We like to believe that music changes the world. That singing together heals communities. That an orchestra in an underprivileged neighbourhood does more than just teach people to play notes. I often hold on to that hope. It feels intuitively right, morally attractive, and maybe even politically useful. But what do we really mean when we say music brings about “social transformation”? And do we actually measure that properly?
These questions are central to a recent large review by Noemy Berbel-Gómez, Lluís Ballester-Brage, Laura Serra-Marín and Júlia Mérida-Coli, published in Review of Education. The authors analysed 270 scientific studies on so-called community music projects: initiatives in which people make music together in neighbourhoods, schools, refugee camps, or social organisations.
The premise behind much of this research is clear. Music is assumed not only to benefit individual participants, but also to contribute to broader social change. And then comes the sobering part — though not in the way you might expect.
What the review shows is that most studies focus narrowly on personal outcomes: increased self-confidence, greater well-being, improved social skills, or a stronger sense of belonging. These are valuable effects, without question. But they are often quickly translated into much larger claims such as better integration, stronger communities, social cohesion or empowerment. That is where the scientific foundation becomes shaky.
In many studies, “social transformation” is not clearly defined. It often amounts to little more than participants feeling better or making new contacts. That matters. But it is not the same as a change at the level of a community or a social structure. A person who feels more confident is not necessarily a sign that a neighbourhood has fundamentally changed.
Another issue is methodological. Much of the research is small-scale, local and predominantly qualitative. Interviews, observations and case studies dominate the field. These can offer rich and moving stories, but they make it difficult to draw general conclusions. There is often no comparison with other interventions, or with situations where no music project took place at all. As a result, we rarely know whether music does something unique or whether it simply provides a pleasant setting for social interaction.
Perhaps the most striking gap concerns collective processes. How do durable social networks emerge? Do relationships persist after the project ends? Do power relations within a community actually shift, or do they remain largely intact? These questions are seldom examined systematically. Instead, the focus remains on individual experiences such as “I feel heard” or “I belong”. That makes the broader claim of social transformation vulnerable.
The authors also warn against romanticising music. Music is not automatically inclusive or unifying. It can also exclude, emphasise differences or reinforce existing norms. In some contexts, making music together can even generate tensions or make certain groups less visible. Social processes are never purely positive, and musical practices are no exception.
At heart, this is a measurement problem. We use big words like integration, empowerment and transformation, but we lack robust frameworks to investigate them properly. We need evaluations that look not only at individual feelings, but also at group dynamics, context and sustainability over time.
This has clear policy implications. Community music projects are often funded with the idea that they will help solve social problems. That may well be true. But then we must also be honest about what we know and what we do not know. Otherwise, we risk burdening arts projects with expectations they cannot meet, or replacing evidence with good intentions and compelling stories.
The conclusion of this review is not cynical, but mature. Music can be meaningful. It can connect people. It can support personal growth. But if we want to speak seriously about social transformation, we need to be precise about what exactly changes, for whom, for how long, and why.