Person-First or Identity-First?

When I started my PhD, I was occupied for several months with almost exclusively the question on whether I would say “autistic people” or “people with autism.” Before I say more, you should know that I do not identify as autistic myself – albeit disabled – but grew up with an autistic sibling.

I stumbled across the question of language in the way that most researchers probably do, in that I had never really thought about it, since ‘it does not concern me personally, right?’. Through that I kind of offended a friend of mine who has a kid with autism and explained to me that there is a difference in person-first or, as they called it, ‘label-first’ language. Not wanting to offend anyone, I dove into the debate. And it’s an intense one.

The main argument for person-first language is that by othering and labeling individuals through adjectives, they become second to their label and are not seen as individual. Others have elaborated on this more, but that seems to be the gist of it. If you say ‘a person with autism’ it becomes more about the person than about the autism. But then, does it?

According to Jim Sinclair, this way an autistic person is also seen as separate from their autism, which is contrary to their experience. Why can they be called clever and handsome, but not autistic? Why is it not a person with smartness or a person with beauty?

Ultimately then, it does not seem to matter how we talk about a condition grammatically, but rather, which semantics are attached to it. For those who are into that, there is some research on how the grammatical construct does not necessarily alter semantic meaning.

However, those who connect the construct with a certain meaning care very much about how we talk about autism and autistic individuals. You might have noticed by now that I am mostly using identity-first language and those who know me also know that I strongly advocate for using it. This is not only because I find Sinclair’s argument slightly more convincing, but also because a large survey in the UK indicated that autistic individuals tend to prefer it as well. Sue Fletcher-Watson recently analysed the results of the survey and shows how professionals dealing with autism according to the numbers care emphatically more about which language is used and strongly prefer person-first constructs whereas autistic people just indicate a slight preference. The conclusion then is that mixing grammatical constructs might be the least offensive way for everyone involved. I disagree.

I perfectly understand the need to work with the autism community at large, which includes autistic individuals, their families, their carers and professionals. But what is language if not a political statement. If there is a debate on how language is used and a large portion of the people we try to address is telling me ‘please call me autistic’, then who am I to judge them wrong? I need to listen to them and I need to listen to them more than to professionals who do have more power in this conversation already. Using identity-first language to me is a political act that acknowledges the preferences of those who are marginalised all too often.

This is my conclusion, but this is not the only way to do it. I don’t think there is a ‘correct’ way to talk about autistic individuals. First of all, I also change my language if I talk about an individual who prefers e.g., person first language (see above when talking about my friend’s kid). Second of all, I don’t think it helps anyone if people just try and do what is ‘politically correct’ without thinking about the circumstances that does make something correct or incorrect. That is why I might ask you why you are making a certain choice. I want to learn more about the different arguments surrounding the language and re-evaluate my own use. For me, it is most important that I make an informed choice in how I speak, especially when speaking about something that is outside of my personal experience, such as being a non-autistic researchers talking about autistic children.

Ultimately, and I think that is where everyone in the debate agrees on, consider the preference an individual has when talking about them specifically. Only that way we can start establishing a person-first attitude regardless of the grammatical constructs we use.

Join the Conversation


  1. Thanks for this Katta. I agree with a lot of what you have to say here. However, I’d like to clarify the conclusion of my blog post discussing the results of the large survey you reference above (paragraph 6).

    First, a central point in my post is that there is no empirical evidence for a preference on how to talk about autism when looking exclusively at the responses of people on the spectrum. So desiring to adopt the preferred language of the a/Autistic community (as opposed to the autism community more generally) does not solve the problem of what language to use. There’s a separate argument from theory, but based on the numbers, there is no majority preferred term among people with an autism diagnosis (or self-identifying as such).

    Following from this, in the post I consider an (admittedly controversial and provoking) option which is to weigh the opinion of a broader array of stakeholders in the balance, also taking into account the extremity of their feelings on the topic. However I do not advocate for this, I merely present it as a way to consider the data. In fact, you apply a version of this approach yourself in your choice to refer to a friend’s child using person-first language in line with parent (but not necessarily autistic) preference.

    Finally, the point I want to make here, is that my conclusion that a mix of language could be appropriate is largely based on the mixed opinions among autistic people specifically, not on differences in opinion between (for example) people with autism and practitioners.

    Thanks for the post. A nice piece and I hope you’ll forgive me for using the opportunity to defend my interpretation!

    1. Oh no, I love having this discussion and I thank you very much for defending your interpretation. I especially like the clarification you made as it adds your intention to my reading of the post.

      I do agree that the empirical evidence is thin, but that to me is also not the only thing that counts, but rather that the argument about felt experience is essential to my choice of language as well. If I saw correctly, that was the main argument people who directly commented on your blog post made as well.

      When talking about my friend’s kid — due to them living on a different continent and me having seen the kid the last time when they were three years old — the closest I come to knowing about their self-identification is talking with their parents about them. I do see how it comes across slightly besides my argument, if that context is missing, so thank you for giving me the chance of providing it.

Leave a comment

Your email address will not be published. Required fields are marked *

6 + nine =

This site uses Akismet to reduce spam. Learn how your comment data is processed.