affordances
Compiled with AI assistance and selectively reviewed. Please consult the cited passages; the books remain the authoritative source. About this companion
Editorial synthesis · provisional
Definition
Affordances are possibilities for action that arise between an agent’s capacities and a particular environment. The same feature can offer an action to one body and deny it to another. In the ecological account developed in the manuscripts, these relations are real even when unnoticed; they include invitations, obstacles and risks rather than merely desirable options or properties belonging to objects alone.
Explanation
AB’s affordances-for emphasizes that there is no body-independent inventory of what a room offers. Geometry, strength, skill, fatigue, tools and social provision can change what is possible. The opportunity precedes its actualization and need not be perceived correctly.
RB2 distinguishes an affordance from the ecological information that specifies it and from the organism’s corresponding effectivity. Direct perception is Gibson’s debated claim, not an unqualified guarantee. The stair-climbing ratio is an attributed study result, limited by the later strength-and-balance example; the backpack example expressly illustrates an idea more than it proves a perceptual effect. Learning, calibration and tools can alter or disclose possibilities.
Technical applications retain their own status. RB2 describes robotic representations of action–object–effect relations, then distinguishes represented value, installed weighting and a system’s own stakes. Affordance learning supplies no automatic passage to mattering. PB’s hom-set treatment is a qualified formal representation of what a body can do; the surrounding AI-assisted skeleton requires verification and does not supply felt experience. These models are not identical with the real relations in Gibson’s account.
RB1 uses broader possibility-field and developmental metaphors without supplying a local ecological definition. RB3 extends the relational pattern to institutions and technical environments that distribute, enable or obstruct agency. What a regime affords and who bears its costs remain ethical questions, not guarantees that available actions are good.
Scope and attribution
AB and RB2 provide direct complete definitions. PB and RB3 provide compressed practical formulations and same-book development; RB1 has broader contextual extensions rather than a local ecological definition. Technical representations and PB’s formal model retain distinct status and are not identities with ecological relations.
Attribution
AB — The Actual Book
AB defines affordances as real action possibilities relative to a particular body and environment. Its affordances-for formulation rejects an agent-independent inventory, while the stair, gap and branch examples show differences of capacity and risk. Possibilities precede their actualization and need not be noticed. Dimensions and skills both matter; the wheelchair example also preserves the role of ramps, lifts and social provision. The same book applies the vocabulary to regimes that enable or constrain action, without equating constraint with domination.
The exact glossary definition and body text are source evidence, not a new author-approved lexicon definition.
The staircase claim is the manuscript’s empirical report, with Warren/effectivity/direct-perception debate named in its notes. Skill and capacity also alter the answer; no universal numerical limit is inferred.
Affordances, bodily time orientation and neural prediction converge without becoming interchangeable; an invitation is not a guarantee of the action’s occurrence.
The analogy to unrealized possibilities at other scales is the author’s extension; it does not redefine every quantum possibility as an ecological affordance.
Regimes distribute possibilities and costs, while constraints can enable agency as well as limit it; this does not make every constraint domination.
Authorial body and glossary definitions with Gibson-attributed origin, explicit empirical-note context and qualified ethical extension.
Occurrence locators: 42, 70, 85, 86, 89, 92, 93, 94, 95, 102, 103, 107, 109, 110, 113, 117, 174; Glossary: 223; Thinker register: 229; Notes: 241, 242, 245.
PB — The Possible Book
PB’s authorial manifesto describes a living field of what can be eaten, avoided, touched, entered, repaired or lost, adding fear or love only for a life that feels. The Tractatus treats affordances as relations between a soliciting world and a body that can, with capacities and limits shaped by tools and situation. Categoria represents what a body can do through outgoing morphisms and a hom-set, but explicitly requires verification of its AI-assisted formal skeleton. The reproduced AI exchange remains a separate provocation.
The same stair can be available to one body and a limit to another; the body and environment jointly condition the relation.
Hom-set-as-affordance is labelled established/plausible in the source; dagger-as-flesh is separately labelled unverified. The global skeleton remains a draft, not a proof, and hom-sets do not supply what-it-is-like or a binding ought.
AI claims about generating affordances or negotiating mutual affordances remain reproduced AI voice and do not establish an actual agent’s own stakes or a new authorial definition.
Authorial manifesto/Tractatus and a separately qualified authorial formal proposal; reproduced AI source voice stays distinct.
Occurrence locators: 36, 98, 146, 147, 153; Reproduced AI exchange: 16, 17, 18.
RB1 — The Real Book, Volume I: The Possible Is Real
RB1 uses affordance language for structured fields of possibilities, embodied ways of thinking, genetic/developmental potential and environmental/institutional relations. Its description of nature’s affordance for order is an authorial possibility metaphor in the scientific-inquiry discussion. These passages provide cross-domain context for the relation between capacity, conditions and what can happen, but no local ecological definition or quantitative affordance criterion.
The source itself presents affordances among metaphors for becoming. These broad uses do not authorize an ecological definition for quantum events or every unrealized possibility.
Institutional and historical applications retain their own level and mechanisms; a metaphor of order does not make every successful theory or institution the same kind of affordance.
Authorial philosophical and scientific-context extensions; no wholesale attribution to Gibson or independent experimental validation.
Occurrence locators: 10, 236, 239.
RB2 — The Real Book, Volume II: Becoming Is Being
RB2’s ecological core treats affordances as real relations between an organism’s abilities and an environmental configuration. Their existence differs from their correct perception, and ecological information specifies them without being the affordance itself. Body dimensions, dynamic effectivities, learning, tools and context shape what can be done. Direct perception remains Gibson’s debated position; the stair ratio is an attributed result with a strength/balance qualification, and the backpack example has an explicit experimental-demand caveat. Robotics and design model or expose possible actions, but the source expressly separates learned representation and installed weighting from a system’s own mattering. Juarrero’s dynamical interpretation emphasizes the constraints that shape possibilities while the parallel theories remain distinct.
Relative to an organism does not mean created by being perceived. The same opportunity or threat may be overlooked or misjudged.
Gibson’s direct-perception account is distinct from the claim that affordances exist, and its empirical reach remains debated; information pickup is not the action possibility itself.
The approximately 0.88 leg-length ratio is the manuscript’s report of Warren’s study, not a universal threshold. The later older-adult example adds strength/balance limits, and the backpack example is expressly attributed to experimental demand rather than perception.
Effectivities name the organism’s capacities; body-scaled and action-scaled affordances differ in emphasis, while both remain relational possibilities. Their existing specialized candidates are not merged here.
Learning and calibration can fail to transfer across a new posture or skill. Tools can create new effectivities and affordances; mere perception of a possibility does not guarantee its realization.
Technical action–object–effect representations and simulations are models or applications. The source says affordance learning supplies representation and installed weighting, not the third achievement of mattering. Homeostatic and interoceptive architectures are proposals, not evidence that present learning already produces experience.
Ecological psychology, enactivism and phenomenology remain parallel interpretations. The Juarrero account describes relational constraints shaping possibilities and likelihoods, rather than a universal equation for affordance.
Umwelt is the lived domain of significance, while milieu names wider viability conditions. Social histories of injury, training, exclusion and support affect what is available; relationality does not imply arbitrary projection.
Language and new forms can open possibilities that were not previously actionable; this does not turn the possible into a ready-made, fixed inventory.
Author’s Gibson/Chemero-supported ecological account with separate Turvey, Warren, developmental, technical, design and Juarrero treatments; source quotations and scientific limitations remain explicit.
Occurrence locators: 3, 4, 26, 48, 49, 50, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 107, 112, 116, 117, 137, 138, 143, 145, 146, 147, 148, 154, 156, 158, 166, 177, 178, 181, 199, 233, 246, 273, 281, 287, 291, 293.
RB3 — The Real Book, Volume III: Murmurations
RB3 attributes the practical appearance of handles and stairs to Gibson’s affordance account, then applies action-possibility language to technical and institutional environments. Regimes distribute invitations, pressures, prohibitions and expectations; infrastructures can make some conduct easy, costly or impossible, while mutual aid can afford repair. These are ethical applications of relational possibilities rather than a new ecological definition of every social system. The source asks who or what pays for what a regime enables, and separates institutional answerability from a model’s own vulnerability. Reproduced AI appendix claims remain attributed provocation.
Affordances concern capacities and limits, not neutral geometry alone.
Institutional correction need not transfer human vulnerability to a model. The claim about world-coupled understanding is contested; possible engineered stakes are left open.
Technical environments reshape purposes and agency, and their material support and distributed costs remain part of ethical assessment. Available actions or stabilized affordances are not guarantees of moral success.
Moral attunement is a learned, transindividual ethical sensitivity to salient opportunities, risks and asymmetries. It remains a separate registered concept, not a synonym for affordance itself.
The appendix’s generated-affordance and mutual-affordance claims are reproduced AI text, not independent evidence of actual artificial experience.
Gibson-attributed practical account and authorial ethical/technical applications, with separately attributed AI appendix.
Occurrence locators: 155, 156, 173, 210, 218, 220, 222, 224, 228, 244, 279, 284, 285, 295; Reproduced AI exchange: 399, 400, 401.
Sense distinctions
Affordances — capacity-relative possibilities for action
Retain one source-confirmed capacity-relative action-possibility sense, with explicit specialized, technical, formal and social treatments.
The direct definitions consistently describe what can be done in relation to a particular agent and environment. Body-scaled, action-scaled, tool-mediated and social cases vary the capacities and conditions rather than introduce a different referent. The robotics section explicitly calls its learned structures representations of action/object/effect relations and protects the analogy with an own-stakes distinction. PB’s hom-set mapping also represents what the body can do and retains local and global verification caveats. These modeled uses must not be substituted for the real ecological relation, but their presence does not by itself establish an independent formal measure or a different lexical core. RB1’s more general possibility vocabulary is expressly metaphorical/contextual. Separate existing effectivity, perception, calibration, robotic-learning and competition candidates retain their identities and future assessment.
Background used in the editorial explanation
Domain-specific mechanisms and wider analogies remain distinguished. Original definitions retain their scope; local cross-references do not rewrite them. Named theories and mechanisms are distinguished from neighboring background, homonyms and the author’s own synthesis. Reproduced AI passages retain their separate voice.