Capabilities and abilities overlap in everyday speech, but across most professional fields the two words carry genuinely different meanings. An ability is something you can do right now, a present-tense skill or function. A capability is broader: it includes not just what you can do, but the potential range of things you could do under the right conditions, with the right resources, or after the right development. That distinction sounds subtle, but it reshapes how economists measure well-being, how rehabilitation professionals assess patients, how business strategists think about competition, and how AI researchers describe what large language models actually accomplish. The gap between the two words turns out to matter more than most people assume.
Where the Two Words Actually Diverge
In casual conversation, calling someone “capable” and calling them “able” feel interchangeable. You might say a colleague is capable of managing a project or able to manage it, and nobody would notice a difference. The reason the words blur together is that both point at a person’s power to do something. But outside of casual talk, the emphasis shifts. An ability usually refers to a demonstrated or existing skill: you have the ability to swim because you already know how. A capability, on the other hand, points toward a wider set of potential achievements, including ones that depend on opportunity, context, or choice. You might have the capability to become a competitive swimmer given the right training, coaching, and pool access, even if your current ability is limited to a shaky backstroke.
This is not just semantic hair-splitting. Fields as different as economics, medicine, ecology, and military strategy have each settled on definitions that hinge on this same core idea: ability is what is, capability is what could be. That pattern keeps showing up because the distinction solves a real problem in each field. Measuring only what someone does right now misses the freedoms, constraints, and potential that shape their situation.
The Capability Approach in Economics and Human Development
The single most influential formal treatment of the word “capability” comes from the economist and philosopher Amartya Sen, whose framework has shaped decades of policy work on poverty, inequality, and well-being. In Sen’s terminology, a “functioning” is what a person actually does or is: being well-nourished, being literate, participating in community life. A capability, by contrast, is the freedom a person has to choose from the full set of functionings available to them.1Handbook of Social Choice and Welfare. Functionings and Capabilities Two people can have identical functionings on paper, say, both are currently unemployed, but vastly different capabilities if one has education, health, and social connections that make re-employment realistic while the other does not.
This reframing changed how development economists think about poverty. Traditional measures like GDP per capita or household income capture what people have. Sen argued that what matters is what people are free to do and become. A country where women are legally barred from driving has restricted their capability set even if household incomes are high. The United Nations Human Development Index, which blends life expectancy, education, and income, was directly inspired by this framework. It treats development not as growing the economy but as expanding human capabilities.
The practical payoff is real. Policies designed around the capability approach look different from policies designed around income alone. Instead of asking “how much money do people have?” they ask “can people actually convert their resources into lives they have reason to value?” That question forces attention to barriers like discrimination, disability, geographic isolation, and political exclusion, none of which show up clearly in income statistics.
Rehabilitation and the Clinical Meaning of Capacity
Healthcare and rehabilitation have their own version of the same split, formalized by the World Health Organization in its International Classification of Functioning, Disability and Health, known as the ICF. The ICF uses standardized qualifiers to describe what a person can do (capacity) versus what a person actually does in daily life (performance).2PubMed Central. Exploring clinician-reported assessments of capacity and performance qualifiers in the ICF: a scoping review A stroke patient might demonstrate the capacity to walk ten meters in a clinical setting with a flat floor and handrails, but their real-world performance could be far lower because their home has stairs, uneven surfaces, or no one to spot them.
The gap between capacity and performance is where rehabilitation targets its interventions. If capacity is high but performance is low, the problem is environmental: maybe the person needs home modifications, assistive devices, or caregiver support. If capacity itself is low, the focus shifts to therapy aimed at restoring function. Clinicians who confuse the two risk prescribing the wrong intervention entirely, improving someone’s physical ability when the real barrier is an inaccessible living situation.
A related confusion shows up between the clinical concept of capacity and the legal concept of competency. In healthcare, capacity refers to a person’s ability to understand, appreciate, and communicate their medical decisions, and it is assessed by a clinician at the bedside. Competency, however, is a legal determination made by a court, addressing whether someone can participate in legal proceedings or make binding decisions.3PubMed Central. Competency vs. Capacity: Legally Incompetent, Yet Medically Capable? A patient can be found legally incompetent by a judge, perhaps due to a prior guardianship ruling, while still having full medical decision-making capacity as assessed by their physician. The two evaluations answer different questions, use different procedures, and can reach opposite conclusions about the same person. Clinicians who treat the terms as synonyms risk overriding a patient’s actual decision-making ability based on a legal status that may no longer reflect their current mental state.
Dynamic Capabilities in Business Strategy
Business strategy adopted the word “capability” early and gave it a specific meaning that lines up with the broader pattern. An ordinary organizational capability is what lets a firm make a living in the short term: efficiently manufacturing a product, processing customer orders, running a supply chain. A dynamic capability, by contrast, is the firm’s ability to extend, modify, or create those ordinary capabilities as conditions change.4Strategic Management Journal. Understanding dynamic capabilities
The distinction matters because companies regularly fail not because they lack present-tense abilities but because they cannot adapt those abilities to new markets, technologies, or competitive pressures. A firm with strong ordinary capabilities in film photography could execute brilliantly on its existing product line and still collapse when digital photography arrived. What it lacked was the dynamic capability to reconfigure itself. In strategy research, this is treated as a separate and higher-order function, not just “being good at things” but “being good at changing what you’re good at.” That recursive quality is what makes the word “capability” do work the word “ability” cannot: it points to a system’s potential to evolve, not just its current state of functioning.
HR professionals increasingly use similar language. When organizations talk about building workforce “capabilities” rather than just hiring for current “skills” or “abilities,” they are making the same conceptual move: investing in the adaptability and breadth of what employees could contribute, not just matching today’s job descriptions to today’s skill sets. The shift is from static inventory (“what can this person do?”) to dynamic potential (“what could this person do if developed, redeployed, or given new tools?”).
How AI Research Uses “Capabilities”
The word “capabilities” has become almost unavoidable in discussions of artificial intelligence, particularly around large language models. Researchers describe these systems as exhibiting reasoning capabilities across domains like mathematics, logic, scientific question answering, and code generation, but with a significant caveat: it is not well understood how those capabilities develop or how robust they are across different situations.5Journal of Artificial Intelligence Machine Learning and Neural Network. Reasoning capabilities of large language models: a systematic review and meta-analysis of benchmarks, methods, and emergent behaviours (2018–2025)
The choice of “capabilities” over “abilities” here is deliberate, even if researchers do not always spell out why. Calling a model’s performance on a math benchmark an “ability” would imply a stable, reliable skill. Calling it a “capability” leaves room for the uncomfortable reality that the same model might solve a calculus problem correctly in one prompt and fail a structurally identical problem in the next. The word captures something more like a range of possible performance levels, dependent on context, prompt phrasing, and training data, rather than a fixed, reliable competence.
AI safety research pushes this even further. When safety researchers talk about “capability evaluations,” they are asking what a system could do under adversarial or unusual conditions, not just what it typically does on standardized benchmarks. A model’s dangerous capabilities might never surface during normal use but could emerge when the system is jailbroken or given tool access. The emphasis on potential rather than demonstrated performance mirrors the capability-versus-ability distinction in every other field.
Physical Capabilities, Maturation, and What Bodies Can Do
In sports science and kinesiology, “functional capacity” and “physical ability” overlap but are not treated as identical. A young athlete’s current abilities, what they can lift, how fast they sprint, are heavily shaped by where they fall on the maturation timeline. Research on adolescent handball players found large differences in strength, speed, and body size between early-maturing and late-maturing boys. The gap between early and late maturers was around 25 centimeters in height, over 33 kilograms in body weight, and about 20 kilograms in handgrip strength.6PubMed Central. The contribution of growth and maturation in the functional capacity and skill performance of male adolescent handball players Yet maturity status had no effect on sport-specific skills like ball handling and tactical execution.
The implication is that raw physical abilities at a given moment, strength, speed, size, mostly reflect biological maturation rather than training or talent. Meanwhile, sport-specific skills, the capabilities that actually determine competitive success over the long run, develop through practice and coaching, and poorer scores in those skills were linked to lack of training rather than to maturation timing. Coaches who select athletes based only on current physical abilities end up favoring early maturers and overlooking late-maturing players who may have greater long-term capability. The distinction between what a body can do right now and what it could become given time and development is essentially the capabilities-versus-abilities divide playing out on a playing field.
Evolutionary Capabilities and What Species Could Become
Evolutionary biology uses the concepts in a different but recognizable way. A species’ realized niche, the set of environmental conditions it actually inhabits, represents its demonstrated abilities in the wild. But the full range of conditions it could tolerate often extends beyond what it currently experiences. Spatial models of climate change typically assume species will only survive where their current realized niche remains available, but research has shown that species often have physiological limits well beyond their present habitat, and those limits can be further extended through evolutionary adaptation in response to selection pressure from a changing climate.7Global Ecology and Biogeography. Extending spatial modelling of climate change responses beyond the realized niche: estimating, and accommodating, physiological limits and adaptive evolution
This parallels the capabilities framing almost exactly. A species’ current distribution is its “ability” in the present tense: here is where it lives. Its fundamental niche, the full range of conditions its physiology could handle, is its “capability.” And its potential to evolve further under pressure extends that capability set into the future. Conservation models that account only for present-tense abilities, that is, the realized niche, will overestimate extinction risk for species with broad physiological capability and underestimate it for species that are already at their limits.
A vivid example of evolved capability comes from human endurance running. Humans are exceptional long-distance runners in hot conditions, a capability that is rare among mammals and unique among primates. This suite of features, from energy-storing tendons to efficient thermoregulation, probably emerged around two million years ago to help meat-eating hominids compete with other carnivores.8SpringerLink / PubMed Central. The evolution of marathon running : capabilities in humans The interesting point is that most modern humans never use this capability. Almost nobody persistence-hunts prey across a savanna anymore. The ability, in terms of daily practice, is absent for most people. But the underlying capability, the biomechanical and thermoregulatory architecture, is still there. A sedentary office worker and an ultramarathon runner share the same species-level capability set; they differ in realized ability.
Cognitive Reserve and the Brain’s Hidden Potential
Neuroscience offers one more angle that fits the pattern. Researchers studying cognitive aging have observed that some people maintain relatively strong mental performance even as their brains accumulate the kind of pathological changes, such as amyloid plaques or white matter damage, that typically predict decline. To explain this paradox, researchers invoke constructs like cognitive reserve, brain reserve, compensation, and brain maintenance.9PubMed Central. Brain reserve, cognitive reserve, compensation, and maintenance: operationalization, validity, and mechanisms of cognitive resilience
Cognitive reserve is essentially a capability concept. It is not a specific ability you can test with a single task but rather a latent capacity to recruit alternative brain networks or use existing networks more efficiently when damage occurs. Two people with the same amount of brain pathology can have very different cognitive outcomes because one has built up more reserve through education, occupational complexity, social engagement, or other lifetime experiences. Their measured abilities on a memory test might be identical today, but their capability to withstand future decline is not. The person with greater reserve has a larger set of cognitive strategies to draw on if and when further damage occurs.
This is why “brain training” apps and lifelong learning programs frame their pitch around building cognitive capability rather than specific memory or attention abilities. Whether those programs actually build reserve in the way that decades of formal education or complex occupational demands do is a separate and more contested question. But the underlying logic, that what your brain could do in the face of future challenges matters independently of what it does right now, is well-supported and follows the same pattern seen in every other domain.
When the Distinction Leads People Astray
One common mistake is treating capabilities as purely aspirational, as though having a capability means you will eventually realize it. In Sen’s framework, a person living in poverty may have the theoretical capability to become educated, but structural barriers can make that capability permanently unrealized. In business, a firm can possess dynamic capabilities on paper, say, a well-funded R&D department, and still fail to adapt if organizational culture or leadership blocks change. The word “capability” describes potential, not destiny. Confusing the two leads to a kind of optimistic negligence, assuming that because the pieces exist in principle, results will follow without intervention.
The opposite error is just as harmful: reducing everything to present abilities and ignoring capability entirely. This is how talent identification in youth sports ends up biased toward early-maturing kids, how conservation models overcount species headed for extinction, and how clinical assessments can underestimate what a patient could achieve with the right support. If you only measure what someone can do today, you miss the leverage points where a modest change in conditions could unlock a much larger change in outcomes. The capabilities lens is, at bottom, a way of asking: what would be possible if the constraints were different? That question, applied to any domain, tends to produce better decisions than simply inventorying current performance.