Mapping climate policy: inviting and navigating diverse perspectives (Part II)
In our previous post detailing CPR’s approach to navigating context-specific language, we mainly discussed Indigenous Peoples as a vulnerable group. However, for many, being part of an indigenous community is a source of pride and strength — and with good reason. Indigenous communities play a crucial role in protecting much of the world’s biodiversity. Indigenous Knowledge and Indigenous practices can also keep communities safe from the effects of climate change.
Speaking personally, I had not really considered how calling someone “vulnerable” could be disempowering or even offensive. I come from an academic background; among climate change researchers, "vulnerability" is a widely used term with an exact definition from the IPCC, the world's leading climate science panel. Over the years, lots of work has gone into creating a definition that acknowledges structural inequalities and the unique set of circumstances someone might face. So hearing just how strongly people disliked being called vulnerable was a bit unexpected.
This is about more than avoiding offense. We want to speak in a way that our users can recognise and feel represented by. That’s exactly why it is important to invite a wide diversity of perspectives; to speak with people, not about them. Data science tools, such as these maps, are useful but they do not cover all blindspots. As much as possible, we try to develop our tools together with users and those with lived experience.
In the case of the particular term "vulnerable", there was an easy alternative. “Impacted groups” is conceptually very similar to “vulnerable groups” and is generally received better by the people we spoke with. Of course, these terms are not exactly the same: if you're a billionaire whose holiday home was destroyed in a storm, you're impacted by climate change even if you are not particularly vulnerable -- after all, you can just stay at your other home and will likely be fine. Still, we think that for most people "impacted group" will be an acceptable and clear alternative and it even allows us to widen our scope a little: many workers that are at risk of losing their job due to the green transition are not vulnerable to climate change itself, but they are definitely impacted in a meaningful way.
A search from Climate Policy Radar’s mapping tool highlights the widespread use of vulnerable as a descriptor.
Searching ‘impacted’ by comparison.
An additional difficulty here is that preferred language changes over time. For example, the British government in the past used the acronym BAME (Black, Asian and Minority Ethnic). A few years ago, they officially stopped this because the term was found to be divisive and exclusionary. But it is still present in historical documents in our dataset and some other groups continue to use the acronym even now. We think it is important to still surface documents with outdated language, even if we do not use these terms ourselves.
In some other cases, it may be difficult to find an alternative that satisfies everyone. One of the people we interviewed about impacted groups gave us the example of the phrase “urban community”. Especially in the USA, this is racially coded as it is often used to refer to Black communities. In many other countries, though, the phrase "urban communities" does not tend to have ethnic or racial connotations. Alternatives are also a bit unwieldy — “communities living in cities” isn’t just a bit of a mouthful, it also just isn’t how most people working in this field talk about their work.
Overall, CPR is a global not for profit, providing tools for climate action for people from around the world. This means we need to balance the needs of different communities, while also factoring in clarity and usefulness. As a general rule, we want to name things as they are: otherwise, users can’t find what they are looking for. But because language is so varied, that really means naming things as they tend to be understood by most people. Sometimes following what most people use is exclusionary or offensive, and we need to consider what we do when this happens. There isn’t always a clear best answer, so we strive to be deliberate and transparent about these decisions. And of course we will not always get it right, so we need to be flexible too. If you see something in one of our products, or in this blog post, that you think could be improved, please email me at anne (@) climatepolicyradar.org
TL;DR
Because we care about policy, we care about language. Data science methods can help us map and track patterns in that language, but these can be pretty blunt instruments when used naively. We need to do the work to also surface smaller stories and understand the nuances of language. Listening to marginalised voices in particular is crucial. This means we can avoid causing offence but more importantly, it means we do a better job of representing diverse perspectives, which translates to better results,in particular for those who are most in need of climate action.
