HSW-0023
You can possess a capability that nobody else can reliably see.
A student may genuinely understand a concept but fail to show it under the assessment format. A worker may have learned a valuable skill informally but lack a recognised credential. A candidate may have completed a course but still be unable to demonstrate the promised capability. A portfolio may contain impressive work but make it difficult for an outsider to tell what the person actually contributed.
This is a problem of capability legibility: how real ability becomes visible, interpretable and trustworthy to another person or system.
Studying usually focuses on building capability. Eventually capability has to cross a boundary. A teacher, examiner, university, employer, client, team or institution must be able to recognise what is there.
Capability and evidence are not the same thing
The first distinction is essential.
- Capability is what you can actually do.
- Evidence is an observable trace of that capability.
- A signal is the compressed message someone else uses to infer capability.
- A standard helps different people interpret the signal in a similar way.
- Trust determines whether the receiver believes the evidence.
A mark is a signal. A certificate is a signal. A portfolio is a collection of evidence. A live demonstration is evidence. A recommendation is evidence filtered through another person’s judgement. None is identical to the capability itself.
This is why eduKate separates Proof of Capability from the learning process. Proof asks, “What evidence would justify belief that the capability exists?” Legibility adds another question: “Can the receiver correctly understand that evidence?”
School marks are compressed descriptions
An examination score compresses a large performance into a small number. Compression is useful. Institutions could not read every student’s entire learning history before making every decision.
But compression loses information.
Two students with the same mark can have different profiles. One may be conceptually strong but slow. Another may be fast and accurate on familiar tasks but weak on transfer. One may have excellent knowledge with fragile checking. Another may have a narrower knowledge base but very stable execution.
This is connected to HSW-0019: Measurement Error and HSW-0020: Benchmarking. Measurement is necessary, but every measurement needs interpretation.
A legible capability needs a shared language
If one school, one company and one training provider use entirely different words for the same capability, movement becomes expensive. Learners have to repeatedly prove what they can do. Employers struggle to compare candidates. Training providers may teach overlapping material under different labels. Credentials become hard to interpret.
This is one reason standards, notation and common terminology matter. eduKate’s HSW-0017: Knowledge Standardisation explains how shared units, definitions, notation and credentials help knowledge travel. Capability legibility is the receiving-side problem: does the destination understand what the standardised signal means?
In June 2026, the OECD’s work on A Skills-First Labour Market highlighted the need for a common skills language, modular and micro-credential learning, skills-first human-resource practices, career guidance and recognition of prior learning. These mechanisms are attempts to make capability easier to describe and connect to opportunities.
The danger: optimising the signal instead of the capability
Once a signal matters, people learn to optimise for it.
Students can learn to chase marks without building durable understanding. Applicants can collect certificates with little transfer. Portfolios can become polished displays that hide weak independent execution. Institutions can prefer signals that are easy to count rather than capabilities that are harder to measure.
This is a classic systems problem. A measure begins as evidence about a goal and gradually becomes the goal.
The solution is not to abandon measurement. It is to keep the evidence chain connected to independent performance.
A trustworthy evidence chain
A strong capability signal usually has several properties.
- Relevant: the evidence resembles the capability being claimed.
- Independent: the learner can perform without hidden support doing the difficult work.
- Repeatable: success is not a one-off accident.
- Current: the evidence is recent enough for the capability to still matter.
- Interpretable: the receiver can understand the standard and context.
- Verifiable: the evidence has enough provenance to be checked.
- Transferable: where appropriate, the capability survives a new task or environment.
No single evidence form is perfect. A robust system often triangulates.
Examinations make some capabilities highly legible
Standardised examinations have powerful advantages. They create common conditions, common questions, common marking rules and a scalable comparison mechanism. For certain academic capabilities, that makes evidence highly legible.
But examinations sample. They do not observe every possible application. They are often time-bounded. They may privilege capabilities that fit the format. This does not make them invalid. It means the receiver should understand the boundary of the signal.
For the student, this means examination training is partly a translation problem. You must convert what you know into the form the assessment can read: the right method, notation, command-word response, evidence, structure, working and timing.
That is not cosmetic. If capability cannot cross the assessment interface, the marker cannot award invisible knowledge.
Portfolios make different capabilities legible
Some capabilities are difficult to capture in a timed paper: design, sustained research, iteration, long projects, creative production, teamwork, programming, laboratory practice or the ability to improve through feedback.
A portfolio can expose richer evidence. But it creates new interpretation problems. Which work is representative? What did the learner do personally? What tools were used? How much guidance was provided? Was the result revised over weeks or produced independently under pressure?
Good portfolios therefore need provenance: task, constraints, role, process, feedback, revision and final output. The reader should not have to guess what the artifact proves.
Informal learning is often capable but illegible
People learn continuously outside formal courses: at work, through hobbies, in communities, through self-directed study and while solving real problems. The capability can be genuine even when the learning route does not issue a certificate.
In June 2026, the OECD published Giving Informal Learning the Recognition it Deserves, describing informal learning as a major but often overlooked part of lifelong learning. Recognition of prior learning is essentially a legibility mechanism: it creates a process through which capability built outside formal pathways can become interpretable inside formal systems.
This matters for students because the future of learning will not be one clean line from school to qualification to permanent expertise. Capability will accumulate across formal education, self-study, work and changing tools.
AI changes the evidence problem
When tools can generate text, code, summaries, explanations and analyses, a finished artifact becomes less informative by itself. The receiver may need to know what the person understood, what the tool supplied, what decisions the person made and whether the result can be defended or reproduced.
This does not make tool-assisted work worthless. Real workplaces have always used tools. It changes the evidence required.
- Can the learner explain the reasoning?
- Can they verify the output?
- Can they adapt it when conditions change?
- Can they identify a failure?
- Can they reproduce the important part without the exact previous prompt?
- Can they state what remains uncertain?
This connects capability legibility to HSW-0013: Verification Burden. Easier generation increases the value of evidence about judgement.
The student needs two ledgers
A useful learner keeps two mental ledgers.
- Capability ledger: What can I independently retrieve, explain, solve, create, judge and transfer?
- Evidence ledger: What credible evidence do I have that another person could use to recognise those capabilities?
The ledgers should not be confused. An empty evidence ledger does not prove no capability exists. A full evidence ledger does not guarantee the capability is strong.
Studying builds the first ledger. Assessment, demonstration and documentation help build the second.
How to make capability more legible without becoming performative
- Define the capability in observable terms. Replace “good at Science” with a more specific claim.
- Choose evidence that matches the claim. A vocabulary quiz cannot prove experimental design.
- Show conditions. State time, tools, collaboration and support.
- Use standards where useful. Shared rubrics and recognised criteria reduce interpretation cost.
- Keep provenance. Preserve drafts, working, decisions and version history for complex artifacts.
- Retest. Capability should survive a later attempt.
- Transfer. Where the claim is broad, test a new context rather than the memorised example.
Schools, tutors and parents should describe the capability behind the mark
A mark can guide decisions, but the most useful feedback often adds interpretation: what is secure, what is fragile, what failed, what changed and what the learner can now do independently that they could not do before.
This is especially important during transitions. A new teacher, school or tutor needs more than a headline score. That is why HSW-0009: Learning Handover treats learner state as something that must survive a change of system.
Better legibility reduces the amount of learning that disappears simply because the next person cannot see it.
From school to the world
The World Bank’s HCI+ framework, launched with the 2026 Human Capital Report, extends measurement of human capital through higher education, youth transition into work and adult learning in employment. That reflects a wider truth: capability keeps accumulating after school, and systems need better ways to understand those gains and losses.
Source: World Bank, Human Capital Index Plus 2026 methodology.
For the individual learner, the implication is practical. Build real capability first. Then learn how the relevant world reads evidence.
Do not become excellent only at signalling. Do not become excellent only in private.
The legibility rule
If a capability matters, it should eventually be possible to answer three questions.
- Can you do it?
- Can you show that you can do it?
- Can the receiver correctly understand what the evidence means?
Studying creates capability. Good assessment creates evidence. Good standards create a shared language. Good judgement keeps the signal connected to the real thing.
Continue through the eduKate studying system: Study & Learning Methods Hub · Proof of Capability · Knowledge Standardisation · School-to-World Handoff