top of page
  • Writer's pictureChris Wingate

“Person-first” versus “identity-first” language in the autism community

In the past decade or so especially, there’s been an increasingly significant debate of sorts within the autism community over whether or not to use “person-first” or ”identity first” language. Today, I’d like to talk about what “person-first” and ”identity-first” language is, their significance, and perhaps the future of these types of language.

Person-first language

Firstly, what is person-first language? Simply put, person-first language puts a person before their diagnosis. Someone with autism is a person with autism; someone with asthma is an person with asthma; someone with HIV/AIDS is a person with HIV or AIDS, etcetera.

The idea is probably older than you'd think, too: psychologists Carolyn Vash in 1959 and Beatrice Wright in 1960 were both early advocates for what the term now refers to. With that said, “person-first language” as a term and person-first language in general only began to see significant usage in the 1980s and 1990s. The HIV/AIDS community was possibly the earliest and biggest advocate of person-first language, rejecting passive terms such as "victims" and patients" in the 1983 Denver Principles and saying:

We condemn attempts to label us as ‘victims,’ a term that implies defeat, and we are only occasionally ‘patients,’ a term that implies passivity, helplessness, and dependence upon the care of others. We are ‘People With AIDS.’

The term ”person-first language” itself was not used in full until 1988. From this ”fringe” of advocacy, though, person-first language was quick to spread. By 1992, advocacy for person-first language had fully broken into the health and disability mainstream, with organizations such as the American Speech-Language Hearing Association publishing full (and still useful) resources on how to make use of person-first language.

Since the mid-1990s, person-first language has arguably been the default in most disability communities, as well as the medical establishment—it is, for example, mandated in the American Medical Association's manual of style for writing, and commonly accepted or mandated by most government institutions in English-speaking countries. But increasingly in the past decade, person-first language has come under criticism from those who see it as demeaning to who they are—and so, enter identity-first language.

Identity-first language

Identity-first language is more to the point—autistic, asthmatic, blind, deaf—and unqualified, and harder to pinpoint the origins of. Easier to pinpoint are the many reasons autistic advocates profess for its usage: autism is an identity; autism is immutable; autism is a positive to be celebrated, not treated as an ailment; person-first language invalidates autistics; and so on.

A case can be made that identity-first language predates person-first language: the deaf community for example has supported identity-first language and deaf people have rejected the notion of needing to qualify themselves to be seen as valid. Without question, however, identity-first language is almost as old as mainstream person-first language. In 1993, the National Federation of the Blind rejected person-first language—and with the hindsight we have now, committed itself to identity-first language:

We believe that it is respectable to be blind, and although we have no particular pride in the fact of our blindness, neither do we have any shame in it. To the extent that euphemisms are used to convey any other concept or image, we deplore such use. We can make our own way in the world on equal terms with others, and we intend to do it.

In the autistic community, Jim Sinclair's 1999-or-so article Why I dislike "person first" language was the first to call for identity-first language. At the time though, identity-first language had no name.

In the following two decades, identity-first language also languished in apparent obscurity, while person-first language grew in disability communities and the medical establishment. While it's beyond the scope of this article, it seems likely that between 1999 and 2011, identity-first language remained a largely-offline, largely activist-dominated, and very niche subject. In fact, based on Google Search Trends, ”identity-first language” doesn't seem to have even come into usage online until 2011, implying there may not have been an established name for it well into the 2000s. Starting in 2011 though, that changed.

Lydia Brown, writing on her blog Autistic Hoya, was probably the first prominent person online to tie identity-first verbiage together with autism. The Significance of Semantics: Person-First Language: Why It Matters still does not use the specific phrase ”identity-first language”, but the dots are there to be connected, and based on the trajectory of identity-first language on Google Trends Search, it seems like other people connected them. Effectively from her blogpost on, identity-first language comes into being online—and the increased visibility of autistic people and autism activists, both online and offline, no doubt helped and accelerated this process. Today, you seem about as likely to run into person-first language as you are to run into identity-first language, and the gap between them is lessening.

Autistic preferences and the future

With the increasingly established nature of both these forms, it is only natural that there has been conflict between those who advocate for person-first language, and those who advocate for identity-first language. Both forms have their reason for existing, both can be justified on reasonable bases, and both are obviously meant to normalize autism and show that it is not a bad thing to be diagnosed with. This is a debate where both sides have—at least on paper—good reasons for using what they use.

Surveys—at least since 2015—have shown quite clear preference among autistic people for identity-first language, however. Indeed, one of the biggest splits in language usage is a rather alarming one: autistic people, who tend to use identity-based language, versus medical professionals, who tend to use person-first language. Most significantly, this is seen in Kenny et al.'s 2015 study Which terms should be used to describe autism? Perspectives from the UK autism community with "aspie" and "autistic person".

Other surveys have shown similar disparities to Kenny et al.'s, among autistic people versus non-autistic professionals—but unsurprisingly, also among autistic versus non-autistic people generally.

In 2018, Chris Bonnello of Autistic Not Weird did a survey and found autistic people are likely to use "autistic person" and similar language, but non-autistic people (and especially those with no autistic relatives) just as strongly prefer to use "person with autism". A 2020 survey by the Organization for Autism Research also found autism professionals and educators to be most skeptical of identity-first resources—although all groups in the survey were strongly supportive of them.

What this suggests is that there is a disparity in language between autistic self-advocates and the people who they often rely on the most: those closest to them, and those who are entrusted with helping them and treating them when needed. And while this certainly isn't the largest issue for autistic people by any means, it certainly is an issue. If the apparent preferences of most autistic people aren't respected on such a basic issue, it's hard to imagine autistic people trusting their family or medical professionals to be involved when more complicated and pressing issues occur.

Furthermore, there are some increasing concerns that the semantic separation between person and autism that person-first language makes could contribute to violence experienced by autistic people. As observed by Botha, Hanlon, & Williams earlier this year in an article titled Does Language Matter? Identity-First Versus Person-First Language Use in Autism Research: A Response to Vivanti:

The implications of the semantic separation of autism and people are most clear in cases of filicide. Autism is a risk factor for ‘altruistic’ filicide, and in a review of 26 filicide-suicides reported in newspapers between 1982 and 2010 in the USA, 54% of victims were autistic, despite making up between only 1 and 2% of the population. [...] McGuire (2016) highlights how in cases of filicide, parents often fall back on the same linguistic separation—I loved my child very much, but I hated autism, and wanted autism out of my life.

Still, while this section may sound gloomy there's a case for overall optimism here: if you treat Kenny et al., Chris Bonnello, and the OAR's research as equivalent, there's a case to be made that this disparity is slowly lessening between the autism community and those outside of it. And judging by how outside communities adopted person-first language, it seems likely that identity-first language will eventually gain widespread acceptance.

But it's not certain—nothing can be taken for granted like that—and we're not to that day yet. Until we are, the tension between the autism community and those outside of it will probably remain. And while that tension remains, it seems obvious to respect the wishes of the autism community: try to use identity-first language unless specified otherwise, and encourage those around you to use it too. It's not a big shift, but it'll definitely a meaningful one for those around you and the autistic community as a whole.


Botha, M., Hanlon, J., & Williams, G. L. (2021). Does Language Matter? Identity-First Versus Person-First Language Use in Autism Research: A Response to Vivanti. Journal of autism and developmental disorders, 1–9. Advance online publication.

Kenny, L., Hattersley, C., Molins, B., Buckley, C., Povey, C., & Pellicano, E. (2016). Which terms should be used to describe autism? Perspectives from the UK autism community. Autism, 20(4), 442–462.


Recent Posts

See All


bottom of page