Museum Photo Archives and the History of the Art Market: A Digital Approach

Abstract

Digital images with metadata contain unique potential for research into the history of the art market. The embedding of digital images in a database allows for the possibility of an association with their historical context due to the presence of metadata, which includes economic data, such as the provenance chain, as well as information about collecting practices. The database becomes a historical reconstruction of context accompanying the reproductions of the works. In this paper, a case study of a museum photo archive of forgeries illustrates the ways in which digital methods can be helpful in analyzing these contexts. The archive was run by the secret “Verband von Museums-Beamten zur Abwehr von Fälschungen und unlauterem Geschäftsgebahren” (Association of Museum Officials for Defense against Fakes and Improper Business Practices). This archive allows the engagement of early 20th century museums in the art market to be traced within specific genres. The goal of the case study and methodology presented here is to learn more about the economic practices of museums. Specifically, this paper reconsiders a study by Timothy Wilson on fake maiolica, with a new focus on the involvement of museums in the art market.

1. Introduction

The goal of the case study and methodology presented here is to learn more about the economic practices of museums. For this inquiry, we will look at a specific photographic corpus produced in museums, the archive of the International Association of Museum Officials for Defense against Forgeries and Improper Business Practices, which dates from 1898 to 1939 (the official title is in German: Verband von Museums-Beamten zur Abwehr von Fälschungen und unlauterem Geschäftsgebahren). The archive depicts fake objects and was collected and circulated by museum officials in the first half of the 20th century. This was a collaborative archive in which museums kept track of the art market. The relation between museums and the art market was often one of conflict, which becomes particularly evident with problematic objects such as forgeries.

In this paper, we will focus on a digitized version of this archive and discuss methods to analyze such a corpus with questions related to the economy of art in mind. The digital format is key here. Images in a database allow for the possibility of an association with their historical context due to the presence of metadata, which includes economic data, such as the provenance chain, as well as information retained from the collection context. The database becomes a historical reconstruction of context accompanying the reproductions of the works. Furthermore, research information can also be merged with the images. Both ways of adding information to the bare images make them “digital objects” (Hui 2012), with layers of information.

With the specific image corpus of illegitimate objects, this project takes a side route in the digital research of museum collections (Geismar 2018; Kohle 2017) and investigates the gaps in the collection. It is these gaps and negations of the canonized display that the archive of the Association of 1898 represents with its focus on forgeries and fraud. With this hidden counterpart to the history of display in European museums in the early 20th century, a new contribution to museum history becomes possible, which highlights the work of museum officials behind the scenes. The intangible quality of the objects pictured in the archive—most of them are discarded and lost—particularly qualifies them for digital approaches. Employing such an approach is further strengthened by the fact that this physical archive of photographs is lost as well (Mundt 2018, p. 305). There are several sources, however, which allow for the reconstruction of sections of the archive. Traces of it can be found in the internal printed materials of the Association, which are kept in archives and have been digitized (see Section 3 and Supplementary Materials). Copies of some of the photographs can be found in museum archives, for example, in the Museum für Kunst und Gewerbe Hamburg.

This paper is an investigation of the economics of art—one that is focused on institutions, power structures, and how these operate as a system that shapes the economic form of artwork. The internal economic practices of museum officials—e.g., how problematic objects such as forgeries were handled and how museums reacted to and managed the “status change” (Dolezalek 2015) of objects—is directly connected to questions involving the external market—e.g., how objects were prepared, valued, and circulated. All in all, the fight against forgeries forms a fascinating documentation of the relation between museums and the art market in general.

2. Research Goal and Theoretic Framework: Economy of Art

In this paper, I am interested in exploring the economic practice of museums. Economic practice is understood here not only as evaluating art and buying it, but also how museum officials were reacting to the market and intervening in it through their practices. In the digital reconstruction of the lost image archive, the ties and interaction with the market can be traced. This is possible because in the digital space, photographs carry contextual information and this allows us to trace their economic function. We will return to this methodology in detail in the next section, after discussing the relevant economic theory.

Museums do not operate outside of the general economy of art; as institutions of the knowledge economy, they are an integral part of it. Our discussion of a theoretic framework for research into museums and the art market has to start on a broader scale. From an economic perspective, artwork extends beyond their artistic form, i.e., their style, to their economic form—how they are traded and how they are collected. In economic theory, works of art are often seen as singularities (Karpik 2010) and opposed to industrially produced commodities. Some artworks possess a high monetary value, and such pieces are often considered rare (Moulin 2011). A theory of regular dynamics of price building is ultimately in conflict with such specific objects that seemingly evade comparison with others. This is rooted in the production of art, which is different from industrial commodity production; that is, it is very often focused on producing singular objects outside the division of labor but carrying signs of single authorship documented in a signature (Heinich 2008). Consequently, their production is riskier, because the costs and returns are dependent on a high degree of speculation. Because production is risky, the cost calculation becomes speculative; the assessment of a piece’s valuation, which is formulated by artists or managers and marketeers of culture, rely on how to place the work in the larger context of art, hence the importance of the context of collection or corpus/oeuvre.

This also means that knowledge work is fundamental here, as it underpins the standards of the cultural logic of collecting and canonization (Graw 2009, p. 9). Works of art are, after all, not only valuable as rare luxury commodities, but also symbolic goods (Bourdieu 1993). An important factor associated with speculation includes the artist’s popularity, especially for western art beginning in the early modern period, when it was more consciously reproduced by the artists and their backers (Graw 2009). However, apart from this individual aspect of branding and fame, strategies for circulation play a critical role in the economic appreciation of art. Because production is risky, the cost calculation becomes speculative; the assessment of a piece’s valuation, which is formulated by artists or managers and marketeers of culture, rely on how to place the work in the larger context of art, hence the importance of the context of collection or corpus/oeuvre. Everything that is understood as art is ultimately compared with other works. Even more so, most art is characterized by a very strict systematization in a group of works, in a periodic or local context, through the authorship or workshop source or through the corpus of a private or public collection. This amounts to a powerful value system that enables appreciation and commercial estimations of artwork. Luc Boltanski and Arnaud Esquerre recently have called this the “collection form”, in their pragmatic sociology of the pricing of art, describing valuation as a complex communication process (Boltanski and Esquerre 2018, pp. 205, 493).

We can see the collection form as a key referent in the process of artification (Heinich and Shapiro 2012) that can be observed in the collective efforts of art producers, experts, and institutions to justify art.

A key example of the process of artification is when works that have not been produced for the art world, such as non-western cultural objects, are forcefully transformed into art collectibles. This is not an intentional production process but a modification in accumulation.

Other approaches focus solely on the artification process and interpret art as a mostly speculative asset whose value is primarily generated in circulation (Coslor 2010, p. 211) or, as Marina Vishmidt has pointed out, an asset for which production itself is speculative (Vishmidt 2018, p. 36). The collection context then becomes a further basis for the speculation with art. Production as well as accumulation is aimed at this collection form, be it in explicitly financial terms such as in auctions, or more cultural terms such as in museums (Shaked 2015).

Knowledge economies of expertise and canon-building go hand in hand with the speculative economies of price fixing. The speculative economy and knowledge economy, however, can also be problematic terms, because they carry the connotation that this economy is about risky games (speculation) versus expert discourses (knowledge). Instead, we can describe both as strategies that are important for the commerce of art. The question then is how both factors stabilize and destabilize this economy. Deceitful production adapts to the same full circle, or encompassing system, of symbolic goods that encapsulates speculative valuation and knowledge economy exactly by taking part in and destabilizing it. Thus, to research forgeries means to investigate the economy of art (Heinich 2009).

In this paper, I aim to apply these theories to the topic of archival photography in the economic activity of museums. The central question is what this image corpus reveals about economic practices in museums. The photographs are connected to this question most directly, as will be shown in a case study on maiolica (Section 4). The images in this collection did not only represent an economic function of valuation, but some of them were also directly recovered from the market, e.g., from auction catalogues, and repurposed for interventions into the market.

3. Archives of Photographs as a Source for Art Market Studies

Photographs provide traces of the steps of production, accumulation, manipulation, and displacement of art. As documents, they are a static means adapting to dynamic changes and speculations in art commerce. Therefore, they serve as an important tool in the market strategies of museums. The issue of photographs in the museum archival practice (Tschirner 2011) is embedded in larger issues of photography in the discipline of art history, because behind the practices of evaluation in art history lies an archival practice (Osborne 1999, p. 58). Images were produced and deposited in an ordered manner and reproductions were circulated and multiplied as part of a general urge to organize the “disparate” (Alphen 2008, p. 66).

The photographs in this case study were documentation pictures from the internal museum practice stemming from various sources. They carried additional layers of information as archival “photo-objects” containing visual, material, and textual information (Bärnighausen et al. 2018).

This quality can be transferred into the digital realm, through building a database, which keeps the associated information beyond the visual data. In our case of the lost archive, reproductions of many of the photographs (ca. 350 of the 1200 photographs of the timeframe from 1898 to 1927) could be extracted from internal printed materials of the Association, allowing for a reconstruction of parts of the archive. These materials have been scanned in the last years (see Supplementary Materials). Through automatic segmentation of the textual information, different parts of the encompassing information in the text around the images were saved in the database with them. In our case, this included title, description, author name, date, and archival number. As the Association produced registers to their materials (e.g. genre categorization), this dataset was enriched further, with a detailed categorization of each case. The genre register contains more than 300 detailed sub-categories, from leather book covers to Italian bronze aeolipiles (Register über das auf den Tagungen des Verbandes von Museumsbeamten von 1898 bis 1927 behandelte Material an Fälschungen 1929, pp. 3–13). All of this information reaches far beyond the practical question concerning the authenticity of artwork and thereby beyond the historic conclusion the museum officials drew regarding whether a work was original or fake. In fact, the archive documents ways of describing works of art, and by organizing the segmented information from these sources digitally, as tables in a relational database, we built a dataset on the networks and practices of museum officials engaging with the economy of art. Now, the database is able to combine various tables and request chains of information, e.g., which museum official was reporting on which forger and who was focusing on which artistic genre? As this database is a very rough prototype, it cannot be made available to the public yet.

We have to keep in mind that these are digital reproductions of photographs of objects, with both the original photographs and original objects now being lost. What remains are specific visual objects—images in their information context. Only when we deal with the images as “photo-objects” or “digital objects”, are we able to address our research questions pertaining to the economy of art, because now the logic of the archive can be explored through the image set as a relational dataset. For example, visual aspects of museum history, such as the history of display and the history of museum technology (Troelenberg and Savino 2017), become traceable not only in the photographs but in the relation of the photographs to the accompanying data.

In their original context (with accessibility limited to members), the images had a strictly practical use of strengthening knowledge about existing forgeries and new technologies of faking art. Furthermore, they were used for collecting visual evidence in a structured manner and were connected to written data, including reports and registers. For researchers today, the image archive as a whole, understood as a continuity of the visual practice of hundreds of museum workers for a historical investigation, offers more angles. Of interest is the practice and discourse of museum officials facing challenges from the art market, more specifically from art crime. This group of art historians and archaeologists not only reacted to the powerful structures they encountered, but with their archival practice and communication infrastructures, they built up a counterforce to the economic and deceptive strength of the illegitimate as well as the legitimate market. They were well aware of this antagonism. Already during the construction of art history as a scientific discipline in the 19th century, the fight against forgeries played a large role (Lenain 2011, p. 243). This shapes the pre-history of the Association. In its creation at the end of the 19th century, art historians took a step further in this conflict and organized themselves in a secret society to combat forgeries. As knowledge workers in the museum, they were directly in contact with the art trade. At the core of their efforts was an image archive and this image archive is, as I have argued, a most valuable source for inquiries into the economy of art. Beyond the practical question of the fight against forgeries, it contains evidence connected to various aspects of the art market, as can be illustrated through the following case study.

4. Case Study on Maiolica

In my case study, I draw on an existing study into a specific section of our archive by Timothy Wilson, an expert on a specific type of Renaissance-era earthenware called maiolica, characterized by its tin-glazed cover and decoration (Wilson 2011). This enables me to work with a reliable corpus that has already been evaluated, because I am not an expert on maiolica myself. I want to credit Wilson with providing an inspiring example of what experts can draw from this source for their specialized work, and I hope that his work and my reference to it will inspire further studies. Indeed, one goal of my digitization efforts with respect to this archive is not only to enable specialized inquiries into specific topics, e.g., groups of actors or objects, but also specific discourses and practices. However, with reference to an existing study, I also aim to show how fruitful it can be to investigate one special section of the archive from various angles and with focus on the relation with the art market. With my approach in particular, I want to show how future research into commercial aspects of visual material in museum archives can be conceptualized methodologically and orientated towards the theoretical concepts of art and value formulated above. For example, we find that the photographs in the archive were not only coming directly from commercial catalogues but were also used by the museum officials for interventions into the market. As my research (see also

Table 1. Overview of maiolica cases in the Association’s archive.

Year	Archive Number	Mitteilung/Verhandlung	Report Name	Content	Source of the Archival Photograph
1899	2	M: 2	Purgold	Book cover
1899		M: 3	Purgold	Palissy	Literature
1900		M: 31	Brinckmann	Casa Pirota
1900		M: 33	Falke	Zschille, Bardinin	Auction catalog
	105		Skinner	Della Robbia
1905	811 (later)	M: 154–155	Falke	Casa Pirota	Auction catalog
	211	V: 1898	Gröbbels	Della Robbia	Sketch
1909	403	M: 248	Brinckmann	Fayence plates
1910	410–412	V: 1909	Bode	Can and jug	New documentation
1910	417	V: 1909	Vasselot	Albarello	New documentation
1913	748–752	M: 411	Falke	Beckerath, Lepke	Auction catalog
1925	953–955	M: 515	Falke	17 albarelli, 4 plates	Existing documentation
1925		M: 523	Falke	Mengarino
1925	1051	M: 555	Cube	3 bowls
1927	1097–1103	M: 568	Schmidt	Karl Fischer	Dealer’s catalog
1930	1270–1281	M: 656	Rackham	Various collections
1938	1676a–b	M: 767	Rackham	Restoration

) will show, having a digital overview of these images, produced by the method described in the previous section, helps to shift the focus towards these topics.

The material of the Association has been mostly forgotten, which was intended by its actors, who took measures to prevent libraries from collecting its materials. Even after the material was made accessible by the Kunstbibliothek Berlin and the University Library Heidelberg in digital form (see also Supplementary Materials), almost no scholars referred to these sources. One exception is the study by Timothy Wilson, a renowned expert on maiolica, who has analyzed this material and completed a study on how the museum experts worked on fake maiolica. He organized his study chronologically and examined all the proceedings in order to trace the discussions the museum experts had at their regular conferences. In doing so, he followed the discussion of maiolica through the 40 years of existence of the Association. His primary focus concerns discussions of authenticity; however, I am more interested in what these materials reveal about the relation between museums and the art market.

As I argued above, it is through digitization and processing the images as a database that we can reconsider this set of photographs. Because we focus on the photograph archive at the core of the Association, the corpus appears differently than in the linear reconstruction by Wilson, who tells the story from conference to conference, from case to case, from 1899 to 1930. In an overview, supported by digital crossing of sources, we find, for example, that most of the visual archival material for maiolica was not produced by the museum directors but through the collection of documents from the art market. This is the starting point to our question: how was this archival strategy embedded in a larger interaction with the market and its media?

So, while Wilson tracked the forgers and middlemen who were discussed, such as Imbert and Mengarino, I am focusing on the auction and dealer’s catalogs that were investigated by the museum directors. For me, Wilson’s intense study of the discussions around some objects shows how the interest in one specific artistic medium and the question of authenticity conjures a lot of evidence related to how museum directors were monitoring the market. Tracing one genre across a corpus, we can examine how knowledge economies—the gathering, organization, management, and exchange of information—intervened more directly in the market through the collection of evidence. We can also investigate how the process of authenticating was not only a synchronic analysis of style and technological factors but essentially tracking objects and their provenance on the market for decades in a diachronic approach. The photographic reproductions of objects are sources across these time periods that were reproduced and archived for the purpose of keeping track of the originals. The photos document the visual appearance of the object, but they also function as a marker of time and space—when the object was where.

The collection value—the importance of which was described above—is actively deconstructed by these experts, who used the collection catalogs as a source for their archive of forgeries. An auction is the place where collections become highly visible; it is here that the experts approached the visual material and transferred it to their archive as a warning for their colleagues. As a result, the knowledge economy of the archive and the collection economy of the art market were in direct conflict, because the museum officials tactically chose to keep their archive secret in order to strengthen its efficiency for their own means of staying ahead of the forgers.

Following and adding to Wilson’s description of the corpus, we can create a summary of the material on maiolica in the Association’s documentation. The resulting table (Table 1) is different from the tables in the database. While the tables in the database are only used to organize the automatically segmented materials, this table contains basic information from the database and manually added information from the source texts.

The set of maiolica can be described as follows. It is what main museum officials critically had their eyes on in the early 20th century and what allowed them to keep track, down to the moment when collections and other sets, for example, one by workshop, came onto the market. In pointing out single fraudulent works, they were working on the canon and defending a genre. Their defense is not only an intellectual but also a practical one. For example, in 1900, the museum director Otto von Falke proudly reported that by spreading doubts on objects in advance, prices could be pushed down in the Zschille auction of 1899; he even states “in favor of the following auction by Bardini” (Mittheilungen des Museen-Verbandes als Manuscript für die Mitglieder gedruckt und ausgegeben August 1900 1900, Nr. 33). Wilhelm von Bode also describes market interventions in his presentation for the Prague conference (Verzeichnis der im Archiv des Museen-Verbandes bewahrter Abbildungen falscher Altsachen n.d., Nr. 411).

The two main persons reporting on fakes of maiolica were Otto von Falke, from 1899 to 1913 and again around 1924 before resigning in 1927, and Bernard Rackham, who joined in 1925 and reported in 1930 and 1937 (Wilson 2011, p. 6). Rackham was contributing material on fake maiolica in the collections of the Victoria and Albert, Gustave de Rothschild, the Ashmolean in Oxford, and Beit in London (Verzeichnis der im Archiv des Museen-Verbandes bewahrter Abbildungen falscher Altsachen n.d., Nr. 1270–1281). Other relevant members of the Association, for our purposes, were Bode and Jean-Joseph Marquet de Vasselot. Bode was also referenced by Falke as someone engaging in these debates (Mittheilungen des Museen-Verbandes als Manuscript für die Mitglieder gedruckt und ausgegeben August 1900 1900, Nr. 33), and he provided an illustration for the third photo volume of the Association in 1910 (Figure 1). Vasselot also provided an albarello depiction for the same volume. These direct contributions of single cases were the exception for maiolica. Arthur B. Skinner of the Victoria and Albert Museum contributed also, among a larger disclosure on terracotta (Verzeichnis der im Archiv des Museen-Verbandes bewahrter Abbildungen falscher Altsachen n.d., Nr. 101–107).

In large part, this corpus of maiolica in the Association’s archive is a critical archive of art auctions and price lists of workshops. Obviously, an archive dealing with forgeries is also implicitly involved with the art market, and in the archive, we see how explicitly it is doing so. As Table 1 shows, the archival documentation was drawn from various sources: exhibition catalogs, auction catalogs (e.g., Zschille, Bardini, Bourgeois), shop catalogues (e.g., the price list of Karl Fischer in Sulzbach and Georg Fischer in Bamberg—the successors of Fleischmann in Nürnberg), and individual buying offers and gifts to museums. In this sense, the maiolica set is quite representative of how the complete image archive was assembled. The Association’s procedure involved generating lists of what they considered to be inauthentic in auctions (Mittheilungen des Museen-Verbandes als Manuscript für die Mitglieder gedruckt und ausgegeben August 1900 1900, Nr. 33; Mittheilungen des Museen-Verbandes als Manuscript für die Mitglieder gedruckt und ausgegeben im Mai 1905 1906, Nr. 154–155). In the beginning, the Association did not bother with collecting the auction catalogs in their archive, as the archive register shows and only began to do so retroactively. For example, archive Nr. 811 is the auction catalog of the collection, Bourgeois in Cologne 1904, which the Association discussed in 1905; however, Falke only archived the catalog after World War I. This overview, drawn from the various registers and by crossing the processed sources, shows how the museum officials were using art market media as catalogs to build their evidence and also traces how their strategies shifted.

In general, they were using commercial images (Figure 2) along with photographs produced by their personnel on the spot (see Table 1, last column, for the different sources of the photo archive). So, critical objects from their own collections were mixed in with items that were offered to them in their day-to-day business. The metadata of these images points to two art market-related dynamics in museums: networking and building of counter-collections in the depot.

While the Association operated in clandestine—going as far as hiding their archive, the archive inventory, and the internal publications on cases, as well as deliberately evading attention for their conferences—their members used their knowledge to directly intervene in the market.

Late into the existence of the Association, there was even a shift towards more publicity in the form of an exhibition. At the Victoria and Albert Museum in London and in other museums, directors started to actively collect forgeries in the early 20th century. Collecting forgeries became a means of fighting forgeries; when they were taken off the market, they could then be examined. This process can be seen in line with the larger communication strategy that Falke spoke of in 1900 about influencing public opinion in order for prices to drop. These sometimes very secret and sometimes not so secret counter-collections are documented in this dataset, allowing a mostly forgotten part of museum history to become visible for us through the photo archive. The large corpus of the collaborative image archive serves to illustrate a strategy that individual museums applied towards the art market not only to discard problematic objects but also to archive them and to communicate this knowledge for the purpose of warning others.

Later, the forgeries collections moved into the exhibition space. A fascinating shift in exhibition practice took place in the early 20th century, which began on a small, sporadic scale but grew to be regularly held exhibitions of forgeries over the course of the century. The Association itself held such an exhibition in 1937 in Vienna. This exhibition included a fake maiolica plate depicting a boy dated to 1871 (Catalog 1937, Nr. 8). They also included a large set of photographs from the archive in the exhibition—a list of which is held in the Art Historical Museum in Vienna in the archival material on the Museum of Applied Arts. The exhibition featured prominently the work of Alceo Dossena, who had recently been uncovered as an influential forger of Renaissance sculpture.

We should interpret these exhibitions not as a newfound interest in the cultural history of forgery but as a drastic countermeasure of the museums against manipulations in the art market, that is, continuing the fight or the “Abwehr” from the Association’s name. This was one of the rare occasions where the secretive Association decided to go public with their research. Due to the outbreak of World War II, these efforts ended in 1939. Still, a way forward for the museums had been carved out, and from the 1950s on, museums held even larger exhibitions on the topic, most notably in the Palais Royal in Paris in 1955 in cooperation with Interpol and an exhibition in Berlin and Folkwangen in 1976–77 in the course of a large-scale research project, “Typologie der Fälschungen”, started in 1974 (Zacher 1976, p. 202). This research project also returned to the question of a central image archive on forgeries. Digitization of such networked approaches seems to be the reason that these efforts are mostly forgotten only decades later; similar to the earlier secret archive, now scholars focus on databases.

5. Conclusions

Art as a collective singular is very vague and complicated to grasp as an economic structure, as I have shown in a critique of theoretical approaches. An answer to this is to trace a solid set of objects, e.g., a collection context or another culturally specific group, and draw an account of its artification and valuation. Our case on the Association demonstrated that digital processing and analysis of such a source is very helpful in shifting the research perspective from the goal that the Association ultimately had—fighting forgeries and manipulation—to telling a more comprehensive story of museums and the art market. This bigger story will have to be told by manageable sub-sets of the data, not only to stay close to the conflicts between different institutions and actors of interest but also to illustrate complex economic dynamics. Taking the complete set of information from more than 2000 documentations in the archive would not be a corpus helpful for digital analyses, as the scope of the archive is very broad. A useful mapping of places, for example, is difficult to do, because the references can refer to a point of contact, a provenance, a source, a point of sale, an exhibition space, an institution, a workshop, etc. Moreover, prices would be a quite random statistical signal, as the objects differ so much, from delicate crafts items to massive statues. Instead, I can imagine the set of sources of the Association being a valuable site of cooperative work in the future, where interdisciplinary interest among scholars working on the art market and experts on specific genres from art history as well as restoration practice could converge in drawing material from the documentation and build their cases. I hope that my case study, with its economic perspective of a corpus already treated by an expert on the genre, is indicative here. This source on forgeries in the early 20th century provides information, such as names and events, as well as representations of objects, and is very direct in doing so, as this was an internal forum. However, as I have shown with the corpus of maiolica treated by Wilson, it also enables present-day outsiders to look at the interaction of museums and the market based on a dataset of internal information.