© 2013 Attila Dinnyés
© 2023 Attila Dinnyés
Attila Dinnyes Attila Dinnyes
Attila Dinnyes

The epistemic infrastructure of meaning-making
Designcommunication and the constitution of cultural value

Demonstrated by Systematic Literature Reviews of
papers on Co-Creation and Arts Management
(between 2000 and 2024)

oszapo_2
Simple SLR
A keyword-based, PRISMA-screened descriptive review. It groups the 116 papers by co-creation vocabulary in titles, abstracts and keywords: a strict co-creation term, a broader related family (co-production, participation, collaboration, engagement), or none. It sees keyword presence, not full-text depth or whether genuine shared value creation occurs. The keyword layer itself is a long tail: a few frequent terms and a vast number of one-off concepts.
Defining papers: Paul & Criado (2020) · Gupta (2023) · Saragih (2019)
Integrative SLR
Each of the 116 papers is read in full text and coded on six variables (domain, locus, emphasis, depth, category, match), rather than by keyword presence. The integrative basis of the coding is provided by Orthel’s (2022) depth gradient, running from surface mention to foundational inquiry, while the interpretive stance is supplied by the coding (pp. 110–121) and self-reflexive (pp. 263, 276) sections of Horvath and Mitev’s qualitative handbook, by Lincoln and Guba’s (1985) naturalistic inquiry, and by Charmaz’s (2014) constructivist grounded theory. From Lincoln and Guba comes the principle of trustworthiness, warranted intersubjectively rather than through inter-coder reliability (1985, p. 314); from Charmaz, the premise that no method is neutral and that coding is situated interpretation (2014, p. 1076), which requires us to proceed in reflexive awareness of our own professional standpoint. The four category lenses draw on Kaszynska, Klamer, Woronkowicz and Noonan. Designcommunication enters as application, as the epistemic praxis through which the researcher’s interpretation is carried out. Research design and object of study stand in a reflexive relation: designcommunication describes the motivated, meaning-making and value-making collaborative process in which cultural value comes into being, and the same disposition guides the researcher’s attention. This designcommunication correspondence, however, does not presuppose the value-constitutive reading; in the codebook it figures as a test that a study can also fail (registered by the MATCH and CAT variables), and for most of the 116 papers it does fail (MATCH = N). By its designcommunication nature, this stance behaves as the dynamic “research brand” we introduce, modelled on Fekete’s (2024, pp. 95, 98) participative dynamic visual identity: the documented codebook itself remains constant, while its reception varies with the participant’s researcher, reader, or critical role.

Defining papers: Orthel (2022) · Throsby (2010) · Klamer (2012) · Kaszynska (2025) · Fekete (2024) · Galla, Horvath & Cosovan (2024) · Horvath & Mitev (2023) · Lincoln & Guba (1985) · Charmaz (2014) · Woronkowicz, Noonan & Skaggs (2026)

SLR_keyword_match_venn
Framing document [EN]

1. Introduction

This is a two-phase hybrid systematic literature review.

  • Phase 1 is a conventional, PRISMA-conformant structured review (database search, screening, descriptive synthesis). It satisfies the classic genre rulebook (Paul & Criado, 2020): three databases, 116 studies in the accepted 40–500 range, more than ten years of coverage, structured tables, an audit trail.
  • Phase 2 is an integrative extension built on Orthel’s (2022) depth-gradient model. Instead of inventorying what the field writes about, it reads each study in full text and asks what it actually claims, and what it never asks.

Phase 1 is the photograph, Phase 2 the painting: the qualitative researcher constructs a painting rather than a photograph (Charmaz, 2003, as cited in Horváth & Mitev, 2015, p. 121). The classic review tells you what the field writes about; the integrative reading tells you what the studies actually argue, and the questions they never ask.

In a keyword- or abstract-based review every one of the 116 studies is a “co-creation paper.” Reading them in full, however, shows that only a minority genuinely analyse shared content creation. That gap is not noise; it is the central finding, and it cannot be produced any other way.

2. The three layers of the integrative review

The review is structured as three layers of reading, and the distance between the layers constitutes the result.

Layer What it captures The question it answers
1. Metadata Title, keywords, topics, database, methodology, scope (bibliographic surface). What does the study say it is?
2. Hand-read The researcher’s first interpretive pass (an early tension list and a hand-written conceptual map per study). What did you read it as on first contact?
3. Full-text coding The six-variable scheme below, coded from the complete argument. What does the study actually do?

When a study labels itself “co-creation” (Layer 1) but the full text shows otherwise (Layer 3), that is conceptual inflation; when a study genuinely deals with co-creation (Layer 3) but carries no co-creation keyword (Layer 1), that is invisibility. The layers are not redundant measurements of the same thing; they are complementary, and their disagreement is the data.

3. Variables

Six coding variables replace the older keyword/theme columns. They fall into two groups: descriptive (where, with whom, which value type) and interpretive (how deeply, which lens, is it real). The main matrix is DOMAIN × CAT; everything else refines it. A seventh field, RELATIONS, records the conceptual tensions in each study (see Section 5).

Variable Type What it asks Values Where it comes from (anchor)
DOMAIN descriptive, exclusive (7) In which cultural sector does the phenomenon sit? Museum · Heritage & archaeology · Music & performing arts · Visual art & art market · Festival & event · Cultural policy · Urban & spatial Pragmatic; emerged from the corpus, loosely fitted to Throsby’s (2001) cultural-sector taxonomy. Descriptive, not theory-driven.
LOCUS descriptive, primary + optional secondary (5) With which (non-primary-author) party does co-creation happen? Community-led · Visitor/consumer · Inter-professional · Artist–stakeholder · Conceptual Becker’s Art Worlds (1974, 2008): “no art is made alone”; LOCUS records which actor-relation is at stake (Becker cited in Woronkowicz et al., 2026, p. 615).
EMPHASIS descriptive, exclusive (3) Which type of cultural value is foregrounded? Cognitive · Social · Institutional (institutional ≠ financial) Throsby’s (2001) value dimensions and Klamer’s (2012) value spheres, operationalised into three codeable registers.
DEPTH interpretive, exclusive (4) How deeply is co-creation itself examined? L1 Surface (label only) · L2 Context (effect measured, mechanism a black box) · L3 Mechanism (process analysed, value taken as given) · L4 Foundational (the nature of value/co-creation is the question) Orthel (2022, IJHS 28(1), pp. 44–58), depth-gradient method, adapted to four levels. The system’s central lens.
CAT interpretive, primary + optional secondary (4) Through which theoretical lens is the contribution best grasped? A Value & meaning · B Process & actors · C Power & legitimacy · D Mediation & transfer Four anchored lenses (see below). The column of the main matrix.
MATCH interpretive, binary (Y/N) Is this genuine co-creation, i.e. shared control or co-authorship over what is created? Y / N Becker (collaboration ≠ co-creation, in Woronkowicz et al., 2026, p. 615); Luonila & Jyrämä (2019), “intertwined yet distinct”; Brandsen & Honingh (2016). The most delicate decision.

The four CAT lenses (with anchors). These are the spine of the analysis:

  • A — Value & meaning. Kaszynska (2025, Cultural Trends 34(4), p. 470): cultural value as normative, hermeneutic, intersubjectively warranted. Authenticity, meaning-making, value constitution.
  • B — Process & actors. Klamer (2012): co-creation among makers, lovers and users; Becker, Art Worlds: patterned cooperation. Process, roles, networks, the mechanics of collaboration.
  • C — Power & legitimacy. Woronkowicz (2021, JCE 45, p. 520, “market for symbolic meaning”); Noonan (2003, JCE 27, pp. 162, 172, CVM/WTP, cultural goods resist commodification); Bourdieu (1993, 1996), field struggles. Who controls, legitimates, defines; democratisation, decolonisation, instrumentalism.
  • D — Mediation & transfer. Galla, Horváth & Cosovan (2024, Creativity Studies 17(1), pp. 282–294) and Cosovan (2009): designcommunication as a mediating, epistemic structure. This is the white spot, and the dissertation’s contribution space.

On MATCH = N (the inflation indicators). A study is coded N when the author themselves says “co-production”; when “participation” means presence, visiting or consultation; when it is mere social-media engagement; when it is value-in-use / experience value in the service-dominant sense (Vargo & Lusch; Grönroos; Prahalad & Ramaswamy), where the consumer creates their own experience but not the content; lab stimulus; or co-creation as a background term in a governance analysis.

4. The four DEPTH levels (L1-L4)

What it is, and what it builds on. DEPTH is the system’s central lens: it asks not what a study is about, but how deeply it examines co-creation itself. It is built on Orthel’s (2022, International Journal of Heritage Studies 28(1), p. 44-58) depth-gradient method, which stratifies inquiry into three tiers (direct/causal, rigorous secondary, indirect/implied) on Groat and Wang’s (2013) Thinking-Finding-Sorting-Writing cycle (Orthel (2022), p. 45). This gradient is adapted into four levels, which also carry the thesis axis: the difference between L3 (mechanism) and L4 (value constitution).

The four levels.

  • L1 — Surface. “Co-creation” is only a label, or participation means presence/visiting; the concept itself is not examined. Decision: a study lands here when the text at most mentions co-creation (e.g. pure economic history, or a policy description where co-creation is not even a topic). Of the 116 studies, 11 fall here, all MATCH = N.
  • L2 — Context. The co-creative frame is present and the study measures or describes the output, but the mechanism is a black box: it does not unpack how value is created. Decision: it lands here when it measures an effect or experience but leaves the “why/how” unopened. Typical case: most empirical governance and management studies; this is the most populous level (54 studies) and overwhelmingly MATCH = N (only 15% Y).
  • L3 — Mechanism. The study analyses the process (through which actors, steps, tools), but takes value as given. Decision: it lands here when it unpacks the process (steps, roles, tools, gradations) without making value itself a question. Typical case: process-analysing case studies; this is where MATCH tips into the clear majority (76% Y).
  • L4 — Foundational. Co-creation, or the nature of value itself, becomes the object of inquiry (value constitution, ontology, a critical or theoretical rethinking of the concept). Decision: it lands here when the study makes the concept of value or co-creation a question. Typical case: the theoretical, conceptual pieces; a seven-member cluster (#4, #38 Graham, #47 Holt, #67 Luonila, #87 Nakajima, #93 Preece, #98 Sachs Olsen), all MATCH = Y. Anchors: Kaszynska (2025), p. 470 (value as intersubjectively warranted); Sachs Olsen (2022), p. 319 (the spokesperson “brings the voice into being”, i.e. mediation constitutes the voice); for the participatory-art pieces, Bishop, Bourriaud (2002) and Kester (2004).

The classification decision. Levels are read from a study’s argumentative structure, not its topic: what it takes as given, and what it problematises. The heuristic: (a) merely mentions co-creation → L1; (b) measures the effect but leaves the mechanism unopened → L2; (c) unpacks the process → L3; (d) makes the concept itself a question → L4. By definition this needs the full text: studies pre-coded from the abstract or an earlier database can shift in classification on the full reading (e.g. for Amer the CSV-based L2 became L3 in full text).

Borderline cases. There are two kinds, and both are findings in their own right:

  1. Under-coded depth (the L3/L4 boundary). In the REVISIT round we re-read twelve borderline cases at high effort; in three of them DEPTH changed from L3 to L4: #38 Graham (co-production is not only governmental but also an epistemological stake), #87 Nakajima (the nature of the producer/consumer boundary is the stake), #93 Preece (the “composition of value”, i.e. value constitution itself, is the stake). The shallow reading thus systematically under-coded the most important (foundational) layer: half of the seven-member L4 cluster only reached its place on re-reading. Stratifying by depth itself requires a deep reading.
  2. Deep but not co-creation (DEPTH and MATCH are independent). Ten studies sit at L3 or L4 yet are MATCH = N (#19 Chaney, #21 Choi and Burnes, #23 Colomer, #25 Davies, #61 Kristal, #65 Li and Coll-Serrano, #72 Mason, #94 Pulh and Mencarelli, #101 Stadler and Fullagar, #108 Valtysson). Serious analytical machinery is aimed at something else (value-in-use, engagement, governance, intra-organisational collaboration) under the co-creation label. These show that DEPTH and MATCH are two independent axes: depth does not guarantee genuine co-creation. They are also the strongest evidence of conceptual inflation, because here the slippage is systematic, not careless.

Why it matters. The DEPTH × MATCH cross-tab yields a monotone gradient (0%, 15%, 76%, 100% MATCH = Y from L1 to L4, see Section 7): the deeper a study examines co-creation, the more likely it is genuine. Neither axis of this gradient is codeable from an abstract, so it cannot be produced in a classic, keyword-based review.

5. RELATIONS and its five dimensions

Each study is characterised by between three and five X vs Y conceptual tension pairs (e.g. competition vs collaboration, top-down vs bottom-up, intrinsic vs instrumental), which we extracted from the study’s own reading rather than fitting from an external template. Across the corpus this yields 493 pairs.

Through the five dimensions, every pair sorts into one of five families:

Dimension What it captures Typical pairs What it shows
POWER Who controls, decides, legitimates top-down vs bottom-up; authority vs legitimacy; institutional control vs community agency The dominant dimension (61% of studies); N-associated. The field is primarily a legitimacy discourse, often without real co-creation.
AUTHORSHIP Who authors, whose work it is artist vs audience; producer vs consumer (prosumer); single vs collective authorship; co-creation vs co-production The only Y-majority dimension (39:27). Where authorship is the question, co-creation is usually genuine.
SCALAR Local vs global, micro vs macro, physical vs digital heritage branding vs community identity; local scene vs global market Echoes the 76% local scope; the translocal scaling of value is under-theorised.
AXIOLOGICAL What value is, who defines it cultural value vs economic value; intrinsic vs instrumental; authenticity vs commodification The rarest dimension (28%). The value question is the field’s own blind spot, precisely where Kaszynska’s lens applies.
TEMPORAL-ONTO Process vs product, tradition vs innovation preservation vs living heritage; process vs artefact The backbone of heritage studies; “process vs artefact” is the linguistic marker of the L3/L4 boundary.

The five dimensions rest on the following methodological commitments:

  • Axial coding (Strauss & Corbin; Horváth & Mitev, 2015, pp. 108–109): coding is about relations between categories, not just categories.
  • Paralogic / transgressive validity (Lather; Horváth & Mitev, 2015, pp. 66–67): a paradox is data, not noise. A tension pair is kept as a tension, not resolved away.
  • Eisenstein montage (Horváth & Mitev, 2015, pp. 39–40): the collision of opposites produces a new quality (1+1=3). RELATIONS is generative, not an inventory; in the manuscript the dimensions are set against each other, not written as separate chapters.
  • Becker’s “art words” (via Cluley, 2012, p. 212): opposed vocabularies anchor action; the way a study names its tensions is itself a positioning move (e.g. the “art-versus-money” script).

This is also where the integrative review departs from the classic genre: instead of a “similar or contradictory findings” table (Paul & Criado, 2020, p. 4), the contradictions are the field’s problem map.

6. The basis of the coding decisions (methodological stance)

The choices inside every variable are governed by a small set of explicit methodological commitments.

  • Depth over inventory. Orthel (2022): stratify by depth of inquiry (surface → mechanism → foundational), not by theme. This is what makes DEPTH the organising axis rather than topic.
  • No method is neutral; coding is situated interpretation. Charmaz (2014, Qualitative Inquiry 20(9), p. 1076), constructivist grounded theory: the codes are constructed in interaction with the data, not discovered.
  • Trustworthiness, not inter-coder reliability. Lincoln & Guba (1985, p. 314): a single-coder design is warranted intersubjectively through an explicit codebook, auditable databases and a documented REVISIT round, rather than through inter-coder statistics.
  • Interpretive handbook. Horváth & Mitev (2015): the coding (pp. 110–121) and self-reflexive (pp. 263, 276) sections supply the qualitative craft (crystallisation, introspection, the open-work reading via Eco, p. 379).
  • Four value lenses, named. Kaszynska, Klamer, Woronkowicz and Noonan (above) are the theoretical anchors for CAT; Throsby/Klamer for EMPHASIS; Becker / Luonila & Jyrämä / Brandsen & Honingh for MATCH.
  • Designcommunication as application. CAT=D (Galla, Horváth & Cosovan, 2024; Cosovan, 2009) enters as the epistemic praxis through which interpretation is carried out: research design and object of study stand in a reflexive relation. Modelled on Fekete’s (2024, pp. 95, 98) participative dynamic visual identity, the documented codebook behaves as a constant “research brand” whose reception varies with the reader’s role.

Two guardrails that keep the stance honest:

  1. Reflexivity, not self-justification. The value-constitutive reading is not presupposed; in the codebook it is a test a study can fail, registered by MATCH and CAT. For most of the 116 studies it does fail (MATCH = N ≈ 60%). The method-object correspondence is reflexivity, not a circular proof.
  2. Constitutive, not predictive. The co-movement of RELATIONS and MATCH (especially in AUTHORSHIP) is read as the field’s self-description (the linguistic imprint of conceptual inflation), not as “the frame predicts the match.” Both codes come from one full-text reading by one coder, and the co-creation vs co-production tension restates the MATCH question itself; the two are not independent measurements, the corpus is a census not a sample, so no predictive-validity claim is made.

7. Results

Six findings, each with an auditable footprint in the data:

  1. Conceptual inflation. Of 116 studies, only 46 (40%) are genuine content co-creation (MATCH = Y); the other 59% are not what they call themselves. In a keyword review all 116 would count as evidence of co-creation.
  2. A monotone depth gradient. The deeper a study examines co-creation, the more likely it is genuine: MATCH-Y share rises 0% → 15% → 76% → 100% across L1 → L2 → L3 → L4. Neither axis is codeable from an abstract.
  3. The inflation cell. Ten studies sit at L3/L4 and MATCH = N: serious analytical machinery aimed at something other than co-creation (value-in-use, engagement, governance) under the co-creation label. This is the strongest evidence of systematic concept-slippage, supported by the corpus’s own words (“applied too liberally,” Walmsley, p. 117; “inherently emancipative” critique, Sachs Olsen, 2022, p. 316).
  4. The keyword two-layer argument. Among the 32 strict “co-creation”-labelled studies, 9 (28%) are empty (labelled but not genuine); among the 46 genuine ones, 23 (50%) carry no co-creation keyword (invisible to a keyword review), three of them at L4. The two readings are statistically distinct measurements (φ ≈ 0.41), hence complementary.
  5. The white spot, and a double blind spot. Only 9/116 (8%) make mediation/transfer the primary lens (CAT=D); even as a secondary lens it reaches only ~25/116. The four “mediator figures” the corpus does surface appear ad hoc, with no shared theoretical language. This converges with two further gaps: the axiological question (rarest RELATIONS dimension, 28%) and the cognitive/epistemic emphasis (only 18/116). The field debates who decides and who authors, but rarely asks what is created and how meaning transfers.
  6. RELATIONS power-dominance. The field’s leading problem space is power (61% of studies), not authorship; the “co-creation” literature is first of all a legitimacy discourse. Authorship is the only Y-majority dimension, and the axiological dimension is the field’s blind spot.

The contribution (the thesis). Designcommunication is not one form of co-creation but the epistemological infrastructure of the meaning-making process by which cultural value comes into being. It fills exactly the double rift above: it links the Level-4 ontological insight (value is constituted in co-creation, not transferred to it: Kaszynska, 2025, p. 470; Sachs Olsen, 2022, p. 319) to the Level-3 mechanism, operationalising when participation becomes value co-creation.

8. Anchor reference list (name, year, where)

  • Becker, H. S. (1974) Art as Collective Action; (2008) Art Worlds — collaboration ≠ co-creation (in Woronkowicz et al., 2026, p. 615).
  • Bourdieu, P. (1993) The Field of Cultural Production; (1996) The Rules of Art — field struggles. CAT=C.
  • Brandsen, T. & Honingh, M. (2016) — co-creation vs co-production (via Sachs Olsen, 2022, p. 316). MATCH.
  • Charmaz, K. (2003) the “painting, not photograph” metaphor (as cited in Horváth & Mitev, 2015, p. 121); (2014) Qualitative Inquiry 20(9), pp. 1074–1084 — “no method is neutral” (p. 1076); bilingual coding (p. 1078). Constructivist stance.
  • Cluley, R. (2012) — Becker’s “art words” (p. 212). RELATIONS.
  • Cosovan, A. (2009) DISCO: Co&Co Communication. CAT=D.
  • Galla, D. D., Horváth, D. & Cosovan, A. R. (2024) Creativity Studies 17(1), pp. 282–294 — designcommunication. CAT=D.
  • Horváth, D. & Mitev, A. (2015) Alternatív kvalitatív kutatási kézikönyv. Alinea, Budapest. Painting/photograph metaphor (p. 121), coding (pp. 110–121), axial coding (pp. 108–109), paralogic validity (pp. 66–67), montage (pp. 39–40), self-reflexivity (pp. 263, 276), Eco open work (p. 379).
  • Kaszynska, P. (2025) Cultural Trends 34(4), pp. 463–477 — cultural value normative/hermeneutic/intersubjectively warranted (p. 470). CAT=A.
  • Klamer, A. (2012) — co-creation: makers, lovers, users. CAT=B / EMPHASIS.
  • Lincoln, Y. S. & Guba, E. G. (1985) Naturalistic Inquiry — trustworthiness (p. 314).
  • Luonila, M. & Jyrämä, A. (2019) Arts and the Market 10(1), pp. 1–17 — co-creation vs co-production “intertwined yet distinct.” MATCH.
  • Noonan, D. (2003) Journal of Cultural Economics 27, pp. 159–176 — CVM/WTP (pp. 162, 172). CAT=C.
  • Orthel, B. (2022) International Journal of Heritage Studies 28(1), pp. 44–58 — depth-gradient method. DEPTH.
  • Paul, J. & Criado, A. R. (2020) International Business Review 29:101717 — the SLR genre rulebook. Phase 1.
  • Sachs Olsen, C. (2022) — spokesperson “brings the voice into being” (p. 319). CAT=D / L4.
  • Strauss, A. & Corbin, J. — axial coding. RELATIONS.
  • Throsby, D. (2001) Economics and Culture — cultural value dimensions. EMPHASIS / DOMAIN.
  • Vargo, S. & Lusch, R. (S-D logic); Grönroos; Prahalad & Ramaswamy — value-in-use. MATCH=N.
  • Woronkowicz, J. (2021) Journal of Cultural Economics 45, pp. 519–526 — “market for symbolic meaning” (p. 520). CAT=C.
  • Woronkowicz, J., Noonan, D. & Skaggs, R. (2026) Work and Occupations 53(3), pp. 599–634 — Becker (p. 615), Bourdieu/Beckert (p. 617). Benchmark.
Háttérdokumentum [HU]

1. Bevezető

Ez egy kétfázisú hibrid szisztematikus irodalmi áttekintés.

  • A Phase 1 konvencionális, PRISMA-konform strukturált áttekintés (adatbázis-keresés, szűrés, deskriptív szintézis). Teljesíti a műfaj klasszikus szabálykönyvét (Paul és Criado (2020)): három adatbázis, 116 tanulmány az elfogadott 40 és 500 közötti sávban, több mint tíz év lefedettség, strukturált táblák, audit trail.
  • A Phase 2 integratív kiterjesztés, amely Orthel (2022) mélységi gradiens modelljére épül. Ahelyett, hogy leltárba venné, miről ír a mező, minden tanulmányt teljes szövegből olvas, és azt kérdezi, mit állít valójában, és mit nem kérdez meg soha.

A Phase 1 a fotó, a Phase 2 az impresszionista festmény (Horváth és Mitev 2015) – a klasszikus áttekintés azt mondja meg, miről ír a mező, az integratív olvasat azt, hogy mit állítanak valójában a tanulmányok, és milyen kérdéseket nem tesznek fel soha.

Egy kulcsszó- vagy absztraktalapú áttekintésben mind a 116 tanulmány „co-creation”-tanulmánynak számít. Teljes szövegből olvasva kiderül azonban, hogy csak egy kisebbségük elemez valódi, tartalmi szintű közös alkotást. Ez a rés nem zaj, hanem a központi eredmény, és másként nem is állítható elő.

2. Az integratív áttekintés három rétege

Az áttekintés három olvasati rétegből épül fel, a rétegek közötti távolság alkotja magát az eredményt.

Réteg Mit ragad meg Milyen kérdésre felel
1. Metaadat Cím, kulcsszavak, témák, adatbázis, módszertan, scope (a bibliográfiai felszín). Minek mondja magát a tanulmány?
2. Kézi olvasat A kutató első értelmező olvasata (korai feszültséglista és kézzel írt fogalmi háló tanulmányonként). Minek olvastad elsőre?
3. Full-text kódolás A lenti hat változós séma, a teljes érvelésből kódolva. Mit csinál valójában a tanulmány?

Ha egy tanulmány „co-creation”-nek címkézi magát (1. réteg), de a teljes szöveg mást mutat (3. réteg), az fogalmi infláció; ha egy tanulmány valóban “co-creation”-nel foglalkozik (3. réteg), de nincs co-creation kulcsszava (1. réteg), az láthatatlanság. A rétegek nem ugyanannak a redundáns mérései, hanem egymást kiegészítők, és az eltérésük maga az adat.

3. Változók

Hat kódolási változó váltja fel a korábbi kulcsszó- és téma-oszlopokat. Két csoportba esnek: leíró (hol, kikkel, milyen értéktípus) és értelmező (milyen mélyen, melyik lencse, valódi-e). A fő mátrix a DOMAIN × CAT; minden más ezt árnyalja. Egy hetedik mező, a RELATIONS, az egyes tanulmányok fogalmi feszültségeit rögzíti (lásd 4. szakasz).

Változó Típus Mit kérdez Értékek Honnan ered (horgony)
DOMAIN leíró, kizáró (7) Melyik kulturális szektorban zajlik a jelenség? Museum · Heritage & archaeology · Music & performing arts · Visual art & art market · Festival & event · Cultural policy · Urban & spatial Pragmatikus; a korpuszból emelkedett ki, lazán illesztve Throsby (2001) kulturális-szektor taxonómiájához. Leíró, nem elmélet-vezérelt.
LOCUS leíró, elsődleges + opcionális másodlagos (5) Melyik (nem elsődleges szerző) féllel zajlik a co-creation? Community-led · Visitor/consumer · Inter-professional · Artist–stakeholder · Conceptual Becker Art Worlds (1974, 2008): „no art is made alone”; a LOCUS azt rögzíti, melyik aktor-reláció a tét (Becker idézve: Woronkowicz és mtsai. (2026), p. 615).
EMPHASIS leíró, kizáró (3) Milyen kulturális értéktípust állít előtérbe? Cognitive · Social · Institutional (az Institutional nem pénzügyi) Throsby (2001) értékdimenziói és Klamer (2012) értékszférái, három kódolható regiszterré operacionalizálva.
DEPTH értelmező, kizáró (4) Milyen mélyen vizsgálja magát a co-creationt? L1 Felszín (csak címke) · L2 Kontextus (a hatást méri, a mechanizmus black box) · L3 Mechanizmus (a folyamatot elemzi, az értéket adottnak veszi) · L4 Foundational (az érték, illetve a co-creation természete a kérdés) Orthel (2022), International Journal of Heritage Studies 28(1), p. 44-58, mélységi gradiens módszer, négy szintre adaptálva. A rendszer központi lencséje.
CAT értelmező, elsődleges + opcionális másodlagos (4) Melyik elméleti lencsén át ragadható meg a hozzájárulás? A Value & meaning · B Process & actors · C Power & legitimacy · D Mediation & transfer Négy horgonyzott lencse (lásd lent). A fő mátrix oszlopa.
MATCH értelmező, bináris (Y/N) Valódi co-creation-e, azaz megosztott kontroll vagy társszerzőség afölött, mi jön létre? Y / N Becker (a kollaboráció nem co-creation, in Woronkowicz és mtsai. (2026), p. 615); Luonila és Jyrämä (2019), „intertwined yet distinct”; Brandsen és Honingh (2016). A legkényesebb döntés.

A négy CAT-lencse (horgonyokkal). Ezek az elemzés gerince:

  • A — Value & meaning (érték és jelentés). Kaszynska (2025, Cultural Trends 34(4), p. 470): a kulturális érték normatív, hermeneutikai és interszubjektíven igazolt jellegű. Ide tartozik az autenticitás, a jelentésalkotás, az értékkonstitúció.
  • B — Process & actors (folyamat és aktorok). Klamer (2012): co-creation a makers, lovers és users között; Becker, Art Worlds: mintázott együttműködés. Folyamat, szerepek, hálózatok, a kollaboráció mechanikája.
  • C — Power & legitimacy (hatalom és legitimáció). Woronkowicz (2021, Journal of Cultural Economics 45, p. 520, „market for symbolic meaning”); Noonan (2003, Journal of Cultural Economics 27, p. 162, 172, CVM/WTP, a kulturális javak ellenállnak a kommodifikációnak); Bourdieu (1993, 1996), mező-küzdelmek. Ki kontrollál, legitimál, definiál; demokratizáció, dekolonizáció, instrumentalizmus.
  • D — Mediation & transfer (mediáció és transzfer). Galla, Horváth és Cosovan (2024, Creativity Studies 17(1), p. 282-294) és Cosovan (2009): a designkommunikáció mint mediációs, episztemikus struktúra. Ez a fehér folt, és a disszertáció hozzájárulási tere.

A MATCH = N jelei (az inflációs indikátorok). Egy tanulmány N kódot kap, ha a szerző maga „co-production”-t mond; ha a „participation” jelenlétet, látogatást vagy konzultációt jelent; ha puszta közösségimédia-engagement; ha value-in-use vagy élmény-érték a service-dominant logika értelmében (Vargo és Lusch; Grönroos; Prahalad és Ramaswamy), ahol a fogyasztó a saját élményét hozza létre, nem a tartalmat; ha lab-stimulus; vagy ha a co-creation háttérfogalom egy governance-elemzésben.

 

4. A DEPTH négy szintje (L1-L4)

A DEPTH a rendszer központi lencséje: nem azt kérdezi, miről szól egy tanulmány, hanem hogy milyen mélyen vizsgálja magát a co-creationt. Alapja Orthel (2022, International Journal of Heritage Studies 28(1), p. 44-58) mélységi gradiens módszere, amely a vizsgálati mélységet három fokon rétegzi (direct/causal, rigorous secondary, indirect/implied), Groat és Wang (2013) Thinking-Finding-Sorting-Writing ciklusára építve (Orthel (2022), p. 45). Ezt a gradiens adaptálódik négy szintre, amelyek egyúttal a tézis tengelyét is hordozzák: az L3 (mechanizmus) és az L4 (értékkonstitúció) közötti különbséget.

A négy szint.

  • L1 — Felszín (Surface). A „co-creation” csak címke, vagy a részvétel jelenlétet, látogatást jelent; maga a fogalom nincs vizsgálva. Döntés: ide kerül egy cikk, ha a szöveg legfeljebb említi a co-creationt (pl. tiszta gazdaságtörténet vagy policy-leírás, ahol a co-creation nem is téma). A 116-ból 11 tanulmány ilyen (mindegyik MATCH = N)
  • L2 — Kontextus (Context). A ko-kreatív keret jelen van, és a tanulmány a kimenetet méri vagy leírja, de a mechanizmus black box: nem bontja ki, hogyan jön létre az érték. Döntés: ide kerül egy cikk, ha méri a hatást vagy az élményt, de a „miért, hogyan” kibontatlan marad. Tipikus eset: a legtöbb empirikus governance- és menedzsment-tanulmány; ez a legnépesebb szint (54 tanulmány), és túlnyomórészt MATCH = N (csak 15% Y).
  • L3 — Mechanizmus (Mechanism). A tanulmányok itt a folyamatot elemzik (milyen szereplők, lépések, eszközök kerülnek használatba), de az értéket adottnak veszik. Döntés: ide kerül egy cikk, ha felbontja a folyamatot (lépések, szerepek, eszközök, fokozatok), de nem teszi kérdéssé magát az értéket. Tipikus eset: a folyamat-elemző esettanulmányok; itt billen át a MATCH túlnyomó többségbe (76% Y).
  • L4 — Foundational (alapozó). Maga a co-creation, illetve az érték természete válik a vizsgálat tárgyává (értékkonstitúció, ontológia, a fogalom kritikai vagy elméleti újragondolása). Döntés: ide kerül a tanulmány, ha magát az érték vagy a co-creation fogalmát teszi kérdéssé. Tipikus eset: az elméleti, konceptuális darabok; hét tagú klaszter (#4, #38 Graham, #47 Holt, #67 Luonila, #87 Nakajima, #93 Preece, #98 Sachs Olsen), mindegyik MATCH = Y. Horgonyok: Kaszynska (2025), p. 470 (az érték interszubjektíven igazolt); Sachs Olsen (2022), p. 319 (a spokesperson „brings the voice into being”, azaz a mediáció hozza létre a hangot); a participatív művészeti darabokhoz Bishop, Bourriaud (2002) és Kester (2004).

A besorolás döntési logikája. A szinteket a tanulmány érvelési szerkezetéből olvassuk ki, nem a témájából: mit kezel adottként, és mit problematizál. A heurisztika: (a) csak említi a co-creationt: L1; (b) méri a hatást, de a mechanizmus kibontatlan: L2; (c) felbontja a folyamatot: L3; (d) magát a fogalmat teszi kérdéssé: L4. Ez a döntés definíció szerint teljes szöveget kíván: a heurisztikus, absztrakt- vagy előző adatbázis alapján előkódolt cikkek a teljes olvasat révén eltolódhatnak besorolásukat tekintve (pl. Amer esetében a CSV-alapú L2-besorolás full-textből L3-ra módosult).

Két fajta határeset van, és mindkettő önálló módszertani lelet:

  1. Alábecsült mélység (az L3/L4 határ). A REVISIT-körben tizenkét határesetet olvastunk újra magas ráfordítással; ezek közül háromnál a DEPTH L3-ról L4-re változott: #38 Graham (a co-production nemcsak kormányzati, hanem episztemológiai tét is), #87 Nakajima (a termelő és a fogyasztó közötti határ természete a tét), #93 Preece (a „composition of value”, azaz maga az értékkonstitúció a tét). Vagyis a sekély olvasat szisztematikusan alábecsülte a legfontosabb (foundational) réteget: a hét tagú L4-klaszter fele csak újraolvasásra került a helyére. A vizsgálati mélység szerinti rétegzés maga is mélységi olvasatot igényel.
  2. Mély, de nem co-creation (a DEPTH és a MATCH függetlensége). Tíz tanulmány L3 vagy L4 szinten áll, mégis MATCH = N (#19 Chaney, #21 Choi és Burnes, #23 Colomer, #25 Davies, #61 Kristal, #65 Li és Coll-Serrano, #72 Mason, #94 Pulh és Mencarelli, #101 Stadler és Fullagar, #108 Valtysson). Komoly elemzői apparátus irányul valami másra (value-in-use, engagement, governance, szervezeten belüli kollaboráció) a co-creation címkéje alatt. Ezek mutatják, hogy a DEPTH és a MATCH két független tengely: a mélység nem garantálja a valódi co-creationt. Egyúttal ez a fogalmi infláció legerősebb bizonyítéka, mert itt nem hanyagság, hanem szisztematikus fogalomcsúszás van.

Miért fontos. A DEPTH × MATCH kereszttábla monoton gradienst ad (L1-től L4-ig 0%, 15%, 76%, 100% MATCH = Y, lásd 7. szakasz): minél mélyebben vizsgál egy tanulmány co-creationt, annál valószínűbb, hogy az valódi. E gradiens egyik tengelye sem kódolható absztraktból, ezért klasszikus, kulcsszóalapú áttekintésben elő sem állítható.

 

5. A RELATIONS és öt dimenziója

Tanulmányonként minimum három és maximum öt „X vs Y” formátumú fogalmi feszültségpár (pl. competition vs collaboration, top-down vs bottom-up, intrinsic vs instrumental) határozza meg a cikkeket, amelyeket a tanulmány saját értelmezése során vontunk ki, nem külső sablonból illesztettünk rájuk. A korpuszon ez 493 párt ad ki .

Az öt dimenzió révén minden pár az öt család egyikébe sorolódik:

Dimenzió Mit ragad meg Tipikus párok Mit mutat
POWER Ki kontrollál, dönt, legitimál top-down vs bottom-up; authority vs legitimacy; institutional control vs community agency A domináns dimenzió (a tanulmányok 61%-a); N-asszociált. A mező elsősorban legitimációs diskurzus, gyakran valódi co-creation nélkül.
AUTHORSHIP Ki a szerző, kié a mű artist vs audience; producer vs consumer (prosumer); single vs collective authorship; co-creation vs co-production Az egyetlen Y-többségű dimenzió (39:27). Ahol a szerzőség a kérdés, ott a co-creation jellemzően valódi.
SCALAR Lokális vs globális, mikro vs makro, fizikai vs digitális heritage branding vs community identity; local scene vs global market Összecseng a 76%-os lokális scope-pal; az érték transzlokális léptékváltása alulelméletezett.
AXIOLOGICAL Mi az érték, ki definiálja cultural value vs economic value; intrinsic vs instrumental; authenticity vs commodification A legritkább dimenzió (28%). Az érték-kérdés a mező saját vakfoltja, pontosan ott, ahová Kaszynska lencséje illik.
TEMPORAL-ONTO Folyamat vs termék, tradíció vs innováció preservation vs living heritage; process vs artefact A heritage-kutatás gerince; a „process vs artefact” pár az L3/L4 határ nyelvi jelzője.

Az öt dimenzió az alábbi módszertani elköteleződéseken nyugszik:

  • Axiális kódolás (Strauss és Corbin; Horváth és Mitev (2023), p. 108-109): a kódolás a kategóriák közötti kapcsolatokról szól, nem csak a kategóriákról.
  • Paralogikus, illetve transzgresszív érvényesség (Lather; Horváth és Mitev (2023), p. 66-67): a paradoxon adat, nem zaj. A feszültségpárt feszültségként tartjuk meg, nem oldjuk fel.
  • Eisenstein-montázs (Horváth és Mitev (2023), p. 39-40): az ellentétek ütközése új minőséget hoz létre (1+1=3). A RELATIONS generatív, nem leltár; a kéziratban a dimenziók egymásnak feszülnek, nem külön fejezetek.
  • Becker „art words” (Cluley (2012), p. 212 nyomán): az ellentétes szókészletek horgonyozzák a cselekvést; ahogy egy tanulmány megnevezi a feszültségeit, az maga is pozicionálás (pl. az „art-versus-money” szkript).

Az integratív áttekintés itt is eltér a klasszikus műfajtól: a „hasonló vagy ellentmondó eredmények” táblázata (Paul és Criado (2020), p. 4) helyett az ellentmondások maguk a mező problématérképe.

6. A kódolási döntések alapjai (módszertani alapállás)

Az egyes változókon belüli döntéseket néhány explicit módszertani elköteleződés vezérli.

  • Mélység a leltár helyett. Orthel (2022): a vizsgálati mélység szerint rétegez (felszín, mechanizmus, foundational), nem téma szerint. Ettől lesz a DEPTH a szervező tengely, nem a téma.
  • Egyetlen módszer sem semleges; a kódolás szituált értelmezés. Charmaz (2014, Qualitative Inquiry 20(9), p. 1076), konstruktivista grounded theory: a kódok az adattal való interakcióban konstruálódnak, nem feltárulnak.
  • Trustworthiness, nem kódolók közötti megbízhatóság. Lincoln és Guba (1985), p. 314: az egykódolós design interszubjektíven igazolt, explicit codebookon, auditálható adatbázisokon és dokumentált REVISIT-körön keresztül, nem kódolók közötti statisztikával.
  • Értelmező kézikönyv. Horváth és Mitev (2023): a kódolási (p. 110-121) és önreflektív (p. 263, 276) szakaszok adják a kvalitatív mesterséget (kristályosítás, introspekció, az Eco-féle nyitott mű olvasat, p. 379).
  • Négy megnevezett értéklencse. Kaszynska, Klamer, Woronkowicz és Noonan (lásd fent) a CAT elméleti horgonyai; Throsby és Klamer az EMPHASIS-é; Becker, Luonila és Jyrämä, valamint Brandsen és Honingh a MATCH-é.
  • Designkommunikáció mint alkalmazás. A CAT=D (Galla, Horváth és Cosovan (2024); Cosovan (2009)) episztemikus praxisként lép be, amelyen keresztül az értelmezés végbemegy: a research design és a kutatás tárgya reflexív viszonyban áll. Fekete (2024, p. 95, 98) participatív dinamikus vizuális identitásának mintájára a dokumentált codebook állandó „research brand”-ként viselkedik, amelynek befogadása az olvasó szerepe szerint változik.

Két védőkorlát, amely az alapállást őszintén tartja:

  1. Reflexivitás, nem önigazolás. A value-konstitutív olvasat nincs előfeltételezve; a codebookban olyan teszt, amelyen egy tanulmány meg is bukhat, és amelyet a MATCH és a CAT rögzít. A 116 tanulmány többségénél meg is bukik (MATCH = N ≈ 60%). A módszer és a tárgy megfelelése reflexivitás, nem körkörös bizonyítás.
  2. Konstitutív, nem prediktív. A RELATIONS és a MATCH együttmozgását (különösen az AUTHORSHIP dimenzióban) a mező önleírásaként olvassuk (a fogalmi infláció nyelvi lenyomata), nem úgy, hogy „a keret előrejelzi a MATCH-et”. Mindkét kód ugyanabból az egy full-text olvasatból, ugyanattól a kódolótól származik, és a co-creation vs co-production feszültség épp a MATCH kérdését fogalmazza újra; a kettő tehát nem független mérés, a korpusz pedig cenzus, nem minta, ezért semmilyen prediktív érvényességi állítást nem teszünk.

7. Eredmények

Hat eredmény, mindegyik auditálható lábnyommal az adatban:

  1. Fogalmi infláció. A 116 tanulmányból csak 46 (40%) valódi, tartalmi co-creation (MATCH = Y); a maradék 59% nem az, aminek mondja magát. Egy kulcsszóalapú áttekintésben mind a 116 a co-creation bizonyítékaként számítana.
  2. Monoton mélységi gradiens. Minél mélyebben vizsgál egy tanulmány co-creationt, annál valószínűbb, hogy az valódi: a MATCH-Y arány L1-től L4-ig 0%, 15%, 76%, 100%. Egyik tengely sem kódolható absztraktból.
  3. Az inflációs cella. Tíz tanulmány L3 vagy L4 szinten áll, és mégis MATCH = N: komoly elemzői apparátus valami másra (value-in-use, engagement, governance) irányul a co-creation címkéje alatt. Ez a szisztematikus fogalomcsúszás legerősebb bizonyítéka, amelyet a korpusz saját szavai is alátámasztanak („applied too liberally”, Walmsley, p. 117; „inherently emancipative” kritika, Sachs Olsen (2022), p. 316).
  4. A kétrétegű kulcsszó-érv. A 32 szigorúan „co-creation”-címkés tanulmányból 9 (28%) üres (azaz címkézett, de nem valódi); a 46 valódiból 23-nak (50%) nincs co-creation kulcsszava (láthatatlan a kulcsszó-áttekintésnek), közülük három L4 szintű. A két olvasat statisztikailag különböző mérés (φ ≈ 0,41), tehát komplementer.
  5. A fehér folt és a kettős vakfolt. Mindössze 9/116 (8%) teszi a mediációt vagy transzfert elsődleges lencsévé (CAT=D); másodlagos lencseként is csak körülbelül 25/116-ig jut. A korpusz által felszínre hozott négy „mediátor-figura” ad hoc jelenik meg, közös elméleti nyelv nélkül. Ez két további hiánnyal konvergál: az axiológiai kérdéssel (a legritkább RELATIONS-dimenzió, 28%) és a kognitív, episztemikus hangsúllyal (csak 18/116). A mező arról vitázik, ki dönt és ki alkot, de ritkán kérdezi, mi jön létre, és hogyan kerül át a jelentés.
  6. RELATIONS: a hatalom dominanciája. A mező vezető problématere a hatalom (a tanulmányok 61%-a), nem a szerzőség; a „co-creation” irodalom elsősorban legitimációs diskurzus. A szerzőség az egyetlen Y-többségű dimenzió, az axiológiai pedig a mező vakfoltja.

A hozzájárulás (a tézis). A designkommunikáció nem a co-creation egyik formája, hanem a kulturális érték meaning-making folyamatának episztemológiai infrastruktúrája. Pontosan a fenti kettős rést tölti be: az L4 ontológiai belátását (az érték a co-creation folyamatában konstituálódik, nem pedig kívülről adódik hozzá: Kaszynska (2025), p. 470; Sachs Olsen (2022), p. 319) köti össze az L3 mechanizmus-szintjével, operacionalizálva, mikor válik a részvételből közös értékteremtés (value co-creation).

7. Horgony-hivatkozások jegyzéke (név, évszám, tanulmányban elfoglalt hely)

  • Becker, H. S. (1974) Art as Collective Action; (2008) Art Worlds. A kollaboráció nem co-creation (Woronkowicz és mtsai. (2026), p. 615).
  • Bourdieu, P. (1993) The Field of Cultural Production; (1996) The Rules of Art. Mező-küzdelmek. CAT=C.
  • Brandsen, T. és Honingh, M. (2016). Co-creation vs co-production (Sachs Olsen (2022), p. 316 nyomán). MATCH.
  • Charmaz, K. (2014) Qualitative Inquiry 20(9), p. 1074-1084. „No method is neutral” (p. 1076); kétnyelvű kódolás (p. 1078). Konstruktivista alapállás.
  • Cluley, R. (2012). Becker „art words” (p. 212). RELATIONS.
  • Cosovan, A. (2009) DISCO: Co&Co Communication. CAT=D.
  • Galla, D. D., Horváth, D. és Cosovan, A. R. (2024) Creativity Studies 17(1), p. 282-294. Designkommunikáció. CAT=D.
  • Horváth, D. és Mitev, A. (2023) kvalitatív kézikönyv. Kódolás (p. 110-121), axiális kódolás (p. 108-109), paralogikus érvényesség (p. 66-67), montázs (p. 39-40), önreflexió (p. 263, 276), Eco nyitott mű (p. 379).
  • Kaszynska, P. (2025) Cultural Trends 34(4), p. 463-477. A kulturális érték normatív, hermeneutikai, interszubjektíven igazolt (p. 470). CAT=A.
  • Klamer, A. (2012). Co-creation: makers, lovers, users. CAT=B és EMPHASIS.
  • Lincoln, Y. S. és Guba, E. G. (1985) Naturalistic Inquiry. Trustworthiness (p. 314).
  • Luonila, M. és Jyrämä, A. (2019) Arts and the Market 10(1), p. 1-17. Co-creation vs co-production, „intertwined yet distinct”. MATCH.
  • Noonan, D. (2003) Journal of Cultural Economics 27, p. 159-176. CVM/WTP (p. 162, 172). CAT=C.
  • Orthel, B. (2022) International Journal of Heritage Studies 28(1), p. 44-58. Mélységi gradiens módszer. DEPTH.
  • Paul, J. és Criado, A. R. (2020) International Business Review 29:101717. Az SLR-műfaj szabálykönyve. Phase 1.
  • Sachs Olsen, C. (2022). A spokesperson „brings the voice into being” (p. 319). CAT=D és L4.
  • Strauss, A. és Corbin, J. Axiális kódolás. RELATIONS.
  • Throsby, D. (2001) Economics and Culture. A kulturális érték dimenziói. EMPHASIS és DOMAIN.
  • Vargo, S. és Lusch, R. (S-D logic); Grönroos; Prahalad és Ramaswamy. Value-in-use. MATCH=N.
  • Woronkowicz, J. (2021) Journal of Cultural Economics 45, p. 519-526. „market for symbolic meaning” (p. 520). CAT=C.
  • Woronkowicz, J., Noonan, D. és Skaggs, R. (2026) Work and Occupations 53(3), p. 599-634. Becker (p. 615), Bourdieu és Beckert (p. 617). Benchmark.

Related Working Papers

    SLR

    Art and co-creation: an integrative systematic literature review
    Extended abstract

    Keywords: co-creation; cultural value; arts management; systematic literature review; integrative review; designcommunication; mediation; value constitution.

    Co-creation is now one of the organising terms of arts and cultural management. Museums invite visitors to co-curate, festivals are built with their audiences, heritage sites are activated with communities, and cultural policy frames participation as both a democratic good and a source of value. Yet the field rarely asks whether the word names the same practice across its uses. In a review built from titles, abstracts and keywords, every such study counts as evidence of co-creation. The question is whether that count survives a full-text reading.

    Method and data
    I report a two-phase, integrative systematic review of the literature on art and co-creation (135 records, 116 in-scope studies, 2005–2024). Phase 1 is a conventional, PRISMA-conformant structured review that satisfies the classic genre rulebook (Paul and Criado (2020)): three databases, a corpus in the accepted size range, structured tables, an audit trail. It reconstructs the field from its searchable metadata. Phase 2 is an integrative extension built on Orthel’s (2022) depth-gradient model: instead of inventorying what the field writes about, it reads what each study actually argues, and the questions it never asks. Every study was read in full text and coded on six variables (domain, locus of value, emphasis, depth, category, match) and a map of conceptual tensions.

    Findings
    The two readings disagree systematically, and the disagreement is the finding.

    • Only 46 of 116 studies (40 percent) involve genuine shared control over what is created; the rest use co-creation as a label, a participation synonym or a legitimation device, often describing what is closer to co-production (co-production and co-creation are intertwined but not interchangeable; Luonila and Jyrämä (2026)).
    • The likelihood that co-creation is genuine rises monotonically with the depth at which a study examines it: 0, 15, 76 and 100 percent across the four depth levels. Surface-level studies never sustain the claim; foundational-level studies always do.
    • Mediation, the work by which meaning is carried between actors, is the field’s white spot: it is the primary analytical lens in only 8 percent of studies. The corpus produces a handful of isolated mediator figures without a shared theoretical language.
    • Half of the genuine corpus (23 of 46) carries no co-creation keyword in its searchable metadata, so a keyword-only review would lose half the real field, including three of the deepest studies. The surface and depth layers are statistically distinct (φ ≈ 0.41, reported as a quarantined descriptive figure, not an inference).

    Contribution
    The paper offers three things: an integrative review method calibrated to a meaning-laden construct, where the gap between keyword-surface and full-text reading is itself the result of a quantified critique of keyword-based reviews; and a theoretical bridge. I argue that the definitive framework of designcommunication shows that meaning-making can’t be simplfied into a form of co-creation but it’s the epistemological infrastructure of the process by which cultural value is constituted, linking a foundational ontological insight (value is constituted, not measured) to an executable mediating mechanism (Galla, Horváth and Cosovan (2024)).

    Caveats. The corpus is one specific field, not a sample, so every descriptive figure (the 40 percent, the 8 percent, and so on) is a complete count of these 116 studies, not an estimate generalised to a wider population; the figures therefore carry no confidence intervals. The one number that looks like inferential statistics, the φ coefficient, is used only to describe how strongly the keyword surface and the depth reading co-occur, and is not the basis for any claim of statistical significance or generalisation.

    References

    • Cluley, R. (2012). Art words and art worlds: The methodological importance of language use in Howard S. Becker’s sociology of art and cultural production. Cultural Sociology, 6(2), 201–216.
    • Galla, D. D., Horváth, D., & Cosovan, A. (2024). Designcommunication – a facilitator of creativity for all. Creativity Studies, 17(1), 282–294.
    • Kaszynska, P. (2025). Cultural value as meaning-making. Cultural Trends, 34(4), 463–477.
    • Luonila, M., & Jyrämä, A. (2026). Reimagining cultural policy through co-creation and co-production: Turning focus on participation and networks. The Journal of Arts Management, Law, and Society. Advance online publication. https://doi.org/10.1080/10632921.2026.2667897
    • Orthel, B. D. (2022). Linking public health and heritage work. International Journal of Heritage Studies, 28(1), 44–58. https://doi.org/10.1080/13527258.2021.1903969
    • Paul, J., & Criado, A. R. (2020). The art of writing literature review: What do we know and what do we need to know? International Business Review, 29(4), 101717.
    Methodology

    Interpretive coding as a methodological bridge in cultural value research
    Extended Abstract

    Keywords: cultural value; meaning-making; interpretivism; interpretive research; value-based approach; rigour; designcommunication; mixed-methods synthesis.

    Cultural policy speaks the language of measurability; economics is its lingua franca (Kaszynska (2025), p. 471), and the what-works agenda demands numerical evidence. But if cultural value is the value of meaning-making, and is normative, hermeneutic and intersubjectively warranted (Kaszynska (2025), p. 463 and p. 470), then individual-utility aggregation cannot in principle reach it. The difference between measuring and meaning is not one of degree but of kind. The problem, however, is not solved by abandoning measurement. The is whether interpretive, meaning-sensitive work can be done so that it keeps the visible marks of systematicity and rigour without freezing meaning into variables.

    The collaboration and what it contributes
    This is a collaborational work with Patrycja Kaszynska. Her statements: cultural value as meaning-making, requiring an interpretive research approach. This paper contributes four things: what these interpretive approaches are and how they developed; why they are not taken seriously in the current climate; why they should be; and what a concrete, coded corpus can say about all this. The axis of the argument is a distinction between interpretive methods (techniques such as interviewing or ethnography, which can be used inside a positivist paradigm) and interpretivism (a paradigmatic commitment to constructed social reality and a non-neutral researcher). When an interpretive technique has to have a positivist paradigmatic output, meaning is exactly what is lost. False interpretivism is not a theoretical worry but a demonstrable practice.

    The argument.

    • The missing fourth step. The value-based approach (Petrova, Graça and Klamer (2022)) is a serious, good-faith interpretive attempt, which is why it shows the break point. Its first phase genuinely articulates values with participants; its later phases convert them into variables and percentage rankings, and a largest-gap-equals-largest-impact logic freezes meaning into a numerical output. The structural reason is stated in the text: this is what funders look for (p. 122), and the proposed remedy is more data, not deeper interpretation (p. 123). What is missing is a fourth, interpretive synthesis step that returns the measured result to the field of discovered meanings.
    • The form of rigour without quantitative pretence. Noonan’s (2003) meta-analysis supplies the model, not by its content but by its discipline: rigour lies not in the regression but in how the paper epistemically fences what its numbers mean and do not, putting the weight on direction, significance and patterns, not precise values (p. 167), and naming its own model stylised and non-extrapolable (p. 164). The same fence is transferable to coding that stays unambiguously qualitative.
    • The two-layer demonstration. The corpus answers both lacks. The systematic surface (PRISMA screening, keyword description) satisfies the demand for regularity; the full-text interpretive layer keeps meaning as a unit of analysis. Read as a census (n = 116): 78 percent of studies stay on the lower two depth levels; 40 percent genuinely treat co-creation as meaning-making; and the two layers diverge (φ ≈ 0.41). That φ is the key sentence for the measurement-minded reader: measurement is not wrong, but measurement and interpretation together capture more than either alone, because they demonstrably grasp different things.

    Institutional anchor. O’Connor, Whiting and Barnett (2026) give the macro-level twin of Kaszynska’s claim: objective respect points to art as an essential, irreducible dimension of shared social life (p. 11), against the intrinsic-value move that reduces cultural value to subjective choice and infinite metrics. They name the need for mediation but do not give the mechanism. The same structure recurs across the diagnoses: the diagnosis is right, the operationalisation is missing. Designcommunication (Galla, Horváth, Cosovan, 2024) is that missing mechanism: not the content but the frame and medium that make value possible and legible. The same pattern is visible in current arts-management work that locates value in the encounter (Carnwaith and Brown (2014), as cited in Luonila and Jyrämä (2026), p. 5) yet closes by asking how impact can be measured.

    Caveats. The distinction between interpretivism and interpretive methods must not become an immunising shield; the aim is a working methodological argument, not conceptual fence-building. The critique of the value-based approach is structural and unnamed throughout; the Petrova–Klamer circle’s data discipline is genuinely valuable. The corpus figures are a census; n is not a sample.

    References

    • Bevir, M., & Rhodes, R. A. W. (2016). The Routledge handbook of interpretive political science. Routledge.
    • Carnwaith, J. D., & Brown, A. S. (2014). Understanding the value and impacts of cultural experience – a literature review. Arts Council England.
    • Crossick, G., & Kaszynska, P. (2016). Understanding the value of arts & culture: The AHRC Cultural Value Project. Arts and Humanities Research Council.
    • Galla, D. D., Horváth, D., & Cosovan, A. (2024). Designcommunication – a facilitator of creativity for all. Creativity Studies, 17(1), 282–294.
    • Kaszynska, P. (2025). Cultural value as meaning-making. Cultural Trends, 34(4), 463–477.
    • Kaszynska, P. (2026). Culture-as-infrastructure: The functional, the political and the need for poetic. Journal of Cultural Economy. Advance online publication.
    • Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Sage.
    • Luonila, M., & Jyrämä, A. (2026). Reimagining cultural policy through co-creation and co-production: Turning focus on participation and networks. The Journal of Arts Management, Law, and Society. Advance online publication. https://doi.org/10.1080/10632921.2026.2667897
    • Noonan, D. S. (2003). Contingent valuation and cultural resources: A meta-analytic review of the literature. Journal of Cultural Economics, 27(3–4), 159–176.
    • O’Connor, J., Whiting, S., & Barnett, T. (2026). Do give up your day job: Basic income for artists, and other stories. Cultural Trends. Advance online publication. https://doi.org/10.1080/09548963.2026.2655376
    • Petrova, L., Graça, S., & Klamer, A. (2022). Evaluating qualities of cultural production: A value-based approach. Media Practice and Education, 23(2), 112–125.
    • Schwartz-Shea, P., & Yanow, D. (2012). Interpretive research design: Concepts and processes. Routledge.
    Interviews

    Interviewer matrix and the re-reading of qualitative interviews through the SLR coding scheme
    Extended Abstract

    Keywords: co-creation; qualitative interviews; interviewer matrix; arts management; mediation; designcommunication; value constitution; invisible labour.

    The integrative review (Phase 2) produced three results that can only be tested on primary ground: co-creation that is invisible to keyword search (half of the genuine corpus carries no co-creation keyword); mediation as the field’s white spot (primary in 8 percent of studies); and an axiological blind spot (the value-theory tension is rare, 28 percent). The question this phase asks: if working mediation is invisible precisely because it succeeds, do practitioners say aloud what publications do not write down? If yes, the blind spot is not the absence of the phenomenon but a limit of the instrument, and that is the empirical stake of the thesis.

    Method: the interviewer matrix
    Eleven semi-structured interviews (8 from the Indiana University, Bloomington field; 3 Hungarian/European; several multi-speaker). To analyse them I used the same coding and analytical variables as the SLR studies (domain, locus, emphasis, depth, category, match), together with the tension map, which made the interviews directly comparable with the 116-study corpus. With only eleven interviews, the layer illustrates patterns; it does not prove or generalise them. The names of the respondants has been altered.

    Findings:
    The two layers reverse.

    category A (value) category B (process) category C (power/legitimation) category D (mediation)
    SLR corpus (n = 116) 37% 18% 37% 8% (white spot)
    Interviews (n = 11) 9% 0% 9% 82% (dominant)

    The corpus mediation white spot becomes the dominant register in the interview layer. On the depth axis the corpus mass sits at the context level; in the interviews the lower levels are empty and the mass moves to the mechanism and constitution levels (about 82 percent match). This is not a contradiction but a sampling difference: the published corpus carries co-creation mostly as a label or legitimation, while practitioners speak about it at the level of constitution, only they do not name it so and do not write it down. At the finer, response-level profile (about 82 coded statements) the sharpest reversal is on the relations axis: the axiological dimension, the corpus’s third blind spot (13 percent), becomes the dominant fracture line in the interviews (35 percent, +22 percentage points).

    Quotable evidence for the self-evident-hence-unwritten claim. Carrie: “because it’s self-evident, it’s never written down, never researched.” Lydia: “all the good design is invisible […] no one really thinks about that.” The field knows as practice what it does not state as text.

    Four interpretive patterns. (1) An inward/outward map of value validation: foundational depth does not imply a participation bias (the autarkic pole, Corey and David, is depth-4 but match-N), a point echoed in arts-management interviews where some audiences prefer anonymity to involvement (Luonila and Jyrämä (2026)). (2) The axiological axis is a real fault line, appearing as a difference between persons (Anna on measurement, Carrie on constitution), not a coding artefact. (3) Mediation as the epistemological infrastructure of value, the core of the thesis: designcommunication is not the content but the frame and medium that make value possible and legible, linking the foundational ontological insight to a concrete mechanism. (4) Invested, invisible design: value-producing work is structurally invisible to the measuring, funding and recognising systems; this is the living cause of Becker’s sampling error (Cluley (2012)), the mechanism behind the invisible half of the corpus.

    The most robust finding. All eleven respondents articulate success, and none makes money the primary goal; the recurring core is connection and being moved, the survival of value, and leaving things better, with money only as a precondition.

    The connecting link. The value-based critique (Kaszynska (2025), p. 470; O’Connor, Whiting and Barnett (2026), p. 11) correctly names that value must be constituted intersubjectively, but does not give the mechanism. The outcome of the interviews (Phase 3) are showing the missing mechanism already at work in practice, only unnamed: designcommunication is the execution structure, the workflow that turns the critique into practice.

    Caveats. The interviewees are diverse and occupy different areas of the art world, but I came to know the eight US respondents in Bloomington, so it can be assumed that they reached me within the same environment and along a similar framing. The community-centred (outward) pole comes from people working in a small, tightly connected local scene, while the autonomous-artist (inward) pole comes from people in secure academic positions; these privileged conditions are stated openly in the analysis. I have friendly or collegial relationships with several of the interviewees, which made it easy to check my readings back with them, but may also have shaped how they presented themselves. Eleven interviews can only illustrate patterns, and the finer percentage profile reflects my own coding judgement: the overall directions are reliable, the exact decimals are not. All quotations are verified with the interviewees following a member-check protocol (Lincoln and Guba (1985), p. 374–378).

    References

    • Carnwaith, J. D., & Brown, A. S. (2014). Understanding the value and impacts of cultural experience – a literature review. Arts Council England.
    • Cluley, R. (2012). Art words and art worlds: The methodological importance of language use in Howard S. Becker’s sociology of art and cultural production. Cultural Sociology, 6(2), 201–216.
    • Galla, D. D., Horváth, D., & Cosovan, A. (2024). Designcommunication – a facilitator of creativity for all. Creativity Studies, 17(1), 282–294.
    • Karpik, L. (2010). Valuing the unique: The economics of singularities (N. Scott, Trans.). Princeton University Press.
    • Kaszynska, P. (2025). Cultural value as meaning-making. Cultural Trends, 34(4), 463–477.
    • Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Sage.
    • Luonila, M., & Jyrämä, A. (2026). Reimagining cultural policy through co-creation and co-production: Turning focus on participation and networks. The Journal of Arts Management, Law, and Society. Advance online publication. https://doi.org/10.1080/10632921.2026.2667897
    • O’Connor, J., Whiting, S., & Barnett, T. (2026). Do give up your day job: Basic income for artists, and other stories. Cultural Trends. Advance online publication. https://doi.org/10.1080/09548963.2026.2655376
    • Wandel-Brannigan, C., & Matamala, A. (2024). Spectrums of co-creation: An analysis of the co-creation process in the development of the Irish National Opera’s community VR opera, Out of the Ordinary/As an nGnách. The Journal of Arts Management, Law, and Society, 54(5), 295–314. https://doi.org/10.1080/10632921.2024.2442947
    logo_2_d_csakfeher_szuk
    info@dinnyesattila.hu
    +36/70 390 7560