Identity First or Person First Language Explained

Chris Kranz 27 April 2026 8.3 min read
reclaiming language and identity

Identity First or Person First? A Quick Guide Without the Drama

I’d been diagnosed for about three months when it happened. Someone in an online forum - friendly, well-meaning, the sort of person who uses correct punctuation in Discord messages - asked me whether I used identity-first or person-first language. I stared at the screen with the energy of a person who’d just been asked to solve a differential equation at a party. I didn’t know there was a quiz. I hadn’t revised.

So, before anything else: both are valid. Your preference is allowed to change. And you cannot get this wrong. There is no committee. Nobody is issuing citations.

This article is for the person who got a diagnosis sometime in the last few years, possibly decades later than would have been useful, and is now encountering a whole vocabulary they were never handed an instruction manual for. It’s a no-drama zone. We’re just going to look at what these terms mean, why people care about them, and what you might want to do with that information - which could, legitimately, be nothing.

The short answer: use what feels true for you; follow the individual’s lead for others. When in doubt, ask.

What Is Identity-First Language and Why Do Autistic People Prefer It?

Identity-first language puts the descriptor up front: autistic person, disabled person, dyslexic student. Person-first language leads with the human: person with autism, person who has ADHD, student with dyslexia.

Neither is universally correct. They come from different philosophies about how a diagnosis relates to who someone is.

Person-first language came out of disability rights movements in the 1980s and 90s. The intent was genuinely good - push back against clinical labels that reduced people to their conditions. Patient became person with. The logic was: you are a person before you are a diagnosis. Hard to argue with that on its face.

Identity-first language gained ground later, particularly within autistic communities, where a different argument took shape: my neurology isn’t a handbag I’m carrying. It’s not separate from me. It is me. Saying “person with autism” implies autism is an accessory - something bolted on that could theoretically be removed. And for many autistic people, that framing feels dishonest at best, and at worst like a quiet suggestion that they’d be better without it.

Here’s a useful way to think about it, if you’re a late-diagnosed adult sitting in the middle of this going “but which one do I pick”: we don’t say “person with gayness” or “person with Jewishness.” For some identities, the descriptor and the person are the same thing. For others - and this is genuinely true for some people with some conditions - keeping a bit of distance between the self and the diagnosis feels important and protective. Both instincts are real. Both come from somewhere honest.

One thing worth knowing: professional bodies like the APA and ASHA have historically defaulted to person-first in their style guides. Meanwhile, many autistic self-advocates, and a growing number of autistic researchers, prefer identity-first. That gap between clinical convention and community preference is itself a useful piece of context if you’re a newly diagnosed adult trying to work out why your psychologist says it one way and everyone on Reddit says it another.

Is One Option More “Correct” Than the Other?

The most common myth in this space is that person-first language is always more respectful. It’s not. It’s one form of respect, and assuming it’s the only one can actually land badly.

The logic behind “person-first is safer” rests on an assumption that’s worth examining: that a diagnosis is inherently negative. Something to be distanced from. Something you wouldn’t want to be, only to have. And for a lot of neurodivergent people - especially those who spent thirty-seven years wondering why everything felt slightly harder than it seemed to be for everyone else, and then finally got an explanation on a Tuesday afternoon in a clinical psychologist’s office in Basingstoke - the label isn’t a burden. It’s relief. Separating it from the self can feel less like respect and more like erasure.

Research backs this up. Kenny et al. (2016) surveyed the UK autism community and found that autistic adults overwhelmingly preferred identity-first language, while parents and professionals leaned person-first. Jim Sinclair made the case as early as 1999 that saying “person with autism” suggests autism is something separable from the person - and that this separation is itself a value judgment.

But - and this matters - this isn’t only an autism conversation. Community preferences vary considerably:

  • Autistic communities lean strongly identity-first; Kenny et al. (2016) found this was the preference of the majority of autistic adults surveyed
  • ADHD communities are less unified - many use both interchangeably and don’t feel strongly either way
  • Dyslexic communities tend to use identity-first almost by default (“I’m dyslexic”) without anyone treating it as a political stance, which is interesting in itself
  • Deaf communities have used identity-first for decades, often with a capital D to signal cultural identity rather than audiological status

When you’re talking about yourself, there’s no correct answer. When you’re talking about someone else, follow their lead. Or ask. We’ll get to that.

Does It Matter Which One I Use for Myself?

There’s a persistent idea that you need to pick a side and stay there. That consistency is required. That if you say “I’m autistic” on a forum on Saturday and “I have autism” in a work email on Monday, you’ve somehow undermined your own position.

You haven’t. Many neurodivergent adults use both, depending on context, audience, and - frankly - how they’re feeling that day. That’s not inconsistency. It’s code-switching, and most of us have been doing it our entire lives without having a word for it.

This is particularly relevant if you’re still relatively early in a late diagnosis. When you’ve known about your neurodivergence for six months, or two years, or even five, you might not have a settled relationship with the label yet. You might be trying it on in different contexts to see how it fits. Language often follows feeling, and feeling takes time. You don’t have to have your identity fully resolved before you’re allowed to talk about it. That would be an absurd prerequisite for speech.

There’s something I think of as linguistic masking - though I’m not sure anyone else calls it that. It’s the way some late-diagnosed adults default to person-first language in professional settings, not because they prefer it, but because it feels safer. Less confrontational. Less like you’re making your neurodivergence someone else’s problem. “I have ADHD” in a workplace email reads as disclosure. “As an ADHDer” in the same email reads as identity, and identity can feel like a bigger ask from the people around you.

Neither version is dishonest. You’re just choosing different registers for different rooms, which is something humans do constantly with everything from formality to accents to whether they swear in front of their partner’s parents. The only thing worth gently noticing is why you’re choosing - whether it’s genuine preference or a habit of making yourself smaller. And even then, noticing is enough. You don’t have to act on it today.

Why Do Some Autistic People Feel So Strongly About Identity-First Language?

The myth that this is “just semantics” - that it’s a language debate with no real-world consequences - is understandable but wrong.

Language shapes how people think about themselves, and more importantly, how systems treat them. The push for identity-first language, particularly in autistic communities, is tied directly to the neurodiversity movement and a rejection of what’s called the medical model of disability.

The medical model, briefly: the problem is in the person. The person is broken. Fix the person. The social model: the problem is in the environment. The person is different. Change the environment. Identity-first language tends to align with the social model - and this distinction has real consequences for how support, accommodations, and interventions are framed.

Consider what each model implies in practice:

  • Medical model framing leads to therapy aimed at making someone behave more normally, eye contact training, social skills groups that teach you to perform neurotypicality more convincingly
  • Social model framing leads to reducing sensory overload in the office, offering flexible working, stopping the bizarre cultural insistence that looking someone in the eye while they talk to you is a measure of your character

For someone working through workplace accommodations or trying to explain to HR why fluorescent lighting is not, in fact, a minor aesthetic preference, this framing shift matters practically. It changes what you’re asking for and how you’re allowed to ask for it.

Many late-diagnosed adults spent years - decades - trying to fix themselves. Trying harder. Masking better. Burning out and not understanding why. Identity-first language, for some of those people, is part of unlearning the idea that they were the problem all along. It’s not a trivial preference. It’s a reorientation.

That said, none of this means person-first language is harmful. Some people genuinely experience their condition as something they have rather than something they are, and that framing helps them. The point isn’t that one choice is enlightened and the other is backwards. The point is that both carry real meaning, and neither should be used on autopilot.

What Should I Say When I’m Talking About Someone Else?

The myth that asking someone’s preference is awkward or rude has caused more actual awkwardness than any amount of asking ever could.

“Do you have a preference for how I refer to your diagnosis?” takes five seconds. Most neurodivergent people appreciate it. Some won’t have a strong preference and will say so, which is also fine - you’ve still communicated that you thought about it, which is more than most people do.

What’s genuinely awkward is confidently using the wrong term and watching someone’s face do that very specific thing where they’re deciding whether correcting you is worth the social energy. It usually isn’t, so they say nothing, and now you’re both slightly uncomfortable for different reasons.

How to ask, depending on context:

  • In a professional setting (e.g., as a manager or HR professional): “I want to make sure I’m referring to your diagnosis in the way that feels right to you - do you have a preference?” Works well during an onboarding conversation or when discussing accommodations.
  • In a clinical or support setting: “Some people prefer ‘autistic person’ and others prefer ‘person with autism’ - is there a way you’d like me to refer to it?” Simple, neutral, takes ten seconds.
  • In a social or personal context: Listen first. Most people will use their own language naturally. If you’re not sure after a conversation, “How do you tend to talk about your diagnosis?” is low-stakes and genuinely curious rather than clinical.
  • In writing (articles, policies, communications): You can’t ask individuals, so check what the relevant community organisation uses. The National Autistic Society uses identity-first. So do a growing number of academic journals. A style guide from 2009 is not a reliable proxy for current community preference.

The asking is the respectful part, not the specific word you land on afterwards. Most people are not waiting to catch you out. They’re just hoping you thought about it.

In practical terms: if you’re writing something - an HR policy, a workplace communication, an article - person-first has been the institutional default for years. But many organisations are updating their style guides, and community-specific guidance is worth checking. If you’re writing about a specific community, look at what that community actually uses rather than defaulting to what a style guide from 2009 recommends.

In conversation, it’s simpler. Listen to what someone calls themselves and mirror it. If they say “I’m autistic,” say autistic. If they say “I have ADHD,” say that. If you’re not sure, ask.

And if you’re still working out what to call yourself - if the whole thing feels like a vocabulary test you didn’t sign up for - that’s fine. You’re allowed to be undecided. You’re allowed to try different words in different rooms and see which ones feel like yours. Language is a tool, not a loyalty oath. Use it however it’s useful, and put it down when it isn’t.