Who Designed the System? | The Uncomfortable Questions #1

Amanda Carroll

Aug 20

5 min read

Updated: Sep 3

THE UNCOMFORTABLE QUESTIONS | QUESTION #1

Who designed the system?

The short answer

No single person designed the systems autistic people encounter. Schools, workplaces, health care, public services, technology, funding rules, risk management, and social expectations accumulated in layers. Each layer encoded assumptions about what a normal learner, communicator, worker, patient, or citizen would look like.

When access fails, the person may need support. The system may need translation. The environment may need redesign. Often, the honest answer includes all three.

Most systems do not have a single architect. They have layers.

A classroom routine begins because it makes a large group easier to manage. A clinic form begins because the software needs a required field. A workplace rule begins after one bad incident. A social expectation survives because everyone who can follow it forgets it is there. Over time, those layers harden into common sense.

Common sense is often just an assumption that stopped being questioned.

The assumed user

Every system imagines a user, even when nobody writes that user down. The imagined student can sit in fluorescent light, shift attention when a bell rings, learn from spoken group instruction, answer quickly, and show engagement in recognizable ways. The imagined patient can wait, tolerate touch, explain symptoms under stress, and answer a stranger’s questions in order. The imagined employee can infer tone, navigate office politics, change tasks without warning, and perform enthusiasm.

Many autistic people can do some of those things. Some can do them for a while. Some can do them at a cost nobody sees. Some communicate, learn, regulate, or participate in entirely different ways.

The mistake is treating one access pattern as evidence of intelligence, effort, safety, or worth.

When design becomes diagnosis

A diagnosis can open access to support. It can also be used as a container for every mismatch. If a person cannot use a service as designed, the explanation becomes autism. The environment disappears from the analysis.

That is incomplete.

Research on autistic and non-autistic communication shows that information transfer and rapport can depend on the match between people, not simply on a deficit located in the autistic person. Research on first impressions shows that observers’ judgments help create social exclusion. Disability law and universal-design frameworks likewise recognize that access depends on what environments provide.

None of that means disability is imaginary. It means the location of the problem matters because it changes the solution.

Separate the goal from the ritual

The most practical question is: What is the real goal?

If the goal is learning, eye contact may be irrelevant.

If the goal is informed consent, speech may be unnecessary.

If the goal is workplace performance, attendance at an unstructured social event may be irrelevant.

If the goal is safety, immediate compliance may be less reliable than clear language, processing time, reduced sensory load, and a known communication method.

Some requirements are necessary. Many are habits dressed as necessities.

A better design sequence

Observe before interpreting.
Name the actual goal.
Identify every demand the person must meet to reach it.
Remove demands that do not serve the goal.
Offer more than one way to receive information, respond, regulate, and participate.
Ask the person what works.
Measure access and outcome, not performance of normality.
Review unintended harm.

The system can learn

The point of asking “Who designed the system?” is not to find a villain. It is to end the fiction that familiar systems are neutral.

People built them. People inherited them. People can examine them. People can change them.

When access fails, the person may need support. The system may need translation. The environment may need redesign. Usually, the honest answer includes more than one of those things.

Different is not failure. Sometimes the design is.

Educational boundary

Wonder Haven Autism Advocacy provides educational information and practical support. We are not medical doctors, licensed mental health professionals, therapists, or diagnosticians. This material is not medical advice, mental health treatment, therapy, diagnosis, or a substitute for individualized care from a qualified professional.

What the research supports

Communication breakdown can be relational. In a 2020 diffusion-chain experiment, autistic-only information chains did not differ significantly from non-autistic-only chains, while mixed-neurotype chains lost more detail and reported lower rapport. This supports a mismatch interpretation in that setting; it does not explain every communication difficulty. Crompton et al. (2020)
Observers help create social outcomes. A 2017 series of first-impression studies found that non-autistic observers rated autistic people less favorably from brief audio-visual samples and were less willing to interact. The difference was not present when observers read transcripts alone. Sasson et al. (2017)
Community priorities can differ from institutional agendas. A 2024 survey co-designed by autistic and non-autistic researchers asked 225 autistic adults in Scotland to rank research priorities. Mental health and wellbeing, identification and diagnosis, support services, public knowledge and attitudes, and issues affecting autistic women ranked highest. Cage et al. (2024)
Universal design treats barriers as design variables. CAST's UDL Guidelines 3.0 focus on reducing barriers rooted in bias and systems of exclusion while preserving meaningful learning goals. UDL is an education framework, not an autism treatment or a legal standard. CAST UDL Guidelines 3.0

What this evidence does not prove

It does not prove that every institution or professional acts with malice.
It does not prove that autism is only an environmental mismatch or that disability disappears when an environment changes.
It does not mean diagnoses, safety rules, specialized supports, or individual skill-building are inherently wrong.
It does not justify replacing one rigid rule with another. People, settings, risks, communication methods, and support needs differ.

How to read this record

Verified evidence: The linked studies and official policy sources report the findings or duties summarized here.
Wonder Haven's analysis: The system-design framework is our synthesis of that evidence and lived patterns; it is not itself a clinical finding.
Allegations: This article does not present an allegation against a named person or institution as an established fact.
Unresolved: The effects of design choices differ across people and settings, and policy implementation changes over time.

What we are still researching

Which institutional demands create the greatest preventable harm for autistic people and families in West Texas.
How barriers and useful accommodations differ across communication methods, intellectual disability, race, gender, age, diagnosis access, and support needs.
Which changes improve access without increasing surveillance, segregation, or pressure to perform normality.
What autistic people themselves define as success in each setting.

Sources and further reading

Evidence and corrections

Evidence cutoff: August 20, 2026. Law, policy, research, and implementation can change. This article is educational analysis, not legal advice. Wonder Haven will correct material errors and update dated claims when stronger or newer primary sources become available.

Bring the next question

What system should we examine next? BRING THE NEXT QUESTION

Keep Following the Questions

All Uncomfortable Questions

Next: Too Smart to Need Help: When One Score Becomes the Gate →