militant was trying very hard to […] and I saw the Militant in our Labour club when I first went there — it was full of militants — just me and one other used to attend and then me and another woman decided we were going to get rid of them and we did; and we ended up with a really good Labour club and good social events; and as you know the Labour club puts on good social events.

Interviewer: How did you get rid of them?

Respondent: Just by sheer being there and challenging their arguments — they always run off when you challenge them (Devine and Roberts 2005: Respondent int048).

Harnessing believers

But which type of participation ultimately counts? The gratification of members and election winning objectives are less tightly-coupled than they are for, say, Calvinists. Accordingly, those who seem to free-ride on the collective experience of political intimacy and millenarianism without ‘giving’ to it are often the foot-soldiers who get the vote out, or the local councillors who commit so much of their free time. Equally, the party ‘doers’ may well view those primarily motivated by communal belief as free-riding on the local party infrastructure and unbroken party heritage, which ultimately rests on whether it is a serious and electable force. Indeed, part of the skill of party organisers is to harness the true believers and turn them to good purpose. Healey noted in 1952 that it was lucky that early socialism in Britain had a local focus: ‘this sort of [national] parochialism was its greatest strength. The Fabians found socialism wandering aimlessly in cloud-cuckoo-land and set it working on the gas and water problems of the nearest town or village. The modern Welfare State is their monument’ (Healey in Diamond 2015: 91–106).

Because parties have multiple functions, the club good account may not be enough in its focus on participatory benefits, though it does help explain why people may avow irrational beliefs for rational reasons. Why parties suddenly become more extreme, or what limits their move away from the centre, is less obvious. Berman suggests that major shifts in commitment arise when participation is somehow subsidised, giving the example of years spent in Yeshiva by the ultra-Orthodox in Israel (Berman 2000). Perhaps the very low cost of becoming a registered supporter in 2015 — or even the negligible cost of online participation — fits this account.

To be fair, populism also arises from crisis in social democracy and market liberalism alike. More extreme members and supporters may be sending meaningful signals regarding the housing crisis and labour market weakness, particularly if they are young (and Corbyn is relatively more popular, or at least less unpopular, among those aged 18 to 24), though why this should take such an authoritarian tinge is perhaps better-explained by the club model.

Additional insights from both political science and moral psychology are also useful. In a well-known political science paper, Snyder and Ting examine parties as informational devices, and the interactions between parties, candidates, and voters. They note that spatial models of party competition are predicated on the notion that parties are free to move ideologically; in that regard they are not like special interest groups, which are constrained. They examine the conditions under which parties move direction, and don’t follow the Downsian logic of offering a program appealing to the median voter. They look at this in terms of signalling and screening: parties partly serve a function of screening candidates and signalling their preferences. Their effectiveness at doing this affects positioning. Parties that fail to screen out more extreme candidates end up with positions further from the median voter, leading to polarisation:

‘[p]latforms diverge… when the cost of party membership is low or parties have only weak screening technologies. Each party stakes out an extreme position in order to reduce the ideological heterogeneity of its membership, and thereby make its label more meaningful to voters’ (Snyder and Ting 2002: 4 [preprint version]).

By contrast, parties which screen effectively enough to achieve ideological uniformity will fail to win hot races. This paper is very well-known, and is particularly interesting in that it does not rely on a distinction between competence and ideology, one often drawn in political discourse (although also see Bernhardt et al. [2011] which explicitly examines valence and ideology as separate phenomena, and the conditions under which they are more or less correlated).

In other words, parties are effective to the extent that they screen candidates. This model doesn’t account for the ordinary members on which the Iannaccone and Berman models of participation focus. Nevertheless, they do discuss the consequences of parties seeking to maximise the net benefits of their members rather than offices: in that case, we are more likely to see divergence from the median voter. Surplus-maximizing parties care about the variance of the distribution of their members’ ideal points, and parties gain more from choosing relatively extreme positions which attract candidates closer to the platform and who value membership highly (Snyder and Ting 2002: 21–22).

Moral psychology and irrationality

Finally, psychological accounts of irrationality draw attention to our own moral intuition. Cognitive psychologist Paul Bloom has just written on how we should think about irrationality. He argues that it is more respectful to assume that the politically-irrational ‘think just like we do, but have different goals, background beliefs, and priorities’. Variation in rationality may well occur, but less because of personality traits and more because of differences in social context. He also draws attention to the low cost of political engagement driving polarisation:

‘If you want to see people at their stupidest, check out national politics, which is replete with us-vs.-them dynamics and virtue signaling, and where the cost of having silly views is harmless’.

He suggests:

‘if you’re curious about people’s capacity for reasoning, don’t look at cases where being correct doesn’t matter and where it’s all about affiliation. Rather, look at how people cope in everyday life [where stakes are high]… [l]ook at the discussions that adults have over whether to buy a house or where to send their kids to school, or consider the social negotiations that occur among friends deciding where to go for dinner, planning a hike, or figuring out how to help someone who just had a baby’.

Indeed, Bloom’s final recommendation perhaps also explains why the public does focus on the family history and personal choices of leaders. School choice in particular is an apparently (if not in reality) high-stakes decision for each parent, and one to which the British public is highly attuned: hence why the Thick of It writers made use of it to illustrate the haplessness of Nicola Murray. Accordingly, judge other members and leadership alike by their high-stakes and personally-costly decisions. For the individual member, an individual vote or clicking ‘like’ is not very costly. For the leadership, though, all signals will be treated as costly. To judge them is not to be disloyal, but to interpret. Lynch suggests however that dialogue with true believers will be pointless: ‘it merely re-energises their sense of being engaged in a grand moral drama, struggling against forces of darkness within and beyond their movement’.

Taken together, Bloom’s recommendation and the Iannaccone/Berman framework suggest as follows: people should be understood as having good reason for their apparent irrationality; the term ‘cult’ is unhelpful (but ‘sect’ probably is helpful); we should look carefully at what people do as well as their low-cost signalling; and tribal divisions can ultimately be resolved. But the impressions of O’Hara and Lynch suggests that a large section of the current membership simply does not want a party which is competent. This is perceived as ‘managerialist’ (as if good management is a bad thing): O’Hara observed members’ antipathy to ‘media management, spin and sterility’; Lynch a member ‘fed up with a Party that’s just concerned with winning’.

Others are buckling down for a long-term project of turning or out-recruiting the other side, and consider the real tragedy not that the battle will be lost, but that it will take such a very long time. Notably, a number of political opponents consider the withdrawal of Labour from effective opposition and electoral competition to be a tragedy for the public, one described by The Economist yesterday as ‘dangerous’.

We will have to see whether the ultimate result is a wholly explicable death of Labour England, or whether there will eventually be a return to focus on electability. If the latter, this will not simply be down to new members ‘losing faith’, but also because of the deep vein of pragmatism in British civic life which Lynch rightly identifies. Losing is tiring. New members will either peel off, or they will turn into long-standing members learning the art of the possible. That may sound managerialist, and many activists have been turned off by ‘neoliberal consensus’ and the possible as too small a space. Looking at the consequences of large-scale political failure, though, perhaps it’s a little underrated.


Luke Akehurst, ‘It Feels Like There are Two Labour Parties — But Not in the Way You Think’, 12 January 2016,,

Eli Berman, Radical, Religious and Violent: The New Economics of Terrorism (MIT Press, 2010).

Eli Berman, ‘Sect, Subsidy and Sacrifice: An Economist’s View of Ultra-Orthodox Jews’, Quarterly Journal of Economics (2000), 115/3: 905–953, doi: 10.1162/003355300554944

Dan Bernhardt, Odilon Camara and Francesco Squintani, ‘Competence and Ideology’, The Review of Economic Studies, 78/2 (2011): 487–522

Paul Bloom, ‘The Irrational Idea That Humans Are Mostly Irrational: People are smarter than some psychologists would have us think’, The Atlantic, 16 September 2016,

Fiona Devine and John Michael Roberts, Qualitative Study of Democracy and Participation in Britain, 1925–2003. [data collection]. 2nd Edition, 2005. UK Data Service. SN: 5017,

Patrick Diamond (ed.), New Labour’s Old Roots. Revisionist Thinkers in Labour’s History (Andrews UK, 2015).

Charles Gide, Communist and Cooperative Colonies (1930; reprinted New York: AMS Press, 1974).

The Economist, ‘Britain’s One-Party State: Labour’s Implosion Leaves Britain without a Functioning Opposition. That is More Dangerous than Many Realise’, 17 September 2016,

Robin Hanson, ‘Hanson on Signalling: EconTalk Episode with Robin Hanson’, 26 May 2008,

Laurence R. Iannaccone, ‘Sacrifice and Stigma: Reducing Free-riding in Cults, Communes, and Other Collectives’, Journal of Political Economy (1992), 100/2: 271–291, doi: 10.1086/261818

Laurence R. Iannaccone and Eli Berman, ‘Religious Extremism: the Good, the Bad and the Deadly’, Public Choice (2006), 128: 109–129, doi: doi:10.1007/s11127–006–9047–7, preprint version available

Gordon Lynch, ‘Corbyn and the New Political Puritans’, 12 July 2016,

Glen O’Hara, ‘What I Learned at My Labour Nomination Meeting’, 12 August 2016,

James M. Snyder and Michael M. Ting, ‘An Informational Rationale for Political Parties’, American Journal of Political Science, 46/1 (2002): 90–110, preprint version:

Olga Stavrova and Angela Meckel, ‘The Role of Magical Thinking in Forecasting the Future’, British Journal of Psychology (2016), doi: 10.1111/bjop.12187

I research in the social science of culture and religion, moral communities and civic engagement. QStep lecturer at U. Bristol.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store