On the representation of places

The reasons for representing places are manifold, and so are the ways in which they are represented. Travel guides, music, and paintings give an impression of often distant places, and scientific texts seek to represent places and their context objectively. By going beyond the existing semantic discourse on representations and focussing on places more specifically, this article seeks a better understanding of the representations of places. First, naturally occurring representations are discussed as the result of the patterns that are formed by place-making and the very being of a place, such as archaeological finds and traditions. Subsequently, I argue that many of the natural as well as non-natural representations can be understood as ‘representation-as’ in Goodman’s and Elgin’s sense, because they represent places as part of a genre, for example, as a tourist attraction or a retreat. This understanding facilitates the conceptualization of individual representations and their placement in the overall context, which in turn helps laying the foundation for a conceptual framework for ‘platial’ information.


Introduction
Representations of places are ubiquitous, ranging from place names and touristic souvenirs to paintings and narratives, and they are as manifold in nature as places are themselves. 1 This article takes a closer look at what form such representations take. Besides discussing numerous examples, it focusses on two particular types of place representation: those that arise in the context of place-making; and more complex representations that give rise to genres of place representations. Existing classifications of representations, such as the classification into icon, index, and symbol proposed by Peirce (e.g., 1934, CP2.254), refer to representations in general. Here, however, we focus on the various forms of place representation, thus emphasizing the geographical rather than the semantic context. In this way, it is hoped, the gap between the geographical understanding of place and the understanding of representations will be further closed.
In contrast to place representation, both the concept of place and instances of places have been intensively discussed in the geographical discourse. Despite, or perhaps because of, the essential importance of place and place-making ('How places are made is at the core of human geography'; Tuan, 1991), different lenses exist through which place has been viewed. For example, Tuan (1977Tuan ( , 1979 describes place as part of human experience: humans perceive a variety of impressions through their sensory organs, which are reflected in emotions and beliefs, thus enabling us to develop a sense of place. Further descriptions refer, among others, to human movement patterns ('place ballet'; Seamon, 1979;Seamon & Nordin, 1980) and the way we ground experience in places (Malpas, 1999). These phenomenological approaches are opposed by social constructionist ones, which understand place as (to varying degrees) constructed by society (cf., Cresswell, 2004), as well as approaches in the context of non-representational theories (Thrift, 1999(Thrift, , 2008. Common to many of these views is that places are created through our interaction with what exists in space. Accordingly, places are often understood as non-static, and the process of place-making, whether in terms of the experience of the locale or individual and social interaction, is seen as playing a central role (Tuan, 1991(Tuan, , 1977. This article asks how we can reference place. The first part is dedicated to representations that naturally originate in the place-making and the living of a place, a type of representation that has been addressed in the geographical discourse only to a limited degree. After three motivating examples (Section Three motivating examples), it is discussed what constitutes such forms of 'natural' representation and what role intentionality plays in this (Section Natural representations), to then examine two examples in more detail: archaeological finds and traditions as natural representations of places (Section Two instances of natural representations). Starting from the observation that places are often represented as something, e.g., as a touristic or vibrant place in case of a touristic souvenir, the second part examines a certain form of representation called 'representation-as' in more detail. In particular, the corresponding genres established by this type of representation are addressed, and some of the natural and non-natural representations discussed in this article are classified into these genres (Section Representation-as and genres of place representations). Finally, the subjectivity and experienceability of place representations as well as their complexity, are examined (Sections Subjectivity and experienceability and Complexity and supervenience).

Three motivating examples
At first it seems obvious what a representation is. 2 An object (e.g., a photograph or film), called a representamen 3 in the context of this article, denotes a 2 This apparent obviousness should not obscure the fact that representations can be studied from different angles. For example, Lefebvre considers the production of social space as the result of the interplay of three 'moments', among them the 'representations of space' (originally aptly called 'l'espace conçu'; Lefebvre, 1974Lefebvre, , 1991 referring to 'sheets of paper, plans, elevations, sections, perspective views of façades', et cetera (Lefebvre, 1991, p. 361). Given Lefebvre's focus on space rather than place, it seems logical that his attention is not on the last mentioned objects and corresponding individual representations, but on the totality of these, i.e., on the 'global perception of space rather than representations of isolated spots' (Lefebvre, 1991, p. 153). Also, in contrast to this article, Lefebvre conducts his discourse in the context of ideological rhetoric. 3 The term 'representamen' was chosen with reference to Pierce (Harteshorne & Weiss, 1932, Section 228), for the frequently used term of a 'symbol' might suggest a non-physical Footnote 1 (continued) Artistic media such as various forms of text, paintings, music, photography, video, sound, and installation art convey narratives, which often refer to and characterize places (e.g., Aitken & Dixon, 2006;Al-Jumaili, 2020;Bailey, 2020;Barnes, 2019;Casey, 2002;Fowler & Helfield, 2006;Mocnik & Fairbairn, 2018). Besides such artistic form of representation, sound recordings, documentary photography and film, audio guides, itineraries, news paper articles, and many other forms of text often document places (e.g., Beckley et al., 2007;Harrison et al., 2021;Lombard, 2013;Bacqué et al., 2015, pp. 80ff). Art exhibitions and science museums can even set such media in a joint context to present a more complete view, and by doing so they become places on their own. Scientific studies and models lead to publications, which characterize places and provide insight into placemaking (e.g., Bergmann & Lally, 2021;Borsellino et al., 2021;Gustafson, 2001;Westerholt et al., 2020;Wunderlich, 2013), and brochures distributed by tourist offices and travel guides aim to market places (e.g., Lew, 1991;van Gorp, 2012). Even Volunteered Geographic Information and social media content provide references to places in various forms (e.g., Ballatore & De Sabbata, 2020;Lai et al., 2020;Mayer et al., 2020;Zeile & Ahmed, 2020). Besides this, also more physical forms of representation exist, such as souvenirs that help imagining a distant place (e.g., Bonarou, 2021;Hashimoto & Telfer, 2007;Peters, 2011;Woodward, 2014) and a stone brought along on the last holiday trip to help remembering a place visited (cf., Peters, 2011). particular place. The representation then refers to the fact that there is a link between the representamen and the place. It can thus be regarded as a triplet, consisting of the representamen, the place, and the link between the two. Such a link can be of different nature, including but not limited to a mental or sensory one, which adds to the complexity of the representation.
The way the link between the representamen and the place is established is the essential factor that determines the qualities of the representation. In many cases, the viewer cannot avoid drawing such a link and understanding it as a representation, because the representamen has the affordance (in the sense of Gibson, 1977) to cognitively suggest this link in the viewer: the representamen, in the context of the place and the person perceiving it, has the quality of enabling and facilitating the representation, the latter of which is understood here as the action of drawing this link. Such mode of establishing a link is common among photographies and films, but also for many other examples. If such an affordance does, however, not exist, a link can only be constructed intentionally. In both cases, the cognitive perception, previously acquired knowledge, and the general context play a mediating role in the process of linking the representamen to the place, which is often referred to as symbol grounding (Pavese, 2019;Scheider et al., 2009;Scheider, 2012). Potential representamens and links as well as mediating contexts vary greatly, which adds to the multifacetedness and complexity inherent in this link. In order to illustrate this and demonstrate the relevance and necessity for the discourse conducted in this article, three examples 4 will be discussed below: a documentary film exemplifying the ambiguity of such a link; paintings exemplifying the link's subjective nature; and touristic souvenirs exemplifying the link's limitedness in conveying qualities of a place.
As an example of the ambiguity inherent to many representations, consider the documentary 'Dont Look Back' by the American film maker Donn Alan Pennebaker about the singer-songwriter Bob Dylan. While this film, which forms part of the direct cinema genre, is usually considered a portrait of this singersongwriter (Beattie, 2016), the narrative of the footage shot during Dylan's 1965 concert tour in England also tells the story of the backstage with its recurring patterns of action and the complex identity that develops in this place. The conversations, the music, the behavioural patterns, and the interaction of those present convey an almost intimate atmosphere and a sense of what it would be like to be in this place ourselves (Beattie, 2016). However, whether we take the film as a representation of the person Bob Dylan or the backstage (or both) is to some extent at the discretion of the viewer. The film aims at a documentary function and in this sense seeks to provide an authentic insight. Only the selection of the clips presented and the chosen camera perspectives influence the viewer's impression and interpretation, and the way this film is interpreted as a representation is thus ambiguous.
Representations of places need not strive for intersubjectivity but can also incorporate subjectivity. The painting 'De sterrennacht' by Vincent van Gogh, for instance, conveys the impression of a starry night observed in Saint-Rémy-de-Prevence. Anyone who has observed the stars knows that this painting does not intersubjectively 5 represent reality in the way, e.g., astronomical photographs would be able to, but strongly incorporates van Gogh's sensations. It refers to how the painter perceived the environment outside the window of his lunatic asylum, in which he was at the time this painting was created. Also, many of the works by Paul Klee and August Macke are known for deliberately incorporating their idiosyncratic impressions of their journey to Tunisia. Although the places depicted in these works are unfamiliar to many, the image drawn by these works is very expressive 4 Throughout this article, reference will be made to these and further examples to illustrate the issues described. The numerous examples of place representations provided should also be understood in the context of the lack of systematic consideration of such examples in the existing literature. 5 The term 'intersubjective' must be understood here on the level of interpretation: certain kinds of photographs might be interpreted similarly by many, whereas van Gogh's interpretation of this particular painting is likely to differ from others. In general, too, the painting genre offers manifold possibilities to give space to such individual experiences, sensations, and interpretations.
Footnote 3 (continued) nature. Rather, a representamen is understood here as both the object and its interpretation. This motivation departs from Pierce's original motivation (cf., Hardwick, 1977, p. 193). specifically because of its interpretation of physical reality and inclusion of what was experienced, and because large parts of the paintings represent in an abstract and metaphorical 6 manner. What the aforementioned works have in common is that they only represent for the painters themselves an access to their own memories of the places they experienced, while the general public has to resort to the emotions and impressions evoked by the idiosyncratic interpretation of the reality depicted, in order to construct their own mental version of these places.
Representations are often limited in what qualities they can convey of a place. For instance, touristic souvenirs seem in principle much more suitable to create sensory impressions than place names due to their physical nature and the possibility to touch and use these in daily life. Yet, this very physical nature limits the possibilities to create diverse and more complex experiences. While the model of the Tower Bridge can indeed be perceived sensorially, this sensory impression does in many respects not correspond to that of the Tower Bridge itself, e.g., in terms of not reflecting the atmosphere, the sound and dynamics of the tourist activity, and the smell of the River Thames. An image of the distant place that the Tower Bridge represents as a tourist attraction is rather triggered first and foremost by the image that we have already received through other media or personal experience. In this respect and despite bearing at least a visual resemblance, the function of the souvenir is somehow similar to that of a place name. The latter may indicate particular qualities of the place, like the suffixes '-cester' or '-shire' do with respect to the historical nature of the place (cf., Ekwall, 1960;Jepson, 2011;Mills, 2011;Parsons, 2011) but offers otherwise a mere reference without revealing further qualities in more detail.

Natural representations
As we have seen before, many representamens are intentionally created to represent a certain place. A painting like 'De sterrennacht' and a souvenir like the model of the Tower Bridge are not the product of chance but the result of prior planning. The effort of this planning is in turn justified by the function that the representation takes on. The model of the Tower Bridge is meant to help tourists remember the place; scientific studies of places are meant to create and convey knowledge; and tourist brochures are meant to attract tourists. The stone brought back from a holiday is an exception, because it was not created for the purpose of remembering the holiday place, but even in this case the stone is taken home intentionally and can thus be understood in the context of 'home' as the result of an intentional action.
The definition of a representation chosen in this article not only allows for intentionally created representamens. Rather, it allows for any important aspect of a place to serve as a representamen. Consider a beach party as an example. The people attending the party dance, take cocktails at a purpose-built bar, and they queue at a toilet container. Although the movement patterns do not reflect the perhaps unique mood of that particular place, it stands to reason that the tracks in the sand will not resemble those of a day when holidaymakers build sandcastles, sunbathe, and go swimming. The patterns will also differ depending on the style of music performed and corresponding dancing styles. In contrast to an intentionally created map of the movement patterns, the traces in the sand naturally create an image of the place ballet without the need for intentional actions, even though it needs to be acknowledged that the representation is ephemeral in this case.
Representations of this second category are referred to as natural in the scope of this article, because they arise naturally in the process of placemaking and the living of the place and are not imposed from outside. Corresponding natural representamens lack intentionality in the sense of representation. They are part of the process that constitutes the place and therefore it does not even have to be recognized during this process that they can function as representamens. They are often unavoidable if the character of the place is not to be changed. All this distinguishes them from intentionally constructed representations.
Although natural representamens have a livedness in common, meaning that they are through place-making actively but in the context of the representation unintentionally produced by the people present, there are substantial differences in how such a representamen relates to the place itself. In the case of the beach party, the pattern of the place ballet in the sand is only a by-product to which not much attention is paid. It is not the traces but the presence of the beach as a whole that characterizes the party and therefore is eponymous. Guided tours, on the other hand, attempt to acquaint tourists with a place and can at the same time become, together with other tourist activities, important actors that shape the place (Koens et al., 2018), thus representing the place as a touristic one. The cityscapes of Venice and La Barceloneta in Barcelona, e.g., are strongly shaped by tourists, which has an impact on the places themselves and makes them known for this impact (Seraphin et al., 2018;Nofre et al., 2018). Similarly, elected politicians both represent and shape a constituency (Trasher et al., 2015;Heideman, 2020).
The non-intentionality of a natural representamen manifests itself only in the context of the representation. The notes on a community noticeboard or calendar of a coffee house, for instance, are intentionally created to represent offers, announcements, and events (cf., Broadway et al., 2018, pp. 11ff), but in their composition they also form a natural representation of the place. For the place of the coffee house is formed by its livedness including the community noticeboard, even though the notes on the board are not intentionally assembled to represent this place. Although and precisely because the representation is intentional in the context of offers, announcements, and events, non-intentionality is possible in the context of the place representation. That is, if the focus of the intention is on an aspect other than the representation, non-intentionality in relation to the latter is likely. The same applies to graffiti, which often represents the writer through 'tags' (Bloch, 2018), but which in its collectivity also becomes a natural representation of the place and interacts with it (Cresswell, 1992;Dovey et al., 2012).
A representamen is concomitant with its affordance to serve as a representation. If a representamen is non-intentional and arises as part of place-making, then it naturally lends itself to becoming a representation, but the latter is not yet necessarily realized. Rather, the affordance to give rise to a representation is inherent in a representamen and this, in the case of the non-intentionality of the representamen, without the intention of the people involved in the place. It requires the realization of this affordance by establishing an (in most cases) mental link between the representamen and the place. Neither are the patterns of a place ballet in the sand a representation of the beach party nor the community noticeboard of the coffee house, unless corresponding links are established and we interpret them as representations. The existence and non-intentionality of the representamen do yet not imply that the possible representation is also realized. In this sense, there are many natural representamen but far fewer (realized) natural representations.
Even the absence of an aspect that might be expected under certain circumstances can constitute a natural representation. Dialects, which can commonly be considered a representation of a region (Clark, 2019, pp. 1ff, 155ff), often borrow words from other dialects and languages, in particular if there was no previous use for the underlying concepts. Like in Proto-Slavic where there were initially no words for some tree species such as beech, European larch, noble fir, and yew tree, and a word for amber was also missing (Pronk-Tiethoff, 2013, p. 60). It was only in Russian that the word 'jantar' (янтарь; amber) was borrowed from the Lithuanian 'gintãras' and from there again into other Slavic languages (Pronk-Tiethoff, 2013, p. 60). This can be seen as an indication that the Proto-Slavs did not live in areas where these tree species and amber occurred naturally or were transported and traded, which makes it possible to narrow down the homeland of the Proto-Slavs (Pronk-Tiethoff, 2013, pp. 59ff).
Not least, places themselves can become representamens and render natural representations (cf., Warnaby & Medway, 2008). For instance, Salzburg as a frequently visited tourist city is often believed by tourists to be prototypically Austrian, and the image of Salzburg is in turn often shaped by the film 'The Sound of Music' (Graml, 2004;Luger, 1992;Reijnders et al., 2015;Im & Chon, 2008). In this way, the film serves as a representation of this city, and the city as a representation of Austria. The representational function must not, however, be confused with an authentic summary of the entire country (Graml, 2004;Luger, 1992). This is countered by the example of many capital cities, which differ from other cities and rural areas in terms of their political, economic, social, ethnic, and cultural function (Daum, 2005, pp. 1f, 12ff) and thus become a representation of the national state primarily through the intentionality of the constructed function. The use of places as a representamen can even be highly idiosyncratic. For example, my feeling of being in Bohuslän in Sweden is triggered by being on the shore of a certain lake in Herrestadsfjället near the city of Uddevalla. This way, the particular place on the shore of the lake represents Bohuslän to me, but many of my fellow human beings would not perceive this natural representation as such.
The demarcation of natural from non-natural representations shows similarities to Peirce's distinction of how a representamen can relate to an object, in this case a place. Peirce distinguishes between icons, which share qualities with the representamen; indices, which establish a relation by spatio-temporal contiguity; and symbols, which do not establish a relation in any of these ways 7 (e.g., Peirce, 1934, CP2.254). The model of the Tower Bridge can in a sense be understood as an icon, even if the precise mechanism of representation is more complex (see Section Representation-as and genres of place representations); the tracks in the sand after a beach party, by contrast, are an index; and a place name is a symbol according to Peirce's classification. Natural representations have an indexical character in many cases, as in the example of the beach party, because the representamen appears at the location of the place. In contrast to the notion of an index, however, the notion of natural representations deals with the process of place-making and refers to the livedness of the representamen. Not only a spatio-temporal contiguity is presumed, but also the lived participation in the place itself. The latter usually conditions the former, but this is not the case the other way round, e.g., when an indexical reference is created by the placement of labels in an archaeological excavation site: the labels are in spatial contiguity to the various places (like the fire place, or the cattle byre or pigsty) and are thus associated with them, but they do not form part of these (former) places. The same applies to a reporter being at the site and pointing towards a certain place either by a gesture or by words (deixis), without the gesture or the words actually forming part of the very place. Being in a location that relates to a place is not the same than participating in the place-making or contributing to the livedness of the place. In the following, two instances of natural representations are discussed in more detail to illustrate the diversity of these and the idea of livedness.

Two instances of natural representations
In addition to the examples already briefly touched upon, two further ones are examined in more detail in this section: archaeological finds and traditional music as natural representations of places. In doing so, different manifestations of these are discussed and numerous corresponding representamens listed to illustrate the scope of natural representations and to serve as a reference for further research. The reader may skip this section if he or she is more interested in theoretical considerations than in these two illustrative examples.
Archaeology mainly relies on natural representations of places (cf., Renfrew & Bahn, 2020, pp. 400f;Doneus, 2013, pp. 21f;Gamble, 1998). The aim thereby is to reconstruct how humans lived in the past based on pottery shards, utensils, human and animal bones, residues from fire, and other archaeological evidence (cf., Renfrew & Bahn, 2020, pp. 12ff, 188ff). For this purpose, a place is often at least partially excavated and the finds interpreted. The latter were not deposited for archaeologists but emerged as part of the lived place. In this respect, the finds can be understood as a natural representation of the past place: if, in the process of reconstructing the place, finds are perceived not only as possibly damaged parts of a physical environment, but in their totality as the key to understanding the past place, they become a representation of the latter. Besides physical finds, even the absence of an object or material can be a suitable indication. Vertical holes, for instance, may indicate posts that were formerly present (post hole; cf., Trachsel, 2008, p. 149), and a high redundancy (i.e., a multiple occurrence) of certain kinds of finds with simultaneous absence of other to-be-expected kinds can be an indication of a ritually used place (cf., Renfrew & Bahn, 2020, p. 413;Trachsel, 2008, p. 126).
Special attention is often paid to the arrangement of the finds as only this makes their contextualization and further interpretation possible (cf., Renfrew & Bahn, 2020, pp. 369ff;Trachsel, 2008, pp. 162ff, 175). In cases such as the sinking of a ship or a volcanic eruption, the physical aspects of an entire place are archived in their state of a single point in time, and the finds thus represent a snapshot of that place (time capsule; cf., Trachsel, 2008, p. 127). In other cases, the temporal dynamics of a place involving different cultural phases can be traced, making the finds a natural representation of the history of that place. The finds in earth layers of different ages reveal, e.g., the changing use of a rock shelter in Ifri Oudadane, Morocco (cf., Linstädter & Kehl, 2012). Considering finds in context can even reveal further aspects, such as the exchange of goods and information in and between different places (cf., Renfrew & Bahn, 2020, pp. 354ff) as well as the identity of a place (cf., Renfrew & Bahn, 2020, pp. 220ff, 400f, 546ff). When such tracing of context turns out to be difficult, experimental archaeology can re-enact the presumed scenarios and places to compare the findings derived from these to the finds found, i.e., the natural representation (cf., Renfrew & Bahn, 2020, p. 53;Trachsel, 2008, pp. 172ff).
Another important example of natural representation is tradition, which we subsequently examine using the example of traditional music in Sweden 8 (cf., Eriksson, 2017, pp. 11f, 20;Nilsson, 2016, p. 58;Nilsson, 2017, p. 23;Ahlbäck, 1995, p. 1). What qualifies tradition to be a meaningful natural representation of place is the fact that it is a naturally occurring form of habit, which, in case of traditional music, can be understood as an emergent interplay of different musicians and dancers in a platio-temporal context (cf., McDonald, 1996, p. 116). For it is only by building up a shared repertoire and corresponding ways of performing that these can be passed on in the long term. Traditional music must thus be understood as an information flow in place and time (Gustafsson, 2016, pp. 844, 923ff;Lissa, 1970, p. 37;Eriksson, 2017, pp. 193ff;Jersild & Ramsten, 2008;Mocnik, 2018). In particular, the transmission of repertoire and playing styles occurs primarily through recurrent practice as a form of unconscious learning and is thus primarily non-physical. An often misunderstood romanticized belief 9 is that the repertoire and playing style would be the result of the characteristics of the cognitively experienced landscape in which the music is performed (Gustafsson, 2016, p. 804;Eriksson, 2017, pp. 193ff) and that this would create a relationship to the place. Rather, repertoire and playing styles often disseminate widely (Gustafsson, 2016, 8 The geographical context of Sweden was chosen because it is well researched and some aspects apply prototypically. Yet, the considerations apply in similar ways also to traditional music in other areas. 9 In fact, this notion stems from national romanticism, as noted by several authors (Gustafsson, 2016, p. 804;Eriksson, 2017, pp. 193ff). Nevertheless, it should be acknowledged that physical geographies have an influence on the use of an area and thus also on the music created and performed, even if this influence seems limited. For instance, herding calls for cows and other livestock originally arose in grazing areas (cf., Ling, 1997, p. 23), while other songs synchronized work processes in the quarry (cf., Ling, 1997, p. 35). pp. 30ff, 121; Aksdal et al., 2005), and individual pieces can even be traced in large parts of Europe (e.g., Jersild & Ramsten, 2008). The representation of a place in the context of traditional music, however, arises from the fact that the living of tradition is identity forming (cf., Jucu, 2021) and therefore leads to the imprinting of an often local style (Aksdal et al., 2005). The latter is practised in certain places over a longer period of time in slightly changing forms and is accordingly associated with these places.
Traditional music shapes places and can even create them (cf., Eriksson, 2017), because traditional music is socially constructed. This becomes apparent, e.g., when different places are instantiated in the same location through the practice of different dance and music styles (Nilsson, 2016, p. 58). The practice creates spatial patterns that are planned and frequently repeated, thus constituting a prototypical example of a place ballet (cf., Nilsson, 2017, pp. 52ff;Gustafsson, 2016, pp. 263ff;Kaminsky, 2011). This can create embodied experience (Nilsson, 2016, pp. 29ff), form identity (Åkesson, 2013, p. 144;McDonald, 1996, p. 118;Aksdal et al., 2005), become part of mundane life (Åkesson, 2013, p. 143;Aksdal et al., 2005), establish a sense of place (Aksdal et al., 2005, p. 4), create memories (McDonald, 1996, pp. 122f), and it can be spatially limited to an area or locale (Lissa, 1970, p. 37). The great similarity of the descriptions given in cultural and music studies to those given in human geography is no coincidence, for these geographical places are shaped in a unique way by their musical and dance practices. Since this process of shaping is temporal in nature and actively lived, the ever-increasing mobility and the evolving methods of communication lead to a change in the long-established structures, thus creating ever new places of music-making (Aksdal et al., 2005, pp. 20f).
Another aspect of traditional music as a form of place representation is the diversity of possible representamens. While repetitive practice, embodied experience, and identity can serve as possible representamens, a reduction to individual, less holistic aspects is possible and useful, especially when intersubjectivity or even formal description is sought. For example and due to their strong regional variation, individual playing and dancing styles can serve as representamens, as can types of dance to a more limited extent. Even individual pieces of music can serve as representamens, even though the practice of the same piece of music in other places may limit its representational power. Sheet music and dance descriptions as abstractions of a piece of music and the corresponding dance, respectively, aim to be intersubjective to a high degree through their formal nature. Yet, the intersubjectivity of sheet music and dance descriptions stands in contrast to the subjectivity of their interpretation as music (Nilsson, 2016, pp. 29ff), the latter adding to the expressivity of the representation. To mitigate the extensive absence of instructions on playing style and of embodied experience, one might though resort to collections of sheet music instead of the sheet music of individual pieces. 10 The 'platial' dimension of traditional music is not only expressed in the experience itself but also reflected in the naming of dance types and pieces of music (Eriksson, 2017, p. 20;Nilsson, 2017, pp. 90ff, 94f;Gustafsson, 2016, pp. 261f). The names of dance types, e.g., often contain reference to places, although in many cases the names were introduced by musicians and dancers outside the place itself (Rehnberg, 1966, pp. 17f). Among the examples are references to European countries such as Poland (Polska, Polkett, Polonäs, Pollonesse, Polonoise, Pols), England (Engelska, Anglais), Scotland (Schottis), Norway (Norske), and Germany (Tyska polskan); to Swedish and Norwegian landscapes (Bleking, Finnskogspols, Halling, Rørospols); and to towns and areas in Germany (Hamburska, Hamburgare, Lybeckare, Rhenländare, Rheinländare) and in Poland (Mazurka). Pieces of music in Sweden and neighbouring countries are, in addition to the corresponding dance type, mostly referenced with the originator or the oldest known transmitter, including his or her home place (Eriksson, 2017, p. 20), such as 'Polska av Hugo Falk, Varnhem, Västergötland' or 'Bleking efter Carl Aron Hakberg, Skövde, Västergötland'. This kind of reference is ubiquitous and even serves as a spatial index. The importance of localness in the process of composing and music-making, as well as of practising music with the aim of representing a memory or place, is illustrated, e.g., by the compact disk 'Places' by Lena Jonsson (2018). As can be seen by these examples, references to the platial dimension form part of lived tradition, just as lived tradition can represent places.

Representation-as and genres of place representation
In the introduction, we have already touched upon the complexity inherent to the representation of places. In this section, we will shed light on this complexity by building on previous works of Goodman (1976,1984) and Elgin (1983Elgin ( , 1993Elgin ( , 2004Elgin ( , 2009Elgin ( , 2010Elgin ( , 2017, which we refer to as 'GE' in the following. An overview of these works can be found in an article by Frigg and Nguyen (2017).
In order to become a representation of a place P -or, in short, a P-representation 11 -a representamen R must be linked to this very place P. This link between the representamen and the place is not established in general by R mimicing P, but manifests itself in R becoming a reference to P. The model of the Tower Bridge, e.g., does not attempt to look like London as a whole. Rather, the model makes a reference to the Tower Bridge itself and stands therefore symbolically for London. As can be seen from this example, representations, unlike relations of similarity, are asymmetrical in general (Elgin, 2010, p. 4). There are two primary modes to establish such a reference, according to GE: denotation and exemplification.
Denotation is a basic mode of reference. The proper noun 'Tower Bridge', for instance, refers to the actual Tower Bridge not because it would resemble it, but by pure denotation. For it is only through stipulation in the context of prior knowledge that we understand the proper noun as a representation; no similarity can be discerned between the noun and the bridge. Something similar is true of the fork, which is arranged on the table with other cutlery to explain where the Tower Bridge, the Tower of London, and the Tate Modern are located relative to each other, because the fork also bears no resemblance to the Tower Bridge. However, as Elgin notes, the representamen (referred to by her as a 'symbol') is not arbitrary: 'To be a representation, a symbol need [sic] not itself denote, but it needs to be the sort of symbol that denotes.' (Elgin, 2010, p. 2) The latter must again be understood in context: the fork only constitutes a representation by the fact that we intentionally and explicitly assign its representational function by stipulation, and by being in some arrangement at all that affords the fork to be assigned this function.
Many representamen, unlike the latter examples, share qualities with the place they represent, thus rendering exemplification as a more complex mode of reference possible. For instance, the model of the Tower Bridge has the same shape and colour yet not the same size and weight. Accordingly, the model exemplifies some of the qualities of the Tower Bridge while not reflecting others. Even in this case of denotation, however, the mere coincidence of qualities is not sufficient. Being a bridge or being built at the end of the 19th century does not imply being a representation of the Tower Bridge. Accordingly, not every possible representamen is an effective example (Elgin, 2010, p. 6).
Another mode of reference, which is particularly effective with respect to place, is referred to as 'representation-as' by GE (e.g., Elgin, 1983, pp. 141f). It can be illustrated by the example of a model of the Tower Bridge, which is sold as a souvenir to represent London as a tourist city. The model of the Tower Bridge exemplifies the real Tower Bridge, because it has the same shape. The model even reveals the fact that the Tower Bridge is a tourist attraction, because it might be painted in similar yet more vibrant colours than the real bridge and serve as a key chain. At the same time, the model is also understood as a representation of London, although the two have no similarities. Rather, the model denotes London because we are used to the representation of a city through its tourist attractions. The fact that the model exemplifies the Tower Bridge as a tourist attraction here implies that the latter quality is ascribed to the city too. That is, the model ultimately represents London as a tourist city. In more general terms, this situation can be described as follows, according to GE 12 : A 11 GE distinguish between P-representations and representations of P (Elgin, 2009, p. 78;Elgin, 2010, p. 3;Goodman, 1976, pp. 31ff). The difference is that the existence of the place is assumed in the latter case but not in the former. Among such examples where the place does not exist are places idealized or even constructed through media (Avraham, 2000;Avraham & Ketter, 2008;Gold & Ward, 1994;Morgan et al., 2011) and imaginary places, such as utopian and imagined cities (Harvey, 2000, pp. 168ff;Erskine, 1960), places in dreams (Iosifescu Enescu & Hurni, 2018;Iosifescu Enescu et al., 2020), and places in virtual reality (Relph, 2007).
representamen R represents a place P as G if and only if (i) R denotes P and (ii) R represents some place Q thereby exemplifying some quality G of Q that (iii) is imputed onto P. In the example given, R is the model of the Tower Bridge; P, London; Q, the Tower Bridge; and G, the quality of a place being touristic. It should be noted that representation-as presumes the existence of some place Q that can be represented by the representamen R such that this representation exemplifies the quality G, despite the notion 'R represents P as G' not naming any specific place Q.
Representation-as is a widely used mode of reference in the context of places. The painting 'De sterrennacht' represents Saint-Rémy-de-Prevence as a peaceful retreat; the megaliths and their arrangement represent both Stonehenge and the surrounding Salisbury Plain as mystical places; and the tracks in the sand represent the beach as that of a lively beach party. What a place is represented as is determined neither by the representamen nor the place itself but can vary depending on context, including the person experiencing the representamen. For the waiters, as opposed to the visitors, the traces in the sand may represent the working place of a crowded beach party where they had their hands full. While the souvenir of the Tower Bridge represents London as a tourist destination for the holidaymaker, the same souvenir represents London as an overcrowded city with too many tourists for the resident. What some place is represented as is though not arbitrary. For example, the megaliths in Stonehenge cannot represent Salisbury Plain as a party place, and the model of the Tower Bridge cannot represent London as a financial centre.
Representations-as constitute genres 13 . Various souvenirs, such as models of the Tower Bridge, the Eiffel Tower, and the NYC skyline, represent London, Paris, and NYC, respectively, as tourist destinations and are therefore part of the genre that can be called 'representation-as a tourist destination'. The representations of this genre have much in common, as can easily be noticed when entering a tourist shop: they are in many cases colourful, often contain symbols such as flags or resemble a touristic part of the place in a stylized way, and they are often offered cheaply. In fact, each characteristic we attribute to a place constitutes such a genre, many of which feel natural and thus are popular. Among these genres are representations-as your home; a cosy, intimidate, vibrant, safe, peaceful, dangerous, ritual, sacred, or holy place; a place of retreat or of change; a mundane, anonymous, or emotional place; a trading, touristic, artistic, crowded, working, abandoned, or deserted place; a place rich in tradition, being typical Dutch, or representing Sweden; a place to grab a coffee or invite friends to; or a place where you are dancing in a certain way or where you can do sports. Most of these genres have in common that they are about the way we perceive and use places, because places are socially or individually constructed. Representationsas direct our focus on these characteristics.
The notion of genres defined by representations-as is different from that of the type of a place. First, a place can be of a certain type defined by some quality G without being represented-as or even representableas. For instance, to claim that the Tower Bridge is a touristic place, there is no need to discuss a representation of this place. Being a touristic place is rather a quality of the place itself. Secondly, a representationas of some place might not be of the genre G even though the place has this quality G. Such as London being represented as a touristic place by a touristic key chain model of the Tower Bridge. This representation is not of the 'cosmopolitan' genre, even though London is cosmopolitan. And thirdly, a place P can even be represented as G even if G does not apply to P in its entirety. Like the megaliths and their arrangement in Stonehenge represent the surrounding Salisbury Plain as a mystical place despite the chalk plateau not being known for being mystical in general.
In the special case of the places P and Q coinciding, a representation-as G can only exist if P also has the quality G. Conversely, however, it does not follow that every representation-as also represents as G. Although the definition in this special case seems much simpler and partly redundant (R is a representation of P exemplifying some quality G), this case is of practical relevance, as in the example of the Tower bridge: a key chain model of the Tower Bridge represents the latter as touristic (but not necessarily historical), while a more detailed and realistic model of the Tower Bridge in the 19th century represents it as historical but not necessarily touristic.

Subjectivity and experienceability
Place-making and the living of a place can be understood as the result of individual action and experience, which are contextualized by social views and patterns of behaviour (Gibson, 1978;Entrikin, 1991, pp. 6ff;Löw, 2016, pp. 170, 174;cf., Tuan, 1977, pp. 120ff;Ley, 1977). Just as places can be highly individual, so can their representation. The example of a stone brought along on the last holiday trip illustrates this, because it would not represent your memory of the visited place for many, if at all.
Places are sometimes portrayed in a 'distorted' way, and they can create bias. For instance, the news coverage of violence, crime, and social problems often forms a negative image, while cities have an interest in improving the reputation of their places to attract tourism, investments, and business (Avraham, 2000;Avraham & Ketter, 2008;Gold & Ward, 1994;Morgan et al., 2011). Since the representation of a place should not be understood in terms of similarity, as discussed before, but is asymmetrically predisposed, distortion cannot be defined in the deviation of the representamen from the place -this would not even be possible in most cases. Nor can the mental image of this place in the context of the representation be compared with the place itself. Rather, the mental image of the representamen must be compared with the mental image of the experienced place. If both differ, for instance if a travel guide or a news paper article suggests a different image than the one experienced on site, then one might call this representation 'distorted', like when being idealized, stereotyped, or biased.
The contextualization through social views and social behavioural patterns suggests a social construction of places in many cases (Cresswell, 2004, pp. 29ff;Kyle & Chick, 2007;Mayer et al., 2020). Thereby, the mental images of individuals intertwine and converge to a certain degree in a complex social process. Due to individually shaped facets of these mental images, corresponding representations also differ in general, despite their possibly intersubjective or even identical representamen. If an intersubjective representation is aimed for, the representation must accordingly be 'distorted' in that it avoids subjective components. As a result, a kind of 'empty shell' is formed, which is intersubjective and needs to be enriched with personal experience. The description of a place in a travel guide, for instance, first of all conveys intersubjective facts: the place's location, history, affordances, visual appearance, et cetera. In addition to these facts and based on the previous experience of the reader, the description also suggests moods and a sense of place. For if we have already visited places to which a similar description would apply, we enrich the mental image of the place described with these moods and the sense of place of similar places that we have already experienced (cf., Mocnik, 2022).
The subjectivity of a representation is accompanied by the possibility of increased experienceability. The expressivity and subjectivity inherent in artistic media facilitates, e.g., both the mediation of subjective experience and of emotion. When looking at van Gogh's 'De sterrennacht' or Klee's and Macke's works of their journey to Tunisia, we therefore feel and react oftentimes in highly individual ways. This is in contrast to scientific descriptions in publications, which strive for intersubjectivity and only have a narrow scope for interpretation due to their theoryladenness; the description only becomes meaningful in the context of a suitable theoretical framework.
The broader understanding of representation as advocated in this article relates to embodied experiences, processes, and practices. Natural representamens are, in particular and as part of the place-making and living, per se of a processual nature. This is demonstrated by the example of folk music traditions, which refer to the social and embodied practices of performing and dancing and even evolve over time due to their long-term and processual nature (see Section Two instances of natural representations). Such understanding of representations bears similarities to non-representational theories 14 (cf., Thrift, 1996Thrift, , 2008 as it shifts the focus from formal representation to the process of embodied experience. The latter is hereby considered to play a pivotal role because it renders the relationship between the natural representamen and the place.
Representations-as can be seen as a fusion of abstract and experiential representations. If a representamen R represents a place P as G, then only an abstract connection between R and P is assumed: R denotes P. However, this situation is made more complex by the fact that representation-as additionally presupposes the existence of a place Q that is represented by R and thereby exemplifies a quality G of that place, because this quality can be experienced in many cases and offers scope for subjectivity. In the context of the Tower Bridge example, the visual and sensory perception of the key chain model may remind us of the hawker who offers these very key chain models for sale in front of the Tower Bridge. Each time we pick up our key, the visual and sensory cue triggers a subjective memory of the place. This experienceability and subjectivity related to the Tower Bridge is thus transferred to London, although we may not have experienced the latter place in detail and in its entirety.

Complexity and supervenience
Representations vary in complexity. Consider the shallow representation of a place by a dot on a map face, which contrasts with the representation by a painting or a documentary series, reflecting the wider social and political context (cf., Rose, 1994). How complex a representation needs to be depends on a variety of factors, in particular, its function (cf., Anholt, 2010, pp. 38ff).
There exist different ways to characterize the complexity of a representation, such as through personal experience. For example, a representation becomes more multifaceted when we have experienced a place, because this can change the way we link the representamen to the place itself. And the idealized and stereotypical image of a place we have after studying a travel guide becomes more diverse and enriched with embodied experience when we visit the place. If we pick up the travel guide again afterwards, the same text and the same pictures usually trigger a much more complex image than before.
Another possibility to characterize the complexity of a representation is the specificness of the representamen. Here, the focus is on the question of which representamen can be chosen for a given representation. If the choice of the stone brought back from the holiday was made practically by chance, any other stone could have served this purpose too. In this case, the representation is less specific and thus in a sense less complex than if the stone was carefully chosen by a fellow traveller and given as a memento. In the latter case, only this one stone reminds us of that place that is shaped by the fellow traveller. A comparison in terms of being less or more specific is, however, only possible for certain pairs of representations, namely only those the sets of possible representamens of which are in a mereological relation. In this sense, being more specific is only relative and applies only to certain pairs of representations.
It can be conjectured that the totality of a place's natural representations supervenes 15 on the qualities of that place. Indeed, because natural representamens are linked to the place by the place-making, two places never afford fully identical possibilities of natural representation unless they are indistinguishable by their qualities including their natural representamens. Interestingly, however, there is no argument for why the totality of a place's non-natural representations should also supervene on the qualities of that place. On the contrary, it might even prove helpful to establish the demarcation of natural to non-natural representations through supervenience, as can be illustrated by the example of a beach party. The patterns in the sand after two very similar parties lead to virtually identical natural representations, because these patterns are linked to the place by the placemaking and are thus not interpreted differently. This is in contrast to non-natural representations, which depend on the context in which they are conceived. Even though the parties taking place annually on the same beach may virtually be indistinguishable by their qualities, which terms they can be referred to depends on zeitgeist and how language evolves over time.

Conclusion
Representations of places are ubiquitous, and yet in many cases their importance is not noticed. Especially in the case of natural representations, the imprinted patterns are perceived as a means of reconstruction rather than as independent entities. One of the exceptions is branding and marketing with respect to tourism and entrepreneurship, where the image of a place is consciously worked with. As has been discussed, there are common conceptual qualities underlying the various representations despite these differences, suggesting that more attention should be paid to place representations also beyond natural ones. A more thorough discussion of which type of representation is suitable for which purpose could pave the way in this direction.
The genesis of a representation of a place can be very complex, because this process is often much more than just a formal statement that a representamen is supposed to represent the place. Instead, sensory and cognitive processes play a pivotal role, especially in the case of subjectively shaped and strongly experienceable representations. In addition, structural characteristics such as the arrangement of representamens are effective means to imply meaning when the places represented are in a similar arrangement. A profound understanding of these mechanisms and a greater appreciation of subjectivity is a possible key to more expressive representations. Future research might explore such prospects, in particular, in the context of non-representational theories.
So far, the genres of place representation have not been researched well. For instance, it is yet unclear which genres play an important role in our perception and communication. A study of the qualities of such genres could reveal new ways in which places can be effectively characterized, such as by a combination of different genres. The representation of your favourite coffee house either as a coffee house, as one of your favourite places, as a place of retreat, as a crowded place, as an emotional place, or as an old-fashioned place may be informative, but the combination of these representations will characterize this one coffee house much better than a single representation alone and therefore, in extreme cases, even apply to just this one coffee house.
The prospects of place representation in information science, especially of natural place representation, seem to be widely unexplored so far. In particular, the concepts of natural representation and of representation-as offer potential to explore the patterns and rules far beyond pure denotation. This article has primarily been concerned with the mechanisms of how a representamen can refer to a place through the attribution of meaning. Future research might in contrast engage with the temporal development and transformation of representations, the interaction of multiple representations, the limitedness of a representation in terms of the mediated characteristics, the experienceability of representations, and the affordance of a representation to facilitate action.
Acknowledgements I would like to express my sincere thanks to René Westerholt for the numerous and fruitful discussions of what makes place, to Johanna Schwehn for the discussions about archaeological representations of place, and to Cosima Werner and the anonymous reviewers for detailed comments on the manuscript.

Funding
This work was partly supported by the Heidelberg Academy of Sciences and Humanities, project Heterogeneity and convergence in shared data sources -the importance of cognitive coherence in collective decision making.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.