Item (Copy)¶

Promote the single physical or digital copy to a first-class bibliographic entity at the bottom of the Work–Expression–Manifestation–Item ladder, so copy-specific data and operations attach to one instance rather than the shared edition.

Core Idea¶

An Item, in FRBR (Functional Requirements for Bibliographic Records) and its successors IFLA-LRM and BIBFRAME, is the most concrete tier of a four-level bibliographic instantiation hierarchy: a specific, singular physical or digital copy that embodies a particular Manifestation (a specific publication, edition, or format), which in turn embodies an Expression (a distinct realization of a Work), which in turn realizes the Work (the abstract intellectual or artistic creation). The Item is the bottom of the ladder — an individual copy as a single object with its own identifier, location, and history.

The structural move that FRBR makes is to promote the Item to a first-class entity with its own record and identifier, separable from the Manifestation record it participates in. This matters operationally because items and manifestations carry different properties and are subject to different operations. The Manifestation record holds publication-level data shared across all copies: publisher, edition statement, ISBN, format, pagination. The Item record holds copy-specific data that belongs to this particular instance: call number, barcode, acquisition date, shelf location, current circulation status, condition notes, provenance, and any copy-specific annotations or stamps. When a borrower checks out a book, the loan is recorded against a specific Item, not against the Manifestation — because it is a specific copy on loan, not the abstract edition. When a conservator notes water damage, the note belongs to the affected copy, not to all copies of that printing.

In a library catalog implementing this model, a single Manifestation record (say, the Penguin Classics 2003 paperback edition of Pride and Prejudice, ISBN 978-0141439518) may link to dozens or hundreds of Item records — one per physical copy held by the institution. An interlibrary loan query ("do you hold this?") is answered at the Item level, because availability requires knowing not just whether the library holds an edition but whether a specific copy is currently uncirculated. Withdrawal of a damaged copy retires one Item record while leaving the Manifestation and all sibling Items intact.

Analogous Item-level distinctions appear in museum object cataloging (the accession-numbered artifact versus the type or workshop that produced it), in archival finding aids (the specific folder in a specific box versus the series it belongs to), and in digital asset management (the specific file at a specific storage path with a specific checksum versus the intellectual content it embodies). In each case the pattern is the same: one physical or digital instantiation, identified individually, carrying instance-specific metadata, connected upward to a shared higher-level record. This is the FRBR Item concept — the bibliographic tradition's operationalization of the most concrete level of a four-tier Work–Expression–Manifestation–Item hierarchy.

Structural Signature¶

Sig role-phrases:

the singular instantiation — one specific physical or digital copy, the most concrete tier, with its own identifier, location, and history
the four-tier ladder — Work (abstract creation) → Expression (realization) → Manifestation (publication/edition/format) → Item (copy), each rung progressively more concrete
the embodied Manifestation — the shared publication the Item instantiates, holding publisher, edition, ISBN, pagination common to all copies
the Item-first-class promotion — the move that gives the copy its own record, separable from the Manifestation it participates in
the instance-specific metadata — call number, barcode, accession date, shelf location, circulation status, condition, provenance belonging to exactly one copy
the property-placement test — the operational rule: invariant across copies → Manifestation; can vary copy to copy → Item
the cardinality constraint — one Manifestation binds many Items; each Item embodies exactly one Manifestation
the bounded-operation guarantee — a loan, withdrawal, or conservation note attaches to a single Item, leaving siblings and the Manifestation untouched; availability is settled only at the Item tier
the four-tier limitation — the Expression/Manifestation intermediate rungs and bibliographic operations are domain-bound; the underlying type/token (class/instance) distinction is the parent that travels

What It Is Not¶

Not the Manifestation. An Item is this book, not this kind of book — a singular copy, not the edition it embodies. Publication-level data shared across all copies (publisher, edition statement, ISBN, pagination) lives at the Manifestation; the Item carries only what is copy-specific. Attaching a copy fact to the Manifestation wrongly claims it of every copy — a condition note placed there would assert that the whole printing is damaged.
Not the Work or the abstract content. The Item is the most concrete rung of the four-tier ladder, the physical or digital token — not the intellectual or artistic creation (the Work), and not its realization (the Expression). It has its own identifier, location, and history precisely because it is an individual instantiation; reading it as the content it embodies collapses the bottom of the instantiation hierarchy into its top.
Not where shared, publication-level data belongs. The Item holds only instance-specific metadata — call number, barcode, accession date, shelf location, circulation status, condition, provenance. Data that is invariant across copies belongs higher up and is inherited by every Item; duplicating publication-level facts onto each copy (or, worse, cloning a whole record per copy) is the modeling error the property-placement test exists to prevent. The bookkeeping should scale with genuine variation, not with copy count.
Not implied by holding the edition. Holding a Manifestation guarantees the abstraction is held but says nothing about whether any individual copy is free; availability is settled only at the Item tier, where individual copies are distinguished. An interlibrary-loan "do you hold an available copy?" needs a specific uncirculated Item, not merely the edition — so "we hold this edition" and "we have a copy available" are different claims living at different levels.

Scope of Application¶

The Item lives within bibliographic and cultural-heritage modeling — the subfields aligned with the FRBR / IFLA-LRM / BIBFRAME four-tier Work–Expression–Manifestation–Item ladder, where the most concrete instantiation is promoted to a first-class entity and copy-level operations bind to one copy; its reach is within that instantiation-hierarchy domain (the class/instance, serial-numbered-banknote, drug-lot, and build-artifact cousins are co-instances of the parent, the type/token distinction, which most need only at a two-level split, not FRBR's intermediate Expression and Manifestation rungs).

Library cataloging — the home: a single Manifestation record links to per-copy Item records, so loans, withdrawals, and conservation notes are recorded against a specific copy and "how many do we hold / which are circulating?" become well-typed Item-level queries.
Museum object records — the accession-numbered artifact (the Item) versus the type or workshop that produced it (the Manifestation-level "type"), carrying instance-specific acquisition history, condition, and conservation events.
Archival finding aids — the specific folder in a specific box (the Item) versus the series and subseries it belongs to, the higher-level groupings.
Digital asset management — the specific file at a specific storage path with a specific checksum (the Item) versus the intellectual content it embodies, settling availability and integrity at the instance tier.
Interlibrary loan and holdings queries — the availability claim resolved at the Item level, because "do you hold an available copy?" needs a specific uncirculated copy, not merely the edition.

Clarity¶

Naming the Item makes legible the difference between this book and this kind of book — the line a flat catalog record constantly blurs. Without the concept, copy-specific facts have no settled home: a call number, a barcode, a water-damage note, a circulation status, an ownership stamp all get attached to "the record," and it is unclear whether they describe the edition or the object on the shelf. Promoting the Item to a first-class entity gives each property a level to live at, and the cataloger's recurring question — "where does this datum belong?" — becomes answerable by rule: publication-level data (publisher, edition statement, ISBN, pagination) sits at the Manifestation and is shared across all copies; instance-level data (location, condition, accession date, provenance, loan status) sits at the Item and belongs to exactly one copy. The four-tier Work–Expression–Manifestation–Item ladder turns a single muddled "where does this go?" into four well-typed slots with operational tests for each boundary.

The payoff is that copy-level operations and queries acquire a precise referent. The concept makes legible why a loan is recorded against an Item rather than a Manifestation — a specific copy is out, not the abstract edition — and why withdrawing a damaged copy retires one Item while leaving its Manifestation and sibling Items untouched. Questions that a flat record cannot answer well — "how many copies do we hold?", "which copies are currently circulating?", "which copies carry annotations?", and the interlibrary-loan "do you hold an available copy?" — become well-formed, because each is now explicitly a question at the Item level rather than the Manifestation level. The practitioner can ask the sharper question the model is built to support: does this property, this operation, or this availability claim attach to the individual instantiation or to the shared publication it embodies? That single placement decision is what the named Item lets a cataloger get right, and it is the boundary the whole instantiation hierarchy exists to keep clean.

Manages Complexity¶

A holding of any scale is a mass of facts about many copies of many editions — call numbers, barcodes, loan statuses, condition notes, provenance stamps, accession dates — and in a flat record this mass has no natural organization: copy-specific data either duplicates across every copy of an edition or smears into a single record that cannot say whether a datum describes the printing or the object on the shelf. The four-tier Work–Expression–Manifestation–Item ladder, with the Item promoted to a first-class entity, compresses that disorder by giving every property and every operation exactly one level to attach to, decided by a single typed test: does this fact belong to the shared publication or to the individual instantiation? Publication-level data lives once at the Manifestation and is inherited by all its copies; instance-level data lives at the Item and belongs to exactly one. A library holding hundreds of copies of one edition is then not hundreds of redundant records but one Manifestation record plus per-copy Item records carrying only what differs, so the bookkeeping scales with genuine variation rather than with copy count. Copy-level questions — how many held, which circulating, which annotated, is an available copy on hand for interlibrary loan — collapse from ill-posed queries against a muddled record to well-typed queries at a known tier. The sprawling "where does each datum go and how do I count what I hold" problem reduces to one placement rule and a clean partition of properties across four levels.

Abstract Reasoning¶

Holding the Item as a first-class tier under Manifestation, with a clean partition of properties across the four-level ladder, licenses a set of concrete cataloging inferences.

Property-placement — type a datum to its tier by an operational test. The workhorse inference runs from a fact to the level it belongs at: ask whether the datum would be the same for every copy of the edition or whether it can differ copy to copy. If it is invariant across copies — publisher, edition statement, ISBN, pagination — it belongs at the Manifestation and is inherited by all Items; if it can vary copy to copy — call number, barcode, shelf location, circulation status, condition, provenance, copy-specific annotations — it belongs at the Item. The test is decidable and the placement follows from it, so "where does this go?" stops being a judgment call and becomes a typed assignment. The same test, run upward, separates Expression-level facts (translation, significant revision) from Manifestation-level ones (print run, format) and Work-level facts (the abstract creation) from all below — the ladder gives each property exactly one rung, and mis-shelving a datum is detectable as a tier mismatch (a condition note attached to a Manifestation, say, would wrongly claim every copy is damaged).

Diagnostic — read a modeling fault off a query that misbehaves. When "how many copies do we hold?" cannot be answered, or returns the count of editions instead of objects, infer the Item was never promoted to a first-class entity — copy-specific data is smeared into a flat Manifestation record that cannot distinguish "this book" from "this kind of book." When a single edition is represented by hundreds of near-duplicate records, infer the inverse error: instance variation was modeled by cloning the whole record rather than by one shared Manifestation plus per-copy Items carrying only what differs. When a loan, a withdrawal, or a damage note appears to affect an entire edition, infer the operation was attached at the Manifestation level when it should have been recorded against a single Item. Each query symptom (uncountable copies, redundant records, edition-wide side effects) points to a specific placement or promotion fault in how the hierarchy was instantiated.

Interventionist — predict the scope of a copy-level operation. Because copy-level operations attach to a single Item, the cataloger reasons forward from an operation to exactly what it touches: check out a copy and the loan is recorded against one Item, leaving its sibling Items available and the Manifestation untouched; withdraw a damaged copy and one Item record retires while the Manifestation and all sibling Items remain intact; add a conservation note and it binds to the affected copy alone. The predicted blast radius is precisely one Item — never the edition — and that boundedness is the whole point of promoting the Item. Conversely, an edit at the Manifestation (a corrected publisher field) propagates to every copy by inheritance, reaching all Items at once. Reasoning about any change therefore starts by typing it to a tier, because the tier fixes its reach: Item-level edits are local to one copy, Manifestation-level edits are shared across all.

Boundary-drawing — when a new Item, and the cardinality that constrains it. The model forces a branch on what counts as a distinct instantiation: a different physical or digital embodiment in its own location, with its own identifier and custody, is a new Item; a different print run or format is instead a new Manifestation; a different translation or significant revision is a new Expression; the same intellectual creation throughout is one Work. Deciding which boundary a difference crosses is the recurring adjudication the four-tier ladder makes crisp — and it is constrained by a fixed cardinality: each Item embodies exactly one Manifestation, while one Manifestation may bind many Items. Reasoning from that asymmetry, an availability claim must be settled at the Item level (an interlibrary-loan "do you hold an available copy?" needs a specific uncirculated copy, not merely the edition), because holding the Manifestation guarantees the abstraction is held but says nothing about whether any individual instantiation is free — the availability fact lives only at the tier where individual copies are distinguished.

Knowledge Transfer¶

Within bibliographic and cultural-heritage modeling the concept transfers as mechanism, with its property-placement tests and diagnostics intact. The Item discipline — promote the most concrete instantiation to a first-class entity with its own identifier, partition properties across the four-tier Work–Expression–Manifestation–Item ladder by the operational test "invariant across copies or copy-specific?", and bound copy-level operations to a single Item — carries across the FRBR/IFLA-LRM/BIBFRAME-aligned family without translation: library cataloging (loans, withdrawals, conservation notes against a specific copy), museum object records (the accession-numbered artifact versus its type/workshop), archival finding aids (a specific folder in a specific box versus its series), and digital asset management (a specific file at a specific path with a specific checksum versus the intellectual content). A curator moving from library to museum to digital repository reads the same diagnostics (uncountable copies means the Item was never promoted; hundreds of near-duplicate records means instance variation was modeled by cloning rather than by one Manifestation plus per-copy Items; an edition-wide loan or withdrawal means an operation was attached at the wrong tier) and the same placement rule in each. These are co-instances of one instantiation-hierarchy discipline, not analogies, because each gives every property exactly one rung and settles availability at the tier where individual copies are distinguished.

Beyond cultural-heritage modeling the honest reading is case (B): underneath the FRBR Item lies the substrate-independent type/token distinction (Peirce 1906) — equivalently the class/instance distinction in computing — and that is the more-general pattern that genuinely recurs across domains, as true co-instances rather than resemblances. Any system that must talk simultaneously about a shared abstraction and its individual instantiations recovers some version of it: class and instance in object-oriented programming; a currency denomination and a specific serial-numbered banknote; a drug product (NDC code) and a specific lot-and-serial bottle; a software version and a specific build artifact with its SHA; a biological species and an individual organism; a melodic theme and its performances; a sentence-type and its written tokens. These are genuine instances of type/token, not metaphors for a library Item, and the cross-domain lesson — model the abstraction and the instance as distinct first-class things with their own identity conditions — should be carried by the type/token (or class/instance) parent. What does not travel out is the home-bound cargo: the four-tier Work–Expression–Manifestation–Item ladder specifically (most domains need only a two-level type/token split, not FRBR's intermediate Expression and Manifestation rungs), the bibliographic property partition (publisher/ISBN at Manifestation, translator/language at Expression, call number/barcode/circulation at Item), the FRBR boundary tests (new print run → new Manifestation; new translation → new Expression), and the cataloging operations (loan, withdrawal, conservation event). So when the cross-domain lesson is wanted — separate the shared abstraction from the individual instance — carry the type/token parent, not the bibliographic "copy Item," whose four-tier ladder and cataloging machinery is a domain application of that distinction at one specific level. Calling a class instance or a serial-numbered banknote a "copy Item" in the FRBR sense borrows the instantiation shape while importing a four-tier ladder that does not fit; that is the analogy boundary, and the cross-substrate lesson belongs to the type/token parent (see Structural Core vs. Domain Accent).

Examples¶

Canonical¶

Take a library's holdings of Pride and Prejudice and walk the four-tier ladder. The Work is Austen's novel as an abstract creation; the Expression is its original English text; the Manifestation is one specific publication — say the Penguin Classics 2003 paperback, ISBN 978-0141439518 — carrying publisher, edition statement, ISBN, and pagination that are identical for every copy. The library owns fifty physical copies of that edition. Each copy is a distinct Item: its own barcode, its own call number and shelf location, its own acquisition date, condition notes, and current circulation status. Crucially, there is one Manifestation record and fifty Item records, not fifty duplicate full records — the shared publication data lives once and is inherited. When a patron borrows a copy, the loan is recorded against that one Item (barcode 39001...), leaving the forty-nine siblings available; when a copy is water-damaged and withdrawn, that single Item record retires while the Manifestation and the other copies stand untouched.

Mapped back: Each physical copy is the singular instantiation; the Penguin 2003 paperback is the embodied Manifestation; giving each copy its own record is the Item-first-class promotion. Barcode, call number, and status are the instance-specific metadata, sorted there by the property-placement test (ISBN is invariant across copies → Manifestation; a condition note can vary → Item). The one-Manifestation-to-fifty-Items relation is the cardinality constraint, and the loan touching exactly one copy is the bounded-operation guarantee.

Applied / In Practice¶

Integrated library systems (Ex Libris Alma, the open-source Koha, and union catalogs like OCLC WorldCat) implement precisely this split to run daily circulation and resource sharing. A bibliographic/manifestation record holds the edition-level data, attached "holdings" and "item" records carry per-copy call numbers, barcodes, and loan status, and the circulation module transacts against items. This is what lets interlibrary-loan systems answer "can I borrow this?" correctly: they must find a specific, currently-uncirculated copy, not merely confirm that some library owns the edition. Withdrawing a lost copy, placing a hold, or flagging a book for repair all resolve to one item without disturbing the shared record or sibling copies across the consortium.

Mapped back: The ILS bibliographic record is the embodied Manifestation and the per-copy holdings/item records are the promoted singular instantiations carrying instance-specific metadata. ILL availability resolving to a specific free copy is the concept's rule that availability is settled only at the Item tier — holding the edition guarantees the abstraction, not a free copy. Circulation, holds, and withdrawals each hitting one item exemplify the bounded-operation guarantee at production scale.

Structural Tensions¶

T1: First-class promotion versus record proliferation (granularity that also costs). Promoting each copy to a first-class entity is what buys precise copy-level operations — the loan against one barcode, the withdrawal of one damaged copy. The entry rightly notes the bookkeeping scales with variation, not copy count, since shared data lives once at the Manifestation. But every copy still needs its own Item record, identifier, and lifecycle, and for holdings of truly interchangeable mass-market copies that per-copy overhead buys little: no one cares which identical paperback is on loan, only that one is. The tension is that item-level granularity is indispensable where copies genuinely differ (annotated, rare, damaged) and near-pure overhead where they do not, yet the model promotes the Item uniformly. The precision and the cost are the same promotion applied regardless of whether copy-level distinctions matter. Diagnostic: Do the copies here carry genuinely copy-specific facts or operations that need individual tracking, or are they interchangeable — making per-Item records overhead the model imposes without payoff?

T2: The decidable placement test versus level-migrating data (facts that will not sit still). The property-placement test — invariant across copies goes to Manifestation, copy-specific goes to Item — is presented as a decidable typed assignment. But real data resists the binary: a printing defect present in an entire print run belongs at the Manifestation yet looks like a copy condition; an annotation shared by a sub-batch of copies fits neither tier cleanly; a fact invariant today (all copies pristine) becomes copy-specific tomorrow (one gets water-damaged). The test assumes each datum has a stable home, while the same property can belong to different tiers depending on scope and time. The tension is that the model's core discipline depends on a clean invariant-versus-variable dichotomy that a substantial minority of bibliographic facts straddle, so "where does this go?" remains a judgment call exactly at the borderline cases that matter. Diagnostic: Is the datum stably invariant across all copies or stably copy-specific — or does it hold for a sub-population or only until a copy-level event, straddling the tier boundary the test presumes clean?

T3: Four-tier precision versus usability and interoperability (WEMI's contested middle). The Work–Expression–Manifestation–Item ladder captures real distinctions a two-level model would blur, and that expressive precision is the concept's contribution. But WEMI is famously hard to apply consistently: the Expression/Manifestation boundary in particular is contested among catalogers, born-digital resources, serials, and aggregates fit the four rungs awkwardly, and IFLA-LRM itself restructured the model partly in response. The tension is that added expressive precision competes directly with usability and cross-system interoperability — a ladder fine-grained enough to place every distinction is also fine-grained enough that different catalogers place the same resource differently, undermining the shared-vocabulary payoff the model exists to provide. More rungs mean more discriminating power and more disagreement about which rung a fact occupies. Diagnostic: Does the resource map cleanly onto all four rungs with agreement among catalogers, or does the Expression/Manifestation split (or an aggregate/serial structure) make the extra precision a source of inconsistency rather than clarity?

T4: Physical-copy intuition versus digital instantiation (a model grounded in scarcity asked to serve abundance). The Item concept is built on the physical copy — one object, one location, one custody chain, availability bounded by scarcity — and that grounding is what makes "record the loan against one Item" and "availability settles at the Item tier" intuitive. Digital assets strain it: a file can be copied infinitely at zero cost, so "how many copies do we hold" and "is a copy available" lose their scarcity meaning, and whether each byte-identical download is a distinct Item or the same one is genuinely unclear. The tension is that the concept is asked to serve digital asset management with a physical-object intuition whose defining feature — a singular instance whose use excludes others — is exactly what digital instantiation dissolves. The checksum-identified file restores identity but not scarcity, so half the Item machinery (custody, circulation, availability) applies and half does not. Diagnostic: Does the copy in question have physical-style scarcity (one use excludes another, custody matters), or is it a freely-reproducible digital instantiation where the Item's availability and circulation machinery no longer has a referent?

T5: Availability at the Item tier versus the abstraction users actually query (right tier, wrong question). The model insists availability is settled only at the Item level — holding the edition says nothing about whether a copy is free — and this is correct for physical circulation. But patrons increasingly ask questions pitched at higher tiers: "can I read this book in any form?" is a Work- or Expression-level query, and the item-first discipline that correctly answers "is this specific copy available?" can obscure the abstraction-level answer the user wanted. The tension is that the tier which is precisely right for one question (physical availability) is the wrong tier for another (does the library provide access to this content at all), and a system optimized to resolve availability at the Item level must climb back up the ladder to serve the increasingly common work-level access question. Precision at the bottom rung competes with answering questions posed at the top. Diagnostic: Is the user asking whether a specific copy is free (Item tier, where availability lives) or whether the content is accessible in any form (Work/Expression tier) — and does the item-first model surface the level the question is actually pitched at?

T6: Autonomy versus reduction (a bibliographic tier or the type/token distinction). Within cultural-heritage modeling the FRBR Item transfers as mechanism — the promotion, the property partition, the placement test, the bounded-operation guarantee carry across library, museum, archive, and digital-asset cataloging as co-instances. But its cross-domain content reduces cleanly: underneath the Item lies the substrate-independent type/token (Peirce) or class/instance distinction, and that is what recurs as genuine co-instances — class and object, currency denomination and serial-numbered banknote, drug product and lot-numbered bottle, software version and build artifact, species and organism. The home-bound cargo is the four-tier ladder specifically (most domains need only a two-level type/token split, not the intermediate Expression and Manifestation rungs), the bibliographic property partition, the FRBR boundary tests, and the cataloging operations. The tension is between a richly specified bibliographic tier and the recognition that its portable lesson — model the abstraction and the instance as distinct first-class things — belongs to type/token. Diagnostic: Resolve toward the type/token (class/instance) parent when the lesson is separating a shared abstraction from its individual instances in any domain; toward the named FRBR Item when the four-tier WEMI ladder and its cataloging operations are actually in play.

Structural–Framed Character¶

The FRBR Item sits at the framed-leaning position — on the framed side because it is a designed modeling construct constituted by a cataloging practice and standard, though held off the framed pole because it renders no verdict; it is engineered infrastructure for describing objects, not a judgment about them. On evaluative weight it points structural: promoting a copy to a first-class entity and partitioning properties across tiers convicts nothing — it is a neutral discipline, and a modeling error surfaces as a tier mismatch, not a moral fault. But on human-practice-bound it is strongly framed: the Item is constituted by a bibliographic practice and dissolves without it. Its content is a set of cataloging operations — loan, withdrawal, conservation note, availability query — and placement decisions made by practitioners against a designed record structure; strip the cataloging practice and there is no "Item," only an object. On institutional origin it is emphatically framed: the Item is an artifact of a specific standards tradition (FRBR, IFLA-LRM, BIBFRAME), and its distinctive machinery — the four-tier Work–Expression–Manifestation–Item ladder, the property-placement test, the cardinality constraint, the FRBR boundary tests — is knowledge-organization infrastructure erected by that tradition, not a distinction any observer-free system exhibits. On vocab-travels it is framed (Manifestation, Expression, the WEMI rungs, the cataloging operations are LIS-specific), and on import-vs-recognize the entry is careful: the cross-domain cousins (class and object, currency denomination and serial-numbered banknote, drug product and lot-numbered bottle, species and organism) are genuine co-instances of the parent type/token distinction, not imports of the FRBR Item — so what recognizes across domains is type/token, while "copy Item," with its four-tier ladder, does not travel.

The one portable structural skeleton is the substrate-independent type/token distinction (Peirce), equivalently class/instance in computing: model a shared abstraction and its individual instantiations as distinct first-class things with their own identity conditions. That skeleton genuinely recurs across programming, currency, pharmacology, versioning, and biology as true co-instances, which is what tempts a structural reading; but it is precisely what the FRBR Item instantiates from the type/token parent at one specific level, not what makes "copy Item" itself portable. The cross-domain reach belongs to type/token — separate the abstraction from the instance — while the domain-accented apparatus that makes this the FRBR Item (the intermediate Expression and Manifestation rungs most domains do not need, the bibliographic property partition, the FRBR boundary tests, the cataloging operations) stays home; importing the four-tier ladder onto a class instance or a banknote, the entry notes, is the analogy boundary. Its character: a neutral but thoroughly standard-and-practice-constituted bibliographic construct, structural only in the type/token skeleton it applies at one rung, framed in all the four-tier cataloging machinery that gives it its character.

Structural Core vs. Domain Accent¶

This section decides why the FRBR Item is a domain-specific abstraction and not a prime, and it carries the case for its domain-specificity — so it is worth being exact about what could lift and what stays home.

What is skeletal (could lift toward a cross-domain prime). Strip the cataloging and a thin relational structure survives: model a shared abstraction and its individual instantiations as distinct first-class things, each with its own identity conditions, so instance-specific facts and operations attach to one instance rather than the shared type. The portable pieces are abstract — a type, its tokens, a partition of properties between them, and operations whose scope is one token. That skeleton is exactly the substrate-independent type/token distinction (Peirce 1906), equivalently class/instance in computing, and it genuinely recurs as true co-instances — not resemblances — in class and object, a currency denomination and a serial-numbered banknote, a drug product and a lot-numbered bottle, a software version and a build artifact with its SHA, a species and an organism. That is the core the FRBR Item shares, not what makes it the FRBR Item.

What is domain-bound. Almost everything that makes the entry the FRBR Item in particular is bibliographic-modeling furniture, and none of it survives extraction. Its distinctive apparatus is the four-tier Work–Expression–Manifestation–Item ladder — most domains need only a two-level type/token split, not FRBR's intermediate Expression and Manifestation rungs; the bibliographic property partition (publisher/ISBN at Manifestation, translator/language at Expression, call number/barcode/circulation at Item); the FRBR boundary tests (new print run → new Manifestation, new translation → new Expression); and the cataloging operations (loan, withdrawal, conservation event, availability query) settled at the Item tier. The decisive test the entry itself supplies: calling a class instance or a serial-numbered banknote a "copy Item" in the FRBR sense borrows the instantiation shape while importing a four-tier ladder that does not fit — that is precisely the machinery that does not survive extraction. Remove the cataloging standard and its record structure and there is no "Item," only an object.

Why this does not clear the prime bar. A prime is a relational structure whose vocabulary travels and whose cross-domain transfer is recognition of the same mechanism, not analogy. The FRBR Item's transfer is bimodal. Within bibliographic and cultural-heritage modeling it travels intact as mechanism — the first-class promotion, the property-placement test, the cardinality constraint, and the bounded-operation guarantee carry across library cataloging, museum object records, archival finding aids, and digital asset management, because each gives every property one rung and settles availability where individual copies are distinguished. Beyond cultural-heritage modeling the named construct does not travel: the cross-domain cousins are genuine co-instances of the parent type/token distinction, not imports of the four-tier Item, and forcing the WEMI ladder onto them is the analogy boundary. When the cross-domain lesson — separate the shared abstraction from the individual instance, each first-class — is needed, it is already carried, in more general form, by the type/token (class/instance) parent, of which the FRBR Item is a domain application at one specific rung. The cross-domain reach belongs to that parent; "copy Item," as named, carries the four-tier ladder and cataloging operations that keep it a bibliographic construct rather than a free-floating prime.

Relationships to Other Abstractions¶

Current abstraction Item (Copy) Domain-specific

Parents (2) — more general patterns this builds on

Item (Copy) presupposes Abstract Work Prime

Calling one copy an Item at the bottom of WEMI presupposes a content identity that persists above its concrete carriers and fixes which variations remain realizations of the same Work.
Item (Copy) is a decomposition of Type–Token Distinction Prime

Removing the child’s frame leaves the reusable structure named by Type Token Distinction.

Hierarchy paths (3) — routes to 1 parentless root

Item (Copy) → Abstract Work → Type–Token Distinction → Abstraction

Show alternative paths (2)

Not to Be Confused With¶

The Manifestation. The edition — the shared publication (publisher, edition statement, ISBN, format, pagination) common to all copies, the rung directly above the Item. An Item is this book; the Manifestation is this kind of book. Attaching a copy-specific fact (a condition note) to the Manifestation wrongly claims it of every copy. Tell: Would this datum be identical for every copy of the edition (Manifestation), or can it vary copy to copy (Item)?
The Work. The abstract intellectual or artistic creation at the top of the ladder — Austen's Pride and Prejudice as such, independent of any text or copy. The Item is the most concrete rung, a physical or digital token; reading it as the Work collapses the bottom of the instantiation hierarchy into its top. Tell: Is it the abstract creation regardless of any embodiment (Work), or one singular copy with its own barcode and location (Item)?
The Expression. A distinct realization of the Work — a particular translation or significant revision — the second rung. It is still abstract (a text, not an object): a new translation makes a new Expression, whereas a new physical copy makes a new Item. Tell: Is the difference a change in the intellectual realization like translation or revision (Expression), or an additional physical/digital instantiation of the same publication (Item)?
The bibliographic record. The catalogue record that describes holdings — easily conflated with the Item it represents. The Item is the modelled entity (the copy as a first-class thing with identifier, location, history); the record is the descriptive surrogate. Cloning a whole record per copy instead of one shared Manifestation plus per-copy Items is the modeling anti-pattern the property-placement test exists to prevent. Tell: Are you naming the copy-as-entity that loans and withdrawals attach to (Item), or the descriptive record structure that documents it (bibliographic record)?
Type/token (class/instance) — the parent. The substrate-independent distinction (Peirce; class/instance in computing) between a shared abstraction and its individual instantiations. The FRBR Item is this distinction applied at one specific rung of a four-tier ladder; class-and-object, a currency denomination and a serial-numbered banknote, a drug product and a lot-numbered bottle, a species and an organism are co-instances of the parent needing only a two-level split. Forcing FRBR's Expression/Manifestation rungs onto them is the analogy boundary. Tell: Is the four-tier WEMI ladder with its cataloging operations actually in play (FRBR Item), or just the general split between an abstraction and its instances (type/token, which carries the cross-domain lesson)?

Neighborhood in Abstraction Space¶

Item (Copy) sits in a sparse region of the domain-specific corpus (63^rd percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Authority Control & Identity (11 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12