Learning patterns/Centering Marginalised Knowledge

From Meta, a Wikimedia project coordination wiki
A learning pattern foroutreach
Centering Marginalised Knowledge
problemThe knowledge of marginalised communities is the knowledge of the majority of the world. How can we best support these communities to bring their knowledge onto Wikimedia projects?
solutionAsking ourselves some critical questions before and as we work with marginalised communities may help us become good allies in centering marginalised knowledge.
created on04:22, 4 December 2017 (UTC)
status:Work In Progress :-)

What problem does this solve?[edit]

You’re a Wikimedian who wants to support marginalised communities to add more knowledge to Wikimedia projects. But you’re not sure how to do this thoughtfully and respectfully, as an active ally. How to begin?


  • Our learning comes from the work we've begun as Whose Knowledge? in 2017, including our support of the Dalit community in India and the diaspora, and queer activists from Bosnia-Herzegovina. It also incorporates insights from the work with the Kumeyaay, a Native American tribe from Southern California, USA and Baja, Mexico.
  • We define marginalised communities as those denied full access to rights and resources, and "left out because of structures of power and privilege". These include women, people of colour, LGBTQI communities, indigenous peoples and others from the global South.
  • You may self-identify as being from a marginalised community - e.g. as a woman - but we hope these questions will help you too. Power and privilege are not static, they work differently in different contexts. For instance, if you're a North American woman in a conversation with a North American man, you may feel disempowered (we hope not!). And when a Latin American woman is in conversation with you, she in turn, may feel disempowered (again, we hope not). It's important to remember that each conversation and relationship will have a different and changing set of power dynamics, not only between you and your immediate partner, but amongst all those involved.

What is the solution?[edit]

Asking the right questions will help you get started on this journey… here are some things you might think about as you begin to work with marginalised communities as an ally.

In this learning pattern, we offer seven critical questions that we ask ourselves, with a few additional questions under each section to help you. We also offer some responses and examples from our experiences that illustrate what we have learnt. Every aspect of this work needs us to be constantly conscious and self-aware of how power works in relationships and processes - in good (empowering) and bad (oppressive) ways.

Why do you want to do this?[edit]

  • Why is it important to you?
  • Who do you think will benefit? (hint, if it’s just Wikimedia, you’re in the wrong place)

Knowing why you want to do this work, is key to how you do it. Your motivations should make it easier for you to support the leadership and expertise of those you work with. If you see the benefits as being primarily to you and your team, or your ego gets in the way, you may want to rethink why you're doing this (and we know it's hard).

Example: When we began our work, we knew that the history and knowledge of marginalised communities is the history and knowledge of the majority of the world. We also knew that easily accessible public knowledge (including Wikipedia) doesn't reflect this majority, either in content or contributors. We could have worked on improving Wikipedia's content and expanded the editor base in many different ways, but we believe that centering the leadership and knowledge of marginalised communities online, and honouring their expertise and experience, is the most important way for us.

Whom do you ask?[edit]

  • Which relationships do you already have with marginalized communities that you can build on?

Trust is at the core of any relationship and collaborative project, but particularly with communities that have been historically oppressed. It takes time to build this trust, so starting with individuals and groups you already know - and who know you - is a good way to begin. Ask them how and with whom you should explore a partnership or project together, and don't be impatient if the introductions take time. Not all explorations will lead to a final project, but every conversation is likely to make you better informed and open to new histories and experiences.

Example: We began exploring a partnership with the Dalit community because Anasuyas had been working for many years to challenge caste-oppression in India and USA, and knew and admired the work of Equality Labs. Seeeko had facilitated an edit-a-thon with Kumeyaay scholar Mike Connolly Miskwish at the Wikiconference North America 2016. Both of these were starting points for our current partnerships.

How do you ask?[edit]

  • How do you ask in ways that build trust, respect, and shared power?
  • How do you know when you're losing trust or taking power?"

You explore potential partnerships with humility, curiosity, and the willingness to be led by what matters to those you're in conversation with. Ask those you're reaching out to, why they might be interested in working with you, and what they are hoping for. See the process as being of uncovering the ground of mutual interest and expectation, and then convening different sets of expertise towards a common goal. But trust, respect and shared power can only evolve when you are willing to shift and change the way you might conceive of the partnership, based on what you are hearing and learning. And you will know you are losing trust, or taking power in the conversation, when you are speaking 70% of the time, and those you're speaking with, are expressing themselves 30% of the time or less. Or there are awkward silences that are not about mutual pauses for reflection, but about discomfort and mistrust. Also be prepared to have the conversations in spaces and places that are comfortable for those you're speaking with.

Example: When we began exploring a partnership with Okvir, the group of queer activists in Bosnia-Herzegovina, they asked us to visit them in Sarajevo. This meant a substantial investment in time, money, and energy. We would never have thought of doing the trip unless we'd been open and willing to listen to our friends at Okvir, and what they felt was the most important way to explore a potential partnership. The trip ended up being the most valuable and effective start to our mutual learning and shared work together. Most significantly, we learnt so much about the high-risk context they work in, that we would never have understood sitting comfortably behind our laptops.

Where do you begin?[edit]

  • Is the priority coming from the community?
  • How do you uncover and support priorities from marginalized communities?

Now you're starting to work together, congratulations! You begin with a process of uncovering mutual needs and expectations, but with the priorities being led by those you're working with. These priorities may surprise you - they may be entirely different from what you expected - but you must be willing to start where the community wants to be. You need to be ready to learn, and possibly to change your previously held opinions or frames of reference. Be very curious and humble - this phase is focused on some version of these questions: What is important to you in doing this work, Why does it matter to you and your community?, What worries you as we begin and why?, What gives you joy, and why?, and What support do you need to do this?. As you do this, the phrases 'I didn't know that' and 'Thank you for helping me learn' should come easily to you! Once you have a sense of the priorities and needs for the project from the community members, you can discuss the ways in which your interests and skills align with these priorities.

Example: When we first began conversations with the Kumeyaay, we assumed that our entry point would be Wikipedia. And some community members were excited about working on Wikipedia and opening up their knowledge to the world. But others reminded us that till 1978, it was illegal for Native Americans to share their cultural and religious practices and knowledge. This oppressive history still makes it difficult for some Kumeyaay to share their knowledge with trust and safety. We needed to be open to having differing priorities set by the community than what we expected. While one member of the group has begun piloting re-writing Wikipedia in her “indigenous decolonization” class as a model for other Kumeyaay teachers, oral history appears to be a more useful entry point with many members of the community.

How do you design to empower and share leadership?[edit]

  • Who leads and who supports?
  • What are the guidelines and shared practices you build together?

Begin with recognising that everyone in a conversation - especially those from marginalised communities - has expertise of different kinds (we know; this isn't how most folks normally think of expertise or knowledge). Sharing leadership is about being aware of what kinds of expertise and skills each of you has, and when and where they need to be used. Beyond all other domains of experience, you're a Wikipedian who knows Wikipedia and its complex processes, and those you're working with are experts in the scholarship and embodied experiences of their communities. How do you bring those different skills together to create a shared design, and who/what else do you need? Start with a set of guidelines that will help you build shared practices and protocols, by asking everyone what they imagine to be good and bad practices in the work you're designing. What do you most want to see, and what do you least want to see, happening? Reverse engineer to a set of guidelines that you're all accountable for, but most especially, *you* are most conscious about holding and respecting.

Two key aspects of this co-design and practice are: security and safety, and time and work rhythms. Please remember that people from marginalised communities work in challenging, often risky, environments. You need to understand that your rhythms of working may not match theirs, and that you need to discuss time frames and work practices that suit everyone involved. You also need to jointly create and share safety and security guidelines, following the principle of "openness with safety": not all information should be freely shared, or "open" without anonymisation, because it could harm community members. Help those you're working with to create user accounts and work on Wikimedia projects in ways that will not compromise their security, online or in real life.

Example: As we began work with the Dalit community, Anasuya was very conscious that she had been born into an "upper" caste (savarna) Indian family, and had to recognise both her own structural privilege and the ways in which she symbolised oppression to those we were working with. Similarly, Siko recognised her power and privilege as a white North American. It meant that in designing our processes, we were always aware of when we might be taking up too much space with our ideas or conversation, and when we needed to step back to be led by the Dalit organisers and scholars. Our shared guideline from the start: that we should be told immediately - in whatever ways felt comfortable - if we ever violated this safe and productive space we were creating. Being upfront about our different positions, and about the guidelines, was a critical part of our shared design. In addition, we worked on shared plans for safety and security, because we knew the organisers were at high-risk in real life, and could be (and had been) harassed online.

Whom else do you bring in?[edit]

  • Who is important to include from the marginalized community’s perspective?
  • Who are the partners and allies you engage?
  • How are expectations set with allies?

So you now have a brilliant set of people working collaboratively together to center marginalised knowledge! But you don't have all the skills and connections needed to do this fully. Inviting others to join works in two ways: first, by making sure the community you're working with knows and shares whom they want to invite; and secondly, for you to suggest people and organisations that can be additional partners. Making sure the new partners are aligned with the shared values and guidelines that you've designed, is key. Principles before technical competence, for instance (and we all know that deeply principled techies exist, they're our friends!).

Example: As we began to support Okvir with their community archive, we realised that an excellent advisor and partner would be Internet Archive (IA). Even though we knew how closely aligned all our organisations were, we first checked with Okvir that they were comfortable with our reaching out to IA, and only then did we convene a meeting between all three groups (note: some meta data conversations are really fun!).

What does “success” look like and to whom?[edit]

  • How do these communities define their own “success”?
  • Are quantitative indicators the most useful?
  • When and how is backlash an indicator of “success”? How do you support the emotional and intellectual labour of backlash?

The "success" of this project may have very different meanings for all of you involved. What are the goals of the community you're working with, and how will they know they've achieved them? is a question that will help you come to a shared understanding of what's important, why, and how to get there. And be prepared to have these goals shift and change, as you do the work. Quantitative metrics may not be where you begin (even though it may well be a secondary outcome of what you do). A combination of transactional (things you can count) and transformational (the important but intangible) metrics is often a good way to "measure" this work. Most importantly, be aware that bringing marginalised knowledge online can lead to significant backlash from those who are challenged by this process. While it is one way to think about "success", it is a painful and often re-traumatising process for marginalised communities. This is when Wikimedians who support this work can be most useful as allies.

Example: We began our work with the Dalit community thinking about quantitative content goals, and "succeeded", some might say. In many ways, we were wrong to do so. Because there was so much editing by a community that is challenging South Asian power structures IRL and still learning Wikipedia principles and processes online, there was significant reversion of the initial work. This has been demoralising to many community members, even as their initial experiences were deeply empowering. But we've learnt from this. We've suggested that the Dalit organisers might focus on a few articles they can improve substantively. And in turn, the Dalit organisers have been deeply appreciative of the Wikipedian allies who not only helped them learn how to edit, but went on to challenge some of the backlash they were facing.

Things to consider[edit]

  • Power, power, power: it begins and ends here. This work is about learning ways to share power (power with), not wield it (power over).
  • Allyship is a verb: how you act is far more important than what you say.
  • Communities are not homogenous: even as you uncover the power relations between you, be aware that within the communities you're working with, they're coping with similar challenges.
  • Amplification, not representation: none of us can represent whole swathes of the communities we identify with. But we can amplify their ideas, experiences, and stories.
  • Never forget: this work is with different generations of communities that have been systematically made invisible, and often brutally barred from sharing their experiences and knowledge.
  • Context matters: there is no formula or blueprint for this work. It's a map you're creating together, a lot of it undiscovered territory.
  • This is not a linear process: surprise! Be prepared to ask some of these questions to yourself and each other many times during the project together. The answers may not be the same the second time around.
  • It's messy: there are no easy answers.
  • Time is relative: every community has a different rhythm and cadence to their work, especially because they're often combating significant challenges in real life. Be prepared to go slower at some points in the process (e.g. when the community is managing critical real life issues), to go faster at others (e.g. if there's backlash to your work together), and to change focus if and when needed at others.
  • Be comfortable with discomfort: it helps you know when things are going well, and especially when they're not.
  • Don't forget to have fun, and create joy (and knowledge) together!


See also[edit]

Related patterns[edit]

External links[edit]