Abstract
While numerous studies have shown diverse effects of rankings, rather little is known about their production. This article contributes to a broader understanding of rankings in society, and does so by focusing on underlying worldviews. I argue that the existence of a ranking and its concrete methodology can be explained by the producer’s paradigmatic assumptions about a world-to-be-ranked. Referring to the sociology of knowledge and studies on commensuration, comparisons, quantification and valuation, I provide a general heuristic to analyze this relation between underlying worldviews and observational regimes through which order is constructed systematically. Presenting empirical results on a ranking for the most famous artists in the world, I show how the review device’s initial problem and its construction of order derive from consistent assumptions about contemporary art, its symbolic structures and its social embeddedness. These findings have implications for both research on rankings and sociology of the arts.
I Introduction: Seeing like a Ranking?
Rankings play a crucial role in modern society because they offer a systematic access to complex phenomena and can have effects in observed fields. While diverse outcomes of rankings have been shown, rather little is known about the concrete production and emergence of specific cases. This article focuses on rankings’ underlying worldviews to expand an understanding of observational tools and their role in society. This broader understanding is necessary because numerous studies have proven multiple mechanisms and effects set in motion by review devices in the field they observe. Rankings can provoke reactivity in organizations and their publics (Brandtner, 2017; Espeland and Sauder, 2007; Martins, 2005). Rankings can shape, formalize and stabilize status competition structures (Brankovic et al., 2018; Sauder, 2006) that provoke agents to strategically adapt (Hazelkorn, 2014). Rankings can foster isomorphism in organizational fields (Wedlin, 2007). Rankings can establish global quality standards (Paradeise and Thoenig, 2013) and link different value regimes like quality of goods, prices and entrepreneurs’ reputation (Gioia and Corley, 2002; Rindova et al., 2005).
In contrast to feedbacks, only selective insights into rankings’ production have been developed (Rindova et al., 2018). When Espeland and Sauder (2016) describe how rankings provoke organizational reactions by offering a shared ‘cognitive map’ for agents in a particular field, the twofold role of such observational tools for a society appears. Rankings generate, stabilize and modify knowledge about specific social spheres through introducing new modes of understanding the world on the one side. On the other, they influence structures and actions based on a diffusion of these particular modes of ordering and evaluating. To advance a sociology of rankings that needs to capture such complex processes, I suggest investigating the epistemology and knowledge that shape the final cognitive map. I argue that every ranking is already based on and shaped by its producer’s paradigmatic cognitive map of a world-to-be-ranked. I put a strong emphasis on how the world is understood in the first place to explain what a ranking can actually ‘see’ (Fourcade and Healy, 2016; Scott, 1998) and model into tight order subsequently. Conceptualizing such a seeing like a ranking as the relation between paradigmatic ways of understanding the world and the specific production of order follows seminal approaches from both sociology of knowledge (Mannheim, 1997) and sociology of science (Fleck, 1979; Kuhn, 1996). An underlying worldview contains ‘[e]pistemological, ontological, methodological, and other kinds of assumptions’ (Abend, 2018: 108) about an object of inquiry. These paradigmatic assumptions determine what can be observed, what kind of information is produced in which way and within which epistemological systems (Knorr-Cetina, 1999).
I argue that a paradigmatic worldview explains not only the sheer existence of a ranking but also its form. This perspective complements explanations for the emergence and success of rankings in contemporary society. Rankings have been described as integral parts of modern knowledge governments (Erkkilä and Piironen, 2018) or functional observational tools in a polycontextural society (Esposito and Stark, 2019). Rankings are also theorized as smooth mechanisms in disciplining regimes (Sauder and Espeland, 2009) or neoliberal technologies in a crisis-shaken global political economy (Welsh, 2020). My approach rather focuses on individual cases. Even if a certain ranking is functional within particular power structures, it is not necessarily installed from above or directly determined by discursive, ideological or economic structures. Equally assuming that a ranking does not emerge randomly, the production of a disclosure device (Hansen and Flyverbom, 2015) in general and a specific option from a variety of review devices (Blank, 2007) in particular, has to make sense to producers and audiences fundamentally.
Presenting empirical results on a ranking of the most famous artists in the world called Kunstkompass (Art Compass), I show how underlying assumptions about art in society shape this production of tight order. In general, compiling any ranking contains and interconnects several operations to derive discrete order from noise (von Foerster, 2003a): defining a category that will be ranked; defining criteria to identify differences between units of this category; defining a way to evaluate these differences; and (in some cases) defining a common metric for quantified evaluations or measurements. These types of observations have been the object of broad sociological research on commensuration (Espeland and Stevens, 1998), comparisons (Heintz, 2010, 2016), quantification (Espeland and Stevens, 2008; Mennicken and Espeland, 2019; Porter, 1995), classifications and categorizations (Bowker and Star, 2000; Fourcade, 2016; Zerubavel, 1996), statistical techniques (Alonso and Starr, 1987; Desrosières, 2002; Hacking, 1990; Porter, 1986), valuation and evaluation (Helgesson and Muniesa, 2013; Hutter and Throsby, 2008; Karpik, 2010; Lamont, 2012), and standardization of measures (Kula, 2016; Zerubavel, 1982).
Following the general constructivist approach shared by this literature, I turn inevitable contingencies in constructed observations into a methodical program. Referring to a concept of observation developed in second-order cybernetics and systems theory (von Foerster, 2003b; Luhmann, 1995), I analytically split up a ranking’s methodology into single observations. I approach each one of them and each interconnection between them as contingent, which allows asking for their plausibility regarding an underlying worldview. This procedure shapes a general research approach towards the social construction of order. If an observational tool does not represent a given world but selectively produces an own world, how is it done and why is it done in this way? I theorize the concrete procedure of constructing a world as an observational regime to capture the consistent observational architecture that regulates the production of order. I argue that producers’ paradigmatic assumptions about a world narrow down alternatives and shape an observational regime.
Researching observational regimes and underlying worldviews connects to general questions on social stasis and flux. Certain art rankings are proper cases because they compare and quantify in a field based on semantics like originality, incommensurability, legitimate multiplicity of criteria, ephemeral symbolic value, and individual perception. My study refers to sociologies of art that understand art as a social sphere with symbolic autonomy. I methodologically consider autonomy a possibility but ask if and how this autonomous sphere is specifically understood from within. Taking Pierre Bourdieu’s (1993, 1996) and Niklas Luhmann’s (2000) theories on art as vantage points, I assume that a multiplicity of worldviews inside the arts stabilize and alter autonomous structures because symbolic resources emerge through mutual recognition based on worldviews.
The article will proceed in four steps. In Part II, I present an analytical framework for reconstructing observational regimes and paradigmatic worldviews. In Part III, I introduce the Kunstkompass and the field of art rankings and sketch their role in the sociology of the arts. In Part IV, I reconstruct the Kunstkompass’ worldview and how it constitutes its observational regime. In concluding Part V, I outline implications for both the sociology of rankings and the sociology of the arts.
II Observational Regimes and Paradigmatic Worldviews
Two questions structure my analysis of a ranking to explain its existence and form: (1) How is the ranking constructed methodologically? (2) Which underlying worldview makes this observational regime plausible? A reverse-engineering of a published ranking successively reconstructs the observational regime that has led to the final list. This observational regime refers to and derives from paradigmatic assumptions about the ranking’s area of interest.
Take an observational tool like the Kunstkompass that ranks artists from all over the world based on their quantified fame. If you want to investigate its observational regime, questions come up about what the ranking actually means by global art and the role of fame in it. Here, ‘art’ needs to be understood in a certain way so that a ranking in general and measuring fame in particular makes sense at all. I argue that this understanding contains paradigmatic assumptions about contemporary art’s place in society, internal structures, and reproductive mechanisms. More specifically, relevant units (artists) and a criterion (fame) need to be defined to evaluate, compare and rank these units. Additionally, epistemological foundations regulate what counts as robust information (quantified art world events indicating fame), which offer suitable insights for targeted audiences.
In-depth analysis is based on the concept of observation and form used in second-order cybernetics and systems theory (von Foerster, 2003b; Luhmann, 1995; Baecker, 1999). In short, any observation is an indication that distinguishes a marked side from an unmarked realm of virtual alternatives (Spencer-Brown, 1969). Researching such an indication sheds light on alternatives that had been excluded. Analytically, I reverse-engineer the production of a ranking to remodel an architecture of single observations that regulate what will be seen. Understanding each observational step and every relation in this observational regime as contingent, I research for reasons why something had made sense in its particular way.
For ordering art worlds, specific categorical units (art works, artist, museum, etc.) need specific comparative (Heintz, 2016) criteria (epoch, media, style, gender, quality, skill, fame, art historical relevance, market performance, visitor numbers, media coverage, etc.). For identifying discrete, numerical differences, a quantifying ranking needs to construct a common metric for comparing highly diverse qualities (Espeland and Stevens, 1998, 2008; Heintz, 2010) to find evaluative differences (Lamont, 2012). Widely acknowledged, quasi-naturalized (Fourcade, 2016) semantical categories (Zerubavel, 1996) play a key role for an observational tool, if it wants an audience to understand its mission. For the arts, certain cultural classifications (DiMaggio, 1987) offer established observational structures to identify and describe cultural artefacts, producers, professionals and institutions. While these semantics are permanently objects of discussion about their legitimacy (Bourdieu, 1996), they are constantly used by professionals and amateurs to describe what is actually going on in the arts. Furthermore, evaluations justify decisions made in the light of an abundant supply of objects, organizations and persons. Therefore, any novel mode of observing the arts navigates a fine line between established conventions and gradual innovations (Becker, 2008).
My analysis of the Kunstkompass is based on the list and additional information published or authorized by its producers. I approached this material with general questions informed by the literature on comparisons and quantification to reconstruct an observational regime: Who made the ranking? What kind of data was used? Which units and which criteria were selected for comparisons? How are units and criteria defined? How are criteria scaled and how are values assessed? Observations made in the production of a ranking should then consistently relate to a set of paradigmatic assumptions about the world-to-be-ordered. In this analytical step, the existence of an observational tool and its single parts is approached by a simple question: Why is it done this way? Starting with the ranking’s sheer existence, reconstructing paradigmatic assumptions begins with identifying the main problem the device is addressing. This ‘call for a list’ (Staeheli, 2012; Ziegler, 2007) defines the ranking’s mission.
I argue that this call for a ranking and hence its function for a targeted audience originate in a particular understanding of the world, namely a sense of chaos and potential order. From a radical-constructivist perspective, neither order nor chaos are ontological or essential facts. Chaos only takes shape for an observer if a necessary proto-order enables initial observations, categorizations and comparisons. Without an assumption of potential order of unit x (e.g. artists, art works, cultural organizations), no informational problem exists that can be solved by a ranking.
III Measuring the Art World: Kunstkompass and Other Review Devices in the Arts
The Kunstkompass is an annual ranking of artists based on quantified fame on a global level. While private enterprises produce rankings to gain resources (Rindova et al., 2018), the mission, method and form of a ranking have to make sense within their own terms regardless of entrepreneurial rationalities and resource mobilization. Therefore, the idea for a ranking like Kunstkompass only appears if there are already assumptions about fame in the art world. In particular, sociologists are familiar with such concepts about the structural mechanisms and unequal distribution of recognition in the arts from their own writings. From a sociological perspective, the symbolic value of an art work is not derived from essential qualities of an artefact or technical skills of its producer, but is socially produced through mutual recognition within a certain symbolic domain and institutional ecology. Although there are major differences, this common sociological approach is shared by studies following diverging conceptions of the arts like field (Bourdieu, 1993, 1996), collective production (Becker, 2008; Peterson and Anand, 2004), functional system (Luhmann, 2000) or professional regimes (White and White, 1993). Here, asking for worldviews aims at understanding how agents in the arts perceive and observe their ecology. So, approaching a multiplicity of worldviews in the arts leads to understanding how symbolic structures and values stabilize and shift.
The sociological baseline towards the arts is that a specific set of institutions and professionals has historically emerged which acknowledge certain persons and objects legitimately as artists and as art works. Furthermore, specific criteria and observational regimes are commonly used to evaluate persons and objects regarding their relevance, quality or novelty. Confronted with a vast abundance of supply and scarce demand, established observational regimes for recognizing producers and artefacts are used for making inevitable selections by curators, artists, museum directors, art historians, collectors, gallerists and critics. However, today’s established ‘aesthetic systems’ (Becker, 2008: 131) and ‘principles of vision and division’ (Bourdieu, 1996: 58) in the arts work well without any quantification or discrete ranking techniques. Furthermore, modern discourse about the arts includes semantics like singularity, originality, incommensurability, qualitative expert verdicts or individual aesthetic perception. While all of them obviously contradict logics of quantification, commensuration, objective evaluation, and standardized measurement, quantifying rankings have appeared in the arts nevertheless.
There are several rankings about museum visitor figures or market prices. These tools obviously embody an ‘allodoxia’ (Bourdieu, 2008) in the arts because they represent external perspectives from ‘hostile worlds’ (Velthuis, 2005: 25ff.) like economy, tourism or cultural politics. But there are also rankings (Quemin, 2013) like Kunstkompass, ArtFacts or Power 100 that claim to observe exactly those symbolic logics and mechanisms of the arts, which seem to resist standardized measurement. Instead of counting material objects (visitors, clicks, dollars), these projects measure symbolic resource distribution by evaluating what they assume to be symbolic resources and produce a ranking organized hierarchically regarding these evaluations.
These lists are known among sociologists (and economists or art historians). Using ranking results for quantitative studies, sociologists have developed center-periphery models that show how certain regional backgrounds are unequally represented in the global art world (Buchholz, 2008, 2018; Buchholz and Wuggenig, 2005; Quemin, 2006, 2013, 2018; Quemin and van Hest, 2015). In this sense, sociological debate about art rankings needs an urgent update based on constructivist research on rankings, quantification, evaluation and commensuration. In the arts as well, rankings can be approached as selective observations that potentially feed back into art’s processual stasis and flux (Buckermann, 2018, 2020). Applying an analytical ‘agnosticism’ (Callon, 1984: 200) to controversial observations of the arts additionally opens up new research perspectives on established observational regimes used by acknowledged professionals. In general, the sociology of the arts is interested in how structures and semantics have emerged that both stabilize art and keep it in flux.
A sociology of plural worldviews in the arts asks how agents in the art world themselves observe their own ecology, which is the social phenomenon called ‘the arts’ with all its material, symbolic, semantic, professional and institutional infrastructures. How do agents navigate through a complex world and how do they establish models for decision-making? How do symbolic value regimes emerge, stabilize and develop without any formal regulation but through mutual recognition? These questions sketch a new agenda for a sociology of the arts that goes beyond traditional epistemologies ranging from l’art pour l’art to institutionalist or idiocentric approaches of explaining what should count as art (Harris, 2010). Discussing the rather odd case of rankings clearly points to the contingency, selectivity and impact of any worldview. Researching them aims at understanding how these different ‘“indigenous” theories-of-the-system-in-the-system’ (Luhmann, 1981: 199, own translation) work and how they relate to the social sphere of art.
The Kunstkompass shows the most famous living artists in the world based on numerical scores that indicate fame. This fame means public recognition by distinguished museums, biennales or magazines. General fame is indicated by individual public facts like exhibitions or critical reviews. The Kunstkompass selects, evaluates and quantifies these symptoms regarding only one kind of fame to compute absolute scores for artists. Scores for individuals are then compared to define relative positions in the ranking. Since its introduction by economist and journalist Willi Bongard 1 in 1970, the Kunstkompass has been published annually (except the year of Bongard’s death). Although the ranking has been published exclusively in German, it is internationally known for listing the most successful artists in the world on a quantitative basis. Furthermore, the Kunstkompass is the archetype and pioneer of systematic observations which investigate criteria and mechanisms that are supposed to be applied in the art world itself. Instead of market prices or visitor figures, ArtFacts or academic studies (Fraiberger et al., 2018; Galenson, 2002) indeed quantify artistic reputation, institutional success, symbolic resonance, or influence.
This is a historical novelty. The invention of the Kunstkompass marks the beginning and refers to a typical modern aspect of art’s autonomy and its differentiated value regimes. While contemporary cases measure and rank the symbolic order of fine arts, there had been quantifying evaluations of artists before. Today’s rankings differ significantly from these historical cases because they emphatically oppose immediately measuring the quality of cultural production. On the contrary, tools that were invented in Europe in the 18th and 19th century dealt with exactly this artistic quality and skill (Spoerhase, 2019). Starting with Roger de Piles’ Balance de peintre (de Piles, 1708: 489ff.), the quality of painters and poets was assessed on numerical scales.While these quantitative results are still used in economic studies today (Ginsburgh and Weyers, 2008; Graddy, 2013), contemporary rankings are not immediately concerned with aesthetic value or exquisite artisanship. Instead, they produce and compile data about reputation, institutional success or influence in the art world. However, I will show that quality has not become obsolete, and especially the Kunstkompass understands artistic quality as an indispensable factor in the emergence of symbolic structures in contemporary art.
The next sections show in detail how fame is assumed to be both a consequence and a proxy of artistic quality (and market prices 2 ). I have analyzed and interpreted 47 issues (c.1200 pages) of the Kunstkompass from 1970 to 2018. The material includes different lists, editorial comments, methodological information, and background stories on the art world. Additionally, I analyzed other publications by the producers (Bongard, 1974; Rohr-Bongard, 2001). Based on a radical-constructivist concept of observation, my qualitative reconstruction was not structured by pre-determined categories because I assumed that a ranking is only plausible within its own consistent worldview.
IV The World of the Kunstkompass
What does art and fame mean for the Kunstkompass and who should care about it? For the Kunstkompass, fame means reputation (or prestige, recognition) of artists within a specific institutional and professional field. Fame, understood as ‘resonance in the art industry’ (KK 2006: 127), 3 materializes in museum exhibitions, biennale participations, acquisitions for public collections and critical reviews in art magazines. All these different phenomena are understood as indicators for the same kind of fame of an artist. These indicators are presented as a matter of fact (Ruhmestatbestand) because they are public and can be verified. For each of these facts, an artist receives a certain amount of ‘fame points’ (Ruhmespunkte). The value of a fact derives from and mirrors the importance of the respective institution. The exact value of an institution’s capacity to ascribe fame is based on a survey among art world professionals. Assembling scores for the most important cultural institutions results in a standardized metric, which can measure the fame of virtually every person in the world. The sum of all points indicates an artist’s absolute fame in the global art world and determines a relative rank. Based on this evaluative comparison, different lists are produced for the top 100 newcomers or highest point increase over time.
Producing Chaos: Abundance and Correlations
This complex observational process has a clear function directed at the list’s main problem. Fundamentally, the Kunstkompass describes a vast chaos of artistic production that is characterized by a high amount of cultural production as well as complex internal differences. Additionally, the rankings state a lack of robust criteria for basic orientation in this chaos of the contemporary and a high insecurity in aesthetic and monetary evaluations. Here, the Kunstkompass’ epistemological consistency becomes obvious because not only these problems emerge from ascribed mechanisms and logics of art in society. Furthermore, I argue that its solution in the form of a systematic, quantifying comparison also makes sense in light of the same assumptions that constitute the call for a list.
Since 1970, the Kunstkompass has been describing a rampant increase of artistic production on an absolute level and in a stylistic dimension. The Kunstkompass regularly claimed that ‘there has never been more art and artists in history than in the present’ (KK 1977: 149) and that this ‘supply of international art is more diverse than ever’ (KK 1970: 143). Social modernization had been widening access successively to high-end institutions, artistic discourses and markets for female artists and artists coming from regions outside Western Europe and the USA. Additionally, there had been supposed a constant increase of legitimate artistic form and content, with a plurality in media, styles or themes. Regarding the 1960s and 1970s, the Kunstkompass described a ‘true inflation of new artistic directions and styles’ (KK 1977: 149) that confronted commercial, institutional or intellectual interests in contemporary art with an unimaginable abundance of artistic producers and artefacts.
In this case, conditions of ‘too little, enough, and too much’ (Abbott, 2014: 2) mean that a vast number of artists and art works make up an abundant supply relative to limited symbolic, financial and organizational resources (like exhibition space, capital, magazine pages, awards, professorships, scholarships, public and scholarly attention, art fair booths and collection shelves). Absolute and stylistic abundances make selections necessary for professionals who manage these limited resources. The Kunstkompass assumes that the multiplication of legitimate art forms makes such selections risky and contested because there is no common or standardized set of evaluative criteria for quality or relevance of new art. For ‘critics, gallerists, museum directors and even artists themselves’ (KK 1970: 143), verdicts about artistic value and historical ‘significance’ (KK 1982: 327) become more insecure because they are confronted with ‘an irritating diversity of direction, varieties and styles’ (KK 1977: 149f.). Additionally, collectors lack rigorous indicators for adequate prices and future market developments.
These problems call for the Kunstkompass. Confusion about contemporary art originates in ‘uncertainty about assessments of quality and value stability’ (KK 1972: 197). The ranking in turn ‘offers an overview about the factual representation of artists – each expressed in a numerical score’ (KK 1974: 92) based on registering the ‘interest artists receive from art experts’ (KK 1983: 308). The ranking claims identifying art that is relevant today to enable assessments of price adequacy today as well as relevance and market performance in the future. Even though the Kunstkompass seems to address a market-oriented audience primarily, it is worth noting that the list’s insights do not derive from market information. Instead, the ranking investigates fame – exclusively generated in the art world – as an allegedly adequate proxy for quality and price. Contrary to many market reports based on realized prices, the Kunstkompass focuses on symbolic differences in artistic production and takes related mechanisms in the symbolically autonomous arts as an essential fact.
Although the Kunstkompass is exclusively displaying artistic reputation, it promises insights into the enigmas of artistic quality and art market performance. These selling points are based on fundamental assumptions about fame, quality and price as differentiated but interrelated value spheres. First, past prices should not inform prognoses about futures prices because the market for new art is too volatile and idiosyncratic. Second, ‘prices paid for contemporary art do not necessarily refer to an artist’s grade’ (KK 1976: 220). Third, quality of contemporary art cannot be measured objectively. However, there is still fame: ‘Due to a lack of objective and generally binding quality standards, art works are not measurable – strictly speaking – not comparable. Measurable – and thus also comparable – is the fame that artists gain over time’ (KK 1973: 102).
The Kunstkompass claims that different values reflect each other and correlate. Fame indicates quality because the population of art world professionals adequately assess quality when they decide what is exhibited, reviewed or collected. ‘Assuming that fame based on expert verdicts is tightly related to artistic quality (statistically spoken: highly correlates), there is a justified probability claim, that the 100 most prestigious artists determined by the Kunstkompass are also the 100 greatest in the present’ (KK 1979: 215). Additionally, the Kunstkompass insists that fame (ergo quality) causes price developments over time. Only ignorant or irrational persons ‘will deny this relation: The bigger the fame of an artist, the higher are usually the prices for his works’ (KK 1976: 220).
These assumptions form a straight line of thought: prices for art works correlate with fame, which is determined by quality. Therefore, fame is supposed to be an indicator for both quality and price, which are the objects of the two main informational problems emerging from the chaos of contemporary art. In this sense, the Kunstkompass answers its own call for a list and claims to provide a rigorous multifunctional indicator: ‘Measurable and therefore comparable is the resonance that art works receive’ (KK 1993: 239), which ‘ensures a maximum of objectivity in evaluating what counts for the value of art today’ (KK 1970: 147). In this sense, the ranking offers informational solutions to informational problems, while solutions and problems both emerge from the ranking’s underlying worldview.
Producing Order: Blackboxing Swarm Intelligence
To tackle diverse problems in evaluating new art, the Kunstkompass has developed a systematic method to construct a discrete order of global art. The ranking investigates larger patterns of recognition in the totality of uncoordinated art world events. Even if buyers or passionate connoisseurs knew about the assumed correlation of fame, quality and prices, they were unable to identify and assess fame on their own. The professional and institutional field, which is granting recognition, is supposed to be confusing and individual verdicts to be unreliable. Here, the Kunstkompass promises objective insights into fame distribution based on standardized and quantified evaluations. The Kunstkompass’ method can be related analytically to two assumptions about social mechanisms: blackboxing and swarm intelligence.
The Kunstkompass understands the institutional and professional art world as a machine that produces fame. This machine supplies a public screen for fame that can be investigated. Even though individual expert verdicts are considered to be corrupted and distorted by individual taste and corrupt interests, the ranking claims that a sufficient number of verdicts produce a reliable image of an artist’s reputation. Although artistic value cannot be measured objectively, the Kunstkompass sees ‘a general consensus among experts about which persons count as great artists’ (KK 1986: 375): ‘However scholars dispute about what – already or still – is art, there is a larger agreement on what is great art and who is an outstanding artist, that one could assume regarding the diversity of cultural production today’ (KK 1977: 152). In conclusion, ‘great art is art which is perceived as such by as many museum directors, curators and publicists as possible. Although expert verdicts diverge in many cases, they concord on numerous occasions. The Kunstkompass measures these congruities and processes them into a ranking of the “100 Greatest”’ (KK 1977: 152). This unconscious consensus of art world professionals is a fact for the Kunstkompass, even though expert criteria for evaluating quality seem to be blurry.
The ranking is not interested in the criteria’s form or emergence but only in their outcomes, which is assumed to be reliable because the population of art world experts are supposed to possess a kind of professional swarm intelligence. While selective outcomes are measured, the complex structures and mechanisms, from which broader patterns of recognition emerge, stay opaque. This blackboxing of the institutional art world enables the Kunstkompass to concentrate exclusively on public outputs like exhibitions or reviews.
Connecting implicit assumptions about blackboxing and swarm intelligence makes not only the final ranking possible but also its constitutive metric. In order to commensurate exhibitions or reviews, the respective institutions need to be evaluated. These comparisons are based on the assumption that the art world itself is structured by symbolical hierarchies. ‘The biggest problem was finding a measure for weighting single museum, exhibitions and publication factors. These factors needed to be valued because they have different significance regarding contemporary artists’ fame’ (Bongard, 1974: 256, own translation). To measure institutional capacities to ascribe fame, the ranking relies on the assumption that art world experts themselves know best about their own symbolic topography. Therefore, Bongard ‘consulted experts to obtain an objective evaluation of institutions mentioned in the survey’ (KK 1970: 147). Once more, the Kunstkompass is not interested in the emergence or logics of symbolic differences and hierarchies of institutions. Once more, the ranking does not ask for in-depth qualitative assessments of individuals but relies on the experts’ swarm intelligence. Once more, this method aims at measuring only the output of this black box. Willi Bongard (1974) wrote that a standardized questionnaire had been sent to 106 art world experts to ask them about the most important institutions in the world. These experts were exclusively located in Germany and Switzerland because these regions were considered decisively neutral and open for international developments. Based on 57 answers, Bongard compiled a quantified institutional ranking of the most important art institutions including a numerical rating score for each one of them. Since the 1970s, there has been no disclosure about the further judgment of institutional fame. Some issues included the institutional list but did not clarify how the rating scores were produced (Rohr-Bongard, 2001). This is important because this mostly invisible institutional ranking is the common metric for all following lists comparing artists.
V Implications for Research on Rankings and the Sociology of the Arts
A certain paradigmatic worldview makes a ranking plausible at all and shapes its concrete production. Assumptions about the ranked field not only generate the main problem, to which an observational tool claims to answer, but also imply certain solutions. In this way, the existence and form of an observational regime can be traced back to underlying assumptions about a world-to-be-ranked. In this final section, I outline resulting implications for both the sociology of rankings and the sociology of the arts.
Worldviews and Sociology of Rankings
My analysis of the Kunstkompass shows that construction of the art world through rankings is based on highly specific premises towards chaos and order in the world. This approach needs to be tested for other fields and other rankings. This research would contribute to a more comprehensive sociology of rankings, which has been focusing on effects rather than the production of rankings and their underlying epistemology.
Furthermore, disputes about rankings can be analyzed from a comparative perspective on multiple epistemologies. If rankings are contested in a particular field, future studies can reconstruct the worldviews of those who oppose a ranking. If rankings are criticized for being incorrect, inappropriate or absurd, investigations of other modes of understanding the world and producing order could show what exactly is the problem. This perspective not only explains the vast resistance to rankings but also clarifies specific aspects of critiques. Probably, certain methodological steps are problematic for some agents. Possibly, different agents are not even talking about the same world.
Investigating a dynamic multiplicity of (competing) worldviews in a given field will inform studies on shifting value systems, related organizational reactivity, competition, and structural shifts of such fields. If rankings today play a crucial role in influencing agents, choices and whole spheres through public observations, there need to be other kinds of observational regimes that function (or had been functioning) equivalently for orientation.
Worldviews and Sociology of the Arts
My findings on art rankings hint at a new departure in the sociology of the arts that concern its field-defining question: How is art possible? While some recent approaches suggest getting closer to the art work itself, its materiality and its aesthetics (de la Fuente 2007, 2010; Alexander and Bowler, 2018), an investigation of multiple worldviews in the arts aims at theoretical innovations regarding autonomous symbolic logics and criteria that structure autonomous art.
First, certain pre-suppositions towards genuinely artistic criteria, respective values and autonomous logics can be reconsidered for sites of cultural production (Sassatelli, 2016). Numerous studies on organizations research artistic logics under pressure by political, economic or religious interests. Cultural organizations are described as ‘pluralistic institutions’ (Zolberg, 1981: 119) that are confronted with non-artistic expectations about the production and display of art (e.g. Alexander, 1996, 2018). Situated in a professional and organizational field (DiMaggio, 1991), museums, for example, deal with ‘tensions of mission’ (Zolberg, 1986) between artistic criteria on the one side and educational, political, and economic goals on the other. However, many research designs referring more or less to concepts of commodification and political instrumentalization of culture (Gray, 2007, 2008) imply a pre-supposed artistic logic, which is then colonized by external forces like power and money. If art museums ‘serve many masters’ (Hooper-Greenhill, 1992: 1), an institution’s specific perspective on art as one master needs to be understood first. Instead of taking autonomy of the arts and a monolithic aesthetic logic for granted, further research will reconstruct what is actually supposed to be under threat by hostile worlds. What is art actually in the perspective of a museum or a biennale? What artistic criteria are actually under pressure and why do they seem to make sense for art world professionals?
Second, there are related questions on autonomous art’s value regimes. Here, Bourdieu’s theory on reversed economies of art and commerce casts a long shadow. Even if empirical changes have taken place since the mid-19th century (e.g. Beckert and Rössel, 2004; Crane, 2009; Graw, 2010; Zahner, 2006; Wuggenig and Rudolph, 2013), studies asking for such shifts start with a pre-supposed differentiation between cultural and economic values, and only then argue for opposition, congruence or blending. Researching multiple worldviews could show how symbolic and artistic value, and its respective economies, is actually perceived inside the arts from various standpoints. If rankings are ‘quasi formal institutions’ (Buchholz, 2016: 47) reinforcing certain modes of evaluations on a global level, they equally hint at a plurality of worldviews that not only linger in different geographical contexts but also in different field positions. If symbolic value based on recognized recognition is at stake, the shape and shift of this mutual recognition needs to manifest in different worldviews.
Third, the Kunstkompass hints at epistemological shifts within the art world about the relation of quality and symbolic order. Assuming quality is still relevant, art world professionals do observe secondary structures like reputation and resonance to deal with radical abundance and to assess the relevance of art. The Kunstkompass describes artistic production but actually only observes art institutions and professionals evaluating this production. This observation of a proxy is based on an evaluation of the art industry’s own symbolic order. Intriguingly, the basic metric (institutional lists) relies on art world professionals’ insights as well. Hence, I assume that such orientation on proxies is common also for other worldviews in the arts. This would mean that self-observation of the art world has taken a soft social turn in an institutionalist direction (Harris, 2010). However, the turn is only a soft one because observations of the social production of value through institutions are still essentially linked to the quality of art works.
Such a quasi-sociologization of understanding art can be conceptualized in systems theoretical terms more generally. For Luhmann (2000: 20), an operational distinction between art works (‘communication through art’) and discourse (‘communication about art’) is integral for art as a function system. Art critics, art historians and curators do not produce art works but describe and evaluate their totality after it has emerged. These ‘self-descriptions’ (Luhmann, 2000: 244ff.) work with semantics correlating with structure (Luhmann, 1980) and ‘theories-of-the-system-in-the-system’ (Luhmann, 1981: 199, own translation), helping to identify references and novelty among art works via sameness and difference (Luhmann, 2000: 244ff., 1990). The Kunstkompass picks up established semantics like artist, style or quality but turns them from tools for ordering to indicators for chaos. To counter a rampant complexity, the ranking introduces fame as a new way for ordering the totality of the arts. In doing so, the Kunstkompass does not immediately describe art works (like established self-descriptions) but investigates a symbolic order of those institutions and professionals that produce self-descriptions.
Observing proxies like fame enabled by such second order self-descriptions of resonance structures is also possible elsewhere in the arts. Moreover, rankings in modern society commonly observe highly differentiated symbolic spheres (e.g. reputation in science) and it is there where they trigger effects. How do different spheres deal with the relation between their particular form of quality and possible proxies? If a ranking should not represent a perceived order, there needs to be another worldview. Confronted with rankings and apparent contingency, all worldviews are then open to both debate in the field and sociological investigation.
Footnotes
Acknowledgements
I am grateful to many colleagues for comments on my research. In particular, I want to thank Cornelia Bohn, Andrea Glauser and Bettina Heintz for constant feedback. This paper benefited considerably from remarks by anonymous reviewers and the editors. The findings are based on research supported by University of Lucerne, SNF-NCCR Iconic criticism (eikones), DFG-Research Training Group Automatisms at Paderborn University, and Bielefeld University.
