Friday 15 December 2017

Book Review - The Cement of Civil Society

A Twitter conversation has just made me realise I never published this book review on here. The proper version is available from the journal here. In sum - the book was so dull, I literally fell asleep on a train reading it. What the author managed to completely miss was that his analysis helped explain the decline of Labour in Glasgow and the rise of the SNP. 

The Cement of Civil Society: Studying Networks in Localities
Mario Diani
Cambridge, Cambridge University Press, 2015, 244+xxii pp, £64.99 (hbk)
ISBN: 9781107100008

Professor Marco Diani is a world-leading scholar on social movements. His book The Cement of Civil Society begins with the widely held proposition that the UK saw an unprecedented rise in protest activity in the 1990s, with growth in visible protest activity on the environment, animal rights, poverty, ethnicity and human rights, and peace. This also included the spread of protest to ‘unusual suspects’ (p.26). This change was paralleled by the growth of opportunities for civic groups to be involved in policy-making within the local state, especially from 1997 onwards.

By focusing on what is termed the voluntary and community sector within two UK cities – Glasgow and Bristol – Diani argues that his study offers particularly novel contributions to the study of local government and urban governance. However, the text very much emerges from a literature on social movements from political science and political sociology, a literature this reviewer is not familiar with. The key innovation is methodological, driven by theory. In his analysis Diani focuses on relational data – that is network data – about the voluntary and community sector, not aggregative data, arguing that: ‘this conceptual and methodological shift leads us to reframe some basic questions about the features of political activism, participation, and civil society in contemporary Western liberal democracies.’ (p.1)

The book begins by setting out a three-fold typology of the voluntary and community sector based on how strong their boundary work is, and the level of resource exchange they engage in: organizational, social movement and coalitional type organisations. The two-by-two grid that leads to this analysis also includes ‘Subcultural/Communitarian’ groups, but these are not a focus of the rest of the book. The second chapter sets out the methodology and mixed methods approaches and the case study choice, arguing that Glasgow had been traditionally dominated by one left-wing party (Labour at this time) and a history of class-based protest whereas Bristol has been more politically plural, with a civil society dominated by middle-class groups associated with new social movements, such as the environment. It should be noted that within each city, Diani only looks at voluntary and community sector organisations in one particular neighbourhood. Thus, Diani suggests, we should see substantive difference between the civil societies if we were to use an aggregative approach to highlight the novel insights of his relational approach.

Chapters three to eight set out the findings of his fieldwork, carried out between 2001-2, in a series of tables and short quotes from the qualitative fieldwork. Chapter three begins the relational analysis by looking at who organisations in both cities form alliances with and why, suggesting that they work with groups that are organisationally similar but who have a higher profile in civic society. Chapter four then conducts a network analysis of civil society in both areas, looking at resource exchange between organisations and social ties. This neatly suggested there were three different networks in both cities, with different levels of resource exchange and embeddedness. Chapter five succinctly relates the findings of chapter four to how organisations viewed themselves, finding congruence and shared protest repertoires among the networks. Chapter six then brings in evidence of engagement in local events to show a continuity across types of networks, protest repertoires and the sorts of events groups were involved with, and the events that link organisations. Chapter seven looks at the centrality of certain organisations in both cities’ networks, with a surprising finding that umbrella councils for the voluntary sector in both cases were not very central, but also attracted different groups in the different cities: coalitional groups in Glasgow and social movement groups in Bristol.

Chapter 8, on the links to local urban governance is probably of greatest interest to readers of this journal. However it falls short on methodological grounds outlined below. The data presented shows that most organisations had contact with the local authority in some way – which should not come as a surprise. The extent to which organisations engaged with public-private partnerships was very high, but this could be down to how they were defined in survey questions, rather than involvement in formal, contractual partnerships. The qualitative data echoes the findings of the broader work on partnerships from this era, with Diana concluding that ‘[e]ach group’s specific experience with council departments and/or partnerships seemed very much mediated by intervening factors such as quality of the civil servants concerned, or the nature of the issues addressed.’ (p.181) A good, and amusing, example of this was the close ties between peace protest groups and the police in Glasgow; the former had to rely on the latter to organise their disruptive protests including providing the police with a handy guide of how many people they expected to be arrested!

The theoretical and methodological innovations of this book will be invaluable to scholars of social movements and civil society in an urban context. As stated above, as a reviewer I am not one of these scholars, therefore the book has a number of weaknesses for a more general reader. Firstly, the data is now extremely dated. Writing from the perspective of Scotland, immense changes have occurred in the 13 years since these data were collected. The Scottish Socialist Party, who are so central to Diani’s analysis and went on to win six seats in the Scottish Parliament in the 2003 elections, are now a spent force in Scottish politics. Most obviously, the class-based politics focused on Labour has been replaced by an (arguably) class-based politics focused on the Scottish National Party, who have come to dominate politics in the west of Scotland. Thus, as studies of the two cities, Diani’s book is mainly a historical account, albeit fascinating at times because of this.

However, the analysis and theorisation offered by Diani does go some way to explain why such dramatic changes in political fortunes could occur so rapidly. The relational approach shows why these voluntary and community sector organisations are the cement of civil society because of the networks they are in. If a political party can successfully replace key nodes in these networks through working with these organisations – as the SNP and latterly the Yes referendum campaign in Scotland did – then a broader change in political outcomes is likely.

A second issue this reviewer has with the book is subjective and epistemological. Diani’s work is set within a tradition of quantitative political science and political sociology with its roots in the “normal” paradigms of North America. Here, it seems, that if a study does not include a regression model then it is not adequate as social science. I’m not fully qualified to comment on the adequacies of the statistical techniques Diani uses, yet for me when these are supported by the rich accounts of his participants the mixed methods really come alive. This is even admitted by Diani when he occasionally introduces a quote as making a point much clearer than the table of data that preceded it. While this reflects an unresolvable issue of epistemological difference, it is a shame that some more of the richness of the qualitative data does not come through in the analysis.

Finally, Diani argues that his relational approach offers new insights into local government and governance. There is no doubt that it does. However, as the discussion in the final chapter shows, suggests this is a product of research design and limitations, rather than intention. The social movement analysis the books sits in, judging by the bibliography, focuses on aggregative analyses of social movements at a national or even international level. As Diani admits in chapter two, the data for a relational analysis at a city-level would be too complex, let alone a national-level. The use of his analysis in specific neighbourhoods was thus a choice of convenience.

As such the analysis technique would be useful for people to replicate to understand the rich linkages between organisations in specific urban neighbourhoods. It would offer a richness of quantitative data to add empirical weight to what we already know about urban governance, and changes over the past 20 years. This type of work would also add to our knowledge the relationship between governance and social capital (c.f. Putnam). The insights of the book do not necessarily transform what we already know about the governance of urban contexts; rather it provides new empirical insights.


This is clear in the final chapter, which as stated draws lessons from the book for two more contemporary changes – the growth of online social networking and the wave of revolution that sped through the Middle-East. That these were national events, and international changes, testifies to the fact that this book speaks firmly to an international literature on national or global social movements. The book is therefore best suited to scholars interested in local government who wish to use its methodology to better understand the relations of governance. 

Is it a bit shit to be gay in the USA?

This is a blog post I should’ve written last week and posted on Monday. Oh well. I’m now wrapping up my small research project on LGBT+ housing and homelessness. I should’ve written this post last week as we launched the two reports on Monday – one for housing providers and one for homelessness service providers.

Our recommendations in both are pretty straight-forward, and should not come as a surprise to readers of my blog post – service providers should routinely ask service users their sexual and gender identity and get over their own cringe. In doing so, we would start to get decent data, but also begin a conversation with service users that is: “we are interested if you are LGBT+ because we realise it might matter to you”. One of the recommendations focusing on homelessness services might seem a bit odd though: we explicitly state that we don’t think LGBT+ specific provision, such as hostels or other supported accommodation, is required in the context of the lives of people who participated in our research in Central Scotland.

A lot of the lobbying for LGBT+ specific provision comes from two concerns. One is that LGBT+ homelessness is an enormous problem; we just have not found evidence for this. In fact, we recreated the methodology of The Albert Kennedy Trust, and surveyed homelessness services in Scotland. We got a very low response rate, and some really ropey data. If I were to make an estimate based on that, I’d say around five per cent of homeless people identify as LGBT+, compared to three per cent of whole population identifying as LGBT+.

Second is a presumption that the cause of homelessness in the case of LGBT+ is family rejection – that is, people come out as LGBT+ and then their families ask them to leave. We really did not find evidence of simple causation like this in our data. For example, two of our gender-queer participants had periods of homelessness because their families were not accepting of their gender identities, but their families were also emotionally abusive and this was just the latest example of this, so they had to leave the family home. In such complex cases, we cannot say for sure, but we could surmise that they would have ended up homeless because of leaving their abusive family whatever their gender identity. Similarly, another bisexual participant became homeless after their relationship with an abusive partner broke down and they started relationships with people of a different sex. Again, the causes of the homelessness are very complex here – we cannot say that the person was homeless because they were bisexual.

Because of these two reasons, we don’t think LGBT+ specific provision is suitable in a Scottish context. What is needed is better training among service providers to make the excellent current service provision more inclusive. 

Now, to get to the subject of this blog post – I also think that the drive for LGBT+ services comes from the USA (and to a lesser extent Canada) and, from what I’ve read among LGB studies, it looks like being LGBT+ in America is really bad. It really struck me when I was reading this paper that compares UK data to a wider literature review. That paper analyses data from the UK longitudinal panel study Understanding Society. It demonstrates that in some categories there is a small negative impact on your life from being LGBT+ in the UK. But the comparison data Uhrig pulls together from the US in particular, is far worse. To give one example that really shocked me: data from the US in 2013 showed that women with same-sex sexual attraction did far worse in terms of educational outcomes. In the UK, lesbians were three-times more likely to be educated than their heterosexual counterparts.

What’s going on then? Why do things seem to be worse in the US? I suspect there’s a lot of methodological things going on here. Firstly, data on sexual and gender identity is pretty poor everywhere, but it seems to be quite staggeringly bad in the US. The main source of data used by many researchers is the US Census, which has allowed same-sex couples to “out” themselves on their forms for a while and say they are a household. There’s three main problems with this: it misses single people, and we know LGBT+ people are more likely to be single; it doesn’t really allow for bisexual people to be recorded anywhere, and certainly ignores transgender people (but then, so do most surveys); and finally, it’s a self-selecting sample, from what I can gather, you don’t have to fill it in if you don’t want to. Whenever you create “prefer not to answer” categories in questions like this, you end up with that being your second biggest category after straight.

It seems there are very few population-level surveys which include LGB, or transgender, questions in the US. This means a lot of the US research that I’ve come across, for example focusing on homelessness, comes from a problem perspective and samples populations with problems, which as people like Prof Mark McCormack point out, leads you to find particularly troubling findings. To give one example, in my hunt for the source of that bloody 25 per cent stat (that a quarter of young homeless people identify as LGBT+), I discovered the root of one of the more bizarre versions – that a whopping 40 per cent of young homeless people identify as LGBT+ – comes from this report. You just have to read the subtitle to work out how they got that stat: funnily enough a lot of people who identify as LGBT+ use LGBT+ services. To be fair on the authors of that report, it seems that the stat has got mangled in translation.

The other methodological issue is the complex intersection of sexual identity and socio-economic status. I’ve only seen glimmers of this in what I’ve read, but I suspect middle class people in the UK feel more comfortable in their sexuality, which might explain why things don’t look too bad in our data.

However, I do wonder if there is something qualitatively different about the experience of LGBT+ in the USA, that it is a more socially conservative society. It certainly seems that social attitudes are marginally more conservative, with a small majority of people in the USA in 2014 still believing same-sex relationships were not “not wrong at all”. This compares to the UK, where in the most recent British Social Attitudes Survey, the vast majority of people think same-sex relationships are “not wrong at all”. More recent data from the US suggests that they are equally supportive of same-sex marriage as people in the UK are of same-sex relationships, however I would caution the conclusion that the US has become very socially liberal, not just because of the current POTUS, but because I think there’s some qualitatively different in the socially-sanctioned institution of marriage, and same-sex relationships more broadly.

As an aside, as I’ve commented before, I think these questions no longer collect accurate data due to social desirability bias, and we need to start asking some more, possibly more explicit questions, to get to the heart of peoples attitudes.


Finally, I think another issue is the lack of a decent welfare state in the US. For example, in terms of homelessness, one UK scholar commented that “[t]he sheer cruelty and vindictiveness of the US system, indeed, is sometimes difficult for Europeans to fathom”. To give one example, if you were a single, young, gay man in Scotland and your family kicked you because of your sexual identity, you would be unintentionally homeless, and your housing authority would have a duty to provide you with housing. I’m sure it’s not as simple as this, and housing authorities would try and wriggle out of it – one shocking example I read was of a housing authority in southern England who said a young man had made himself intentionally homeless because he chose to come out. But despite these cases, and despite the increases in homelessness and rough-sleeping in the UK over the past seven years, homelessness support is much better in the UK than it is in the US. And this cuts across a wider range of social and public services. If your welfare state is stronger, then if you come across a bump in your life, say due to exclusion related to your sexual or gender identity, then it is going to be easier to get your life back up-and-running again. So, I wonder if this is why outcomes do not seem to be as bad for LGBT+ people in the UK compared to the US, and it’s also why we don’t think LGBT+ specific provision is suitable in the UK. We have good mainstream services, we just have to make sure they are inclusive and supportive.