We took stock of the information and evidence we were gathering about ‘severe and multiple disadvantage’, and began to have misgivings about the way we’d framed our work.

We began to explore more fundamentally how knowledge about severe social harm is created, interpreted and used; by whom; using which methods; for what purpose; and under what assumptions, frameworks and mindsets.

We tried to start thinking about our work through a lens of justice and systemic oppressions, rather than focusing (deliberately or unintentionally) on a ‘cohort’ of people who needed fixing.

As well as trying to integrate this into our strategy and internal practice, we also made a number of grants under the banner of ‘knowledge’.

We wanted to move away from an individualised, even medicalised, framing of marginalisation and towards something more systemic.

Whilst ‘severe and multiple disadvantage’ was intended to broaden the conversation about complex and interlocking social harms, we were surprised by the speed at which the new, siloed shorthand of ‘people with SMD’ painted over this and obscured many of the very connections we were trying to highlight.

‘SMD’ became another condition which could be solved one person at a time through the delivery of ‘more targeted and effective interventions’ by services, which in turn were operating in a market-based performance management paradigm with control, ‘value for money’ and artificial predictability as its central features.

We tracked this all the way back to a paternalistic, damage-centred and needs-based framework of worthiness, and longed for something less ‘charity’ and more ‘justice’.

We’d realised that our framing of ‘the problem’ was part of the problem.

We concluded that we’d never ‘complete the jigsaw’ when it came to piecing together a coherent picture of severe and multiple disadvantage.

Every time we switched the lenses we were using to investigate the issue, it completely changed who we were talking about – for example, we knew that the prison population skewed towards white men, whilst other measures like mental health or exploitation illuminated the experience of other groups.

We could never get it ‘right’; so rather than squashing ever more and different groups of people into a ‘cohort’ which was already bursting at the seams, we saw that we needed to change our thinking on a more basic level.

This work enabled us to break away from needs-based definitions of social harm and disadvantage, and opened up conversations and action concerning more fundamental inequalities and oppressions.

It flowed from – and contributed to – ideas and dialogue about justice, equity and liberation, which had sometimes been an awkward fit with our previous strategy.

Ideas about ‘knowledge justice’ also helped us to explore and challenge the fundamental assumptions of the systems we were working in, like who is listened to and believed; which people, methods and evidence are considered credible, robust and legitimate; how decision making works; and how ‘success’ is defined, measured and experienced.

In other words, ‘Whose Knowledge?’ really matters?

By using an open-call grantmaking methods, we were able to reach and fund some people, work and issues we hadn’t engaged with before, and bring people together at the intersection of advocacy, activism and research.

Although there was strong interest from grantees, ‘knowledge frameworks’ was quite a tough sell as a funding programme.

We had to think differently about the levers we held to make or nurture change in this area. The jump from ‘these are some thoughts and reflections’ to ‘let’s do something’ wasn’t an easy one.

We chased our own intellectual tails for some time, and were troubled by the dichotomy between philosophical, chin-stroking critique and ‘real-world action’.

At times it was hard to tell whether we were unpicking ‘hidden wiring’ in the system which could unlock critically important action, or just going off on a philosophical tangent. Even internally, we struggled to dispel the notion that the work was ‘too theoretical’.

Once we took away some of the boundaries governing our work and mission – particularly the stricter definitions of what constituted ‘severe and multiple disadvantage’ – it was hard to put them back up again.

Breadth was a strength, but vagueness was a weakness – and they often went hand-in-hand.

The number of people, communities, issues and organisations we were in touch with broadened considerably, and whilst we gained nuance, we lost some clarity. This change worked well for some grantees, whose analysis was already focused on systems of oppression rather than addressing needs, but others found it confusing or felt a moving of the goalposts – ‘don’t you ‘do’ homelessness anymore’?

Similarly, it was hard to find satisfactory alternatives for some of the frameworks we’d lost confidence in. It became difficult to re-ground ourselves and our partners in different, coherent ideas about scale, progress, performance and accountability.

We’d left one shore, but struggled to reach a different one.

In terms of grantmaking: having convened a group and asked them to work together, whilst some relationships blossomed, we probably overestimated the extent to which people brought together in this way could form a ‘collective’, including one with the power to distribute funds more widely.

This meant we sometimes got stuck on questions about governance, decision-making and what shared priorities could look like, when time would have been better spent maximising the insights people could take away for their own work.

You could argue that ‘knowledge’ was another brick in the crumbling edifice of institutional philanthropy as we saw it, as our analysis became incompatible with our actions and our setup.

We found it hard to square the principles of knowledge democracy – equitable governance and the centring of marginalised voices – with a hierarchical operating model based on the hoarding and gradual release of ‘charitable’ resources.

We became increasingly uncomfortable with our inability to reconcile the tensions inherent in philanthropy: for example, the underlying governance assumption that foundation staff and trustees are uniquely able, experienced, knowledgeable and qualified to make decisions about, and hold responsibility for, money.

Looking into our knowledge practices was one of a number of ways in which we lost confidence in the redeeming features of the models and actions we were still routinely engaging in.