Flags, society and space: Towards a research agenda for vexillgeography

Correspondence Dominic Medway Email: d.medway@mmu.ac.uk This paper proposes that flags are an important but under‐theorised aspect of geographical inquiry and seeks to outline a new field of “vexillgeography,” which critically embraces the spatiality and performativity of flags. We start by outlining the potential of semiotics (and geosemiotics) to further our understanding of flags, focusing on iconic, indexical and symbolic modes of representation in relation to flag design and the marking of territories and boundaries. In so doing, we highlight the importance of flags in the ordering of both Cartesian and relational space. We then examine flags from a performativity perspective, emphasising their role as actants within wider assemblages, and the everyday affective and emotional responses to flying (and viewing) flags. We conclude by outlining an agenda for critical research into vexillgeographies, to explore the multiplicity of ways in which flags (or flag assemblages) can contribute to the construction of spatial meaning.

a symbol on a flag is not a precisely defined category; its boundaries may be indefinite, depending upon the context. (2007, pp. 206-207) Reflecting this, geographers have examined flags as a form of political iconography (Webster, 2011), often in the context of banal nationalism (Billig, 1995) and contestation over flag meanings and the identities they represent (Webster & Leib, 2001). Other research has referenced the role of flags in marking (disputed) territories and claiming sovereignty (Dodds, 2010;McConnell, 2017;Steinberg, 2010). Overall, however, human geographers have neglected the subject of flags. Our paper strives to reveal the potential for moving vexillology into more critical, socio-political spheres of geographical enquiry. Thus, we show how flags act as components within networks of "territorial organizational things" (Garfinkel, 1996, p. 15), and thereby contribute to a fluid denotation and codification of space. Further, we argue that flags act as performative spatial inscriptionsforming one of the many signs and objects "that shepherd our bodies in particular directions" and "provide cues for particular modes of consciousness" (Amin & Thrift, 2002, p. 91).
In examining the interplay between flags and space, we start from a representational perspective, focusing on their semiotic and geosemiotic potentialities. Our argument then addresses vexillological performativity, encompassing embodied and affective interplay with flags, and the spatial implications of this. We conclude by proposing a research agenda for the geographical study of flagsor "vexillgeography."

| SEMIOTICS AND GEOSEMIOTICS IN VEXILLOLOGICAL UNDERSTANDING
Drawing on Peirce's (1965) understanding of semiotic representation, we suggest that flags can exhibit one or more modes of relationship with the territories they signify. The first is iconic; when a flag graphically incorporates a resemblance to a physical aspect of a placee.g., the flag of Cyprus depicting the island's shape. Flags can also be indexical, representing a given place using insignias, with this connection being either observed or inferred. For example, many flags use stars to communicate different place attributes, including: political orientation (e.g., the North Korean flag featuring a communist star); administrative structure (e.g., the US flag where stars equate to states); and/or localised natural phenomena (e.g., the New Zealand and Australian flags incorporating the Southern Cross constellation). Colour combinations in flag designs also means their indexical power in representing place is open to everyday abstraction, as seen in the Swedish flag colours integral to IKEA's brand identity, a phenomenon that resonates with notions of banal nationalism (Billig, 1995). A third symbolic mode of relationship is where the link between sign and signified is arbitrary, but nevertheless agreed on or learned by actors. If the perceived connections between a flag and a place are devoid of iconic or indexical understanding by the viewer, then this symbolic mode forms a default. Thus, one might learn in school that a particular flag represents a country, but with no understanding of why.
These semiotic principles also have relevance for flags representing territories other than the nation, including: small regions (e.g., the flags of the 26 Swiss cantons), large supra-national areal jurisdictions (e.g., the flag of the European Union) and disputed-territory-spanning political areas, such as the Black Standard flag of the Islamic State of Iraq and Syria, or ISIS. Moreover, the ISIS example emphasises that connections between a flag as a sign, and a place as a signified and socially constructed and understood entity, need not be universally perceived or acknowledged. Thus, some people might feel that a national flag misrepresents "their country" by effectively writing them out of "the nation," particularly in places strongly divided by partisan lines of politics, culture and/or ethnicity. These situations can result in vexillological contestation, indicative of wider disputes within the complex interrelationships between society, politics and place. For example, the Spanish political left often adopts the flag of the Second Republic (1931)(1932)(1933)(1934)(1935)(1936)(1937)(1938)(1939), shunning the modern Spanish flag for its perceived Francoist and royalist connections (Blitzer, 2012). Vexillological contestation can even lead to a flag's indexical re-signification, enacted through a political process to create a flag with a more inclusive representation of a place, as evidenced in the 1994 redesign of South Africa's flag.
Geosemiotics (Scollon & Scollon, 2003) emphasises how signs can simultaneously convey meaning to space, but also take meaning from their emplacement. Drawing on this understanding, flags at border crossings are not simply signifying a nation state, but through their positioning are also assuming the role of boundary markers for its physical and jurisdictional extentreflecting the fact that "boundaries are a constitutive prerequisite of territory" (Brighenti, 2010, p. 60). By such means, flags are also a recognised way of marking territory captured in theatres of war. This is attested by a series of iconic images, from the US Marines planting the Stars and Stripes on Iwo Jima during the Second World War to the British Royal Marines (re)raising the Union Jack over Port Stanley in the 1982 Falklands War. Indeed, in the latter case a political narrative about which flag flew in Port Stanley was constructed by opposing sides to convey the legitimacy of their territorial claim to the islands (de Lama, 1982).
Thus, "planting" flags emerges as a metaphor for claiming territorial permanencyperhaps unsurprising considering alternative meanings of the word in terms of "putting down" roots. Nevertheless, when flags mark the territories of opposing forces in theatres of war, they can be regularly uprooted and transplanted with the ebb and flow of conflict. This notion of space on the move and "always under construction" (Massey, 2005, p. 9) is illustrated in news images of Islamic State convoys moving through disputed desert terrain with the Black Standard fluttering from pick-up trucks, conveying the fluid and mobile geographies of ISIS. Thus, we see how flags can mark the territory of relational as well as Cartesian space, linked primarily to a homophily (McPherson et al., 2001) of actors, and "constituted through [those actors'] interactions, from the immensity of the global to the intimately tiny" (Massey, 2005, p. 9). This resonates with Brighenti's idea that "[t] erritory is not an absolute concept, but is always relative to a sphere of application or a structural domain of practice" (2010, p. 61). Territory is, therefore, not only a material thing, but also the abstract construct of human and institutional relations, and thereby an active and dynamic endeavour (Deleuze & Guattari, 1987). Furthermore, territories of relational space can also occur outwith standard, areally defined contexts such as the nation state to connect geographically disconnected and/or diasporic communities drawn together through shared identity, practice or culture.
An additional consideration is how relational spaces can concentrate and dissipate in a very physical sense. This is apparent in the visible realisation of semi-permanent or temporary spatial clusters of affinitive human interaction on the ground (Palmer et al., 2017). These clusters form ideal sites for the vexillological marking of relational territory, evident in displays of the rainbow flag in and around "gay villages" (Binnie & Skeggs, 2004;Nash, 2013). Here, sufficient actors within the LGBTQ+ community are geographically proximate to appropriate a distinct urban quarter and mark it in vexillological terms. Such spatial appropriations can be more ephemeral, manifest in the rainbow flags flown in abundance, and incorporated into clothing and costumes, at Pride parades.
The rainbow flag brings to the fore other aspects of semiotic and geosemiotic representation. First, it demonstrates the multifaceted materialities of flag assemblages, which can transcend notions of a piece of fabric to assume other seemingly more banal, and mundane, reformations and reinterpretations of semiotic powere.g., flags on clothing, badges, tattoos, car stickers. Second, it shows the multiple ways in which the sign of a flag might contribute to the organisation of space. Thus, a rainbow flag present at a Pride parade or seen within a gay village is not simply acting as a boundary marker of "gay space" (Bell & Binnie, 2004). It also fulfils the role of boundary actant, reflecting and potentially reinforcing the divisions between the relational territory of a homophily of actors and those outside. Conversely, the same flag may act as a boundary object (Star & Griesemer, 1989), forming a semiotic bridge between the homophily of actors within a given locale and "others." "Others" might include actors geographically disconnected from the locale, but part of a wider LGBTQ+ community, as well as those from different social (and sexual) worlds. The boundary object concept suggests all these actors may be able to relate to each other through some degree of mutual understanding(s) about what the sign of the rainbow flag represents, reflecting Star and Griesemer's contention that "Boundary objects … are both plastic enough to adapt to local needs and the constraints of several parties employing them, yet robust enough to maintain a common identity across sites" (1989, p. 393).
Flags as boundary objects can therefore unite actors in support of, or opposition to, particular identity positions. One example would be gender, as witnessed in suffragette flags and banners made up of green, white and purple colours. Alternatively, flags can have a clear racial messageevident in vexillological representations and their related performance(s), which may be contested. These would include the Confederate flag, mentioned above, or the Pan-African/Black Liberation flag comprising a red, black and green tricolour (Olivia, 2015).

| FLAGS AND PERFORMATIVITY
Flags as inscription devices are often subject to accepted protocols of display and performance relating to how and when they should be hoisted, flown, lowered, draped or waved. These performances are usually central to ceremonies and rituals designed to promote a sense of belonging, e.g., to "the nation," as recalled in Barthes's vexillologically pertinent account of a Paris-Match cover showing a saluting soldier with his eyes uplifted, "probably fixed on a fold of the tricolor" (1957, p. 116)an example of a performative ceremony that valorises a flag's semiotic power. Repetition of such performances is often critical to their success as it helps to stabilise and reinforce the link between the flag as a sign and the physical and/ or relational territory it signifies. Such repeated and typically ritualised performances can also be nuanced to convey commonly understood emotional and political messages on the space surrounding them, such as a flag hoisted fully in celebration, at half-mast as a sign of mourning, or upside down as a sign of distress. This accords with Denis and Pontille's contention that "the repeated actions of … inscription devices have been described as articulated operations that progressively give scientific claims their substance" (2014, p. 406).
Flag ceremonies also include notions of ritualised performance in a more embodied sensee.g., saluting the flag, standing to attention or bowing the head when it is raised. These learned and near automatic movements may further stabilise the inscription of a flag as a territorial signifier, supporting affective and topophilic connections to a place. This accords with the concept of "embodied nationalism" involving "the quotidian affirmation of the national through momentary encounters of bodies and objects" (Militz & Schurr, 2016, p. 56). It also indicates flags have relevance in the "more-thanrepresentational" perspectives of human geography (Lorimer, 2005), which consider the emotional dimensions of human experiences along with their associated and interwoven spatialities.
Another aspect of flag performativity is where this takes place, which links back to ideas about geosemiotics, and introduces the notion of "hot" nationalism (Billig, 1995). For example, the purposeful act of flag flying in certain spatial contexts assumes an unavoidable performative dimension that can ramp up political heat. In Northern Ireland, the display of the Union Jack, even in the abstracted form of its red, white and blue colours painted on kerbstones to delineate sectarian territorial boundaries in urban space, may be construed by Northern Irish Nationalists as a direct Unionist challenge. Equally, flying the Irish tricolour can be perceived as a counter provocation. Such vexillological tensions can be exacerbated through juxtaposed performances of flag allegiances, when these opposed spaces of community adjoin (Bryan et al., 2010).
This reveals the complex nature of affective responses to the "how" and "where" of flag performances, which may, in turn, be shaped by the cultural and political identity of actors involved in the performance itself, as well as those viewing it. It identifies how a multiplicity of interpretations of a given flag can arise, revealing a potential for performative "slippage" (Butler, 1993) between sign and signified across, and even within, different time-space contexts. Recalling Denis and Pontille (2014), this allows us to move beyond ideas about how flags might stabilise representations of place through inscription, to consider how flag performativity can surface different meanings and feelings for the space(s) in which it occurs. Indeed, it is this slippage which allows flags such as the English flag of St George to exist as complex and multifaceted semiotic assemblages. On the one hand, therefore, this flag can be associated with an unpalatable form of English nationalism, through its symbolic appropriation and "toxification" by right-wing groups such as the English Defence League (Copping, 2012). Yet, it is also routinely hoisted on churches and other English public buildings with little apparent dispute. And, in the enduringly disappointing campaigns of England teams in major football championships, St George's flags are flown from many car windows for the (generally limited) duration of the team's participation. Some, however, have taken issue with such practices. Huw Thomas, Labour Party candidate for the Welsh constituency of Ceredigion in the 2015 UK general election, suggested that such use of flags is only for a "simpleton" or "casual racist," and that cars displaying them should be defaced (BBC, 2015).
Equally, the US flag can be performed and viewed in many different ways, and by many different actors, ranging from its presence as a badge on a US President's lapel, its appropriation in the tattoos of American gay men (Guenter, 2001) or its ritual burning as part of an anti-American protest in the Middle East. In the last case, the political heat of the flag performance is clearly linked to where it occurs, who is doing it and the in-the-moment emotional responses of the flag-burning audience. Furthermore, these emotional responses are likely to differ between those persons situated within the immediate environs of the event, and those who view it from further away through a multitude of media lenses. In an instant, therefore, a flag burning can give a very different meaning to spacefrom a nexus of celebration and/or protest for performers and nearby onlookers, to one of hostility and threat for others. For the latter, the perceived political provocation inherent in performative flag desecration means many nations ban the practice, at least in relation to their own flag (Welch, 2000).
Flag performances and associated emotional responses are further complicated by what can be termed the "intertextual sensibility" of individuals (and/or communities) as flag performers or "gazers." We define this as the ability of actors to appreciate and affectively respond to semiotic nuances arising from (re)combinations, subversions and (re)interpretations of aspects of flag design elements and their subsequent performance. This reflects the fact that where interactions between people and flags as symbols are concerned, "any encounter contains reference to past encounters" (Anderson, 2014, p. 82). An example is the display of a lookalike ISIS Black Standard at London Pride 2015, the key difference being that its apparent Arabic text was actually made up of the white silhouettes of dildo designs. Pastiching and subverting the design of the Black Standard in this manner, and placing it in a Pride parade, raised the "political heat." This was something fully intended by the flag designer/performer. He wished to make a statement that the ISIS flag had "become a potent symbol of brutality, fear and sexual oppression [and] wanted to try and stimulate a dialogue about the ridiculousness of this ideology" (Coombs, 2015, n.p.).
Mindful of the potential offence this might cause if misinterpreted, reaction was reportedly gauged from fellow Pride marchers before hitting the main crowds, but the response was "so good humored and understanding that any worries were dissipated in minutes" (Coombs, 2015, n.p.). As the parade continued, the flag bearer was approached by police officers. They inspected the flag and, "grinning," immediately acknowledged that it was depicting sex toys. However, it seemed that other spectators lacked such intertextual sensibility, as the police explained: "they were getting reports that a man was carrying an ISIS flag through the streets, a misconception that could easily put me in danger. They asked if I would keep the flag concealed. So I put it away" (Coombs, 2015, n.p.).
Reactions to the "fake" Black Standard were clearly mixed. Some appreciated it was a satirical simulacrum, while others, including news channel CNN, believed it was a "genuine" ISIS flag. This media magnification of the performance brought further immediate responses on Twitter. These ranged from anger that an apparent ISIS flag should be present at a Pride Parade: Such reactions indicate the contrasting emotional states and differing intensities of political heat emerging from a single flag performance. In one performative context, therefore, we see how a flag manages to render several readings, indicating the power of vexillological performances to engender a multi-layering of spatial production across both real and virtual worlds.
This spatial-temporal fluidity resonates with other performances of the actual ISIS Black Standard, which demonstrate how vexillgeographies can also involve transnational dimensions, evident when apparently random enclaves of vexillological space occur at considerable distance from their territorial heartland. For example, in 2015 a teenager from the north of England was convicted of plotting acts of jihadist terrorism, and part of the prosecution's admissible evidence was that he had a screensaver of the ISIS Black Standard on his smartphone (Gutteridge & Barnett, 2015). Similarly, in Germany in 2016, the Bavarian Interior Minister drew attention to the fact that an Afghan refugee involved in an axe attack had a "hand painted Isis flag" on his bedroom wall (Yeung, 2016, np). In both instances, vexillological performances in banal settings politically charged the space around the defendants. In this manner, ISIS flag images and depictions appear to be projected and subsequently decoded as an open declaration of the hypermobility of ISIS space, and the fact it can seemingly "metastasise" from the Middle East to other locationsgeographically disconnected from the imagined caliphate, but still very much connected in terms of relational space.

| CONCLUSION: A RESEARCH AGENDA FOR VEXILLGEOGRAPHY
Based on this analysis, we suggest the potential for a rich vein of research examining the interplay between flags, society and space, which we term "vexillgeography." First, future research could explore the multiple ways in which flags contribute to the construction of spatial meaning. Specific lines of investigation might develop the analysis presented here of flags as complex and polysemic texts, open to heterogeneous interpretation and understandings, and constant semiotic slippage (Butler, 1993). Furthermore, it would be instructive to explore how this symbolic mode of relationship may be moderated by different dimensions of a flag's performative context. This approach should shed light on the multitude of possible outcomes for how a flag can inscribe space, as well as providing potentially valuable and revealing insights into the complex socio-economic machinations that underpin this process.
Second, a deeper exploration of vexillgeography requires recognition and interpretation of human affective and emotional responses to flag performances. Although affect and emotion are difficult to disentangle, geographical inquiry has highlighted their importance in shaping social relations, and aligning "some subjects with some others and against some others" (Ahmed, 2004, p. 25). Similarly, we have evidenced some of the ways in which flag performances may provoke spontaneous feelings of pride, anger or even humour. Such volatility in people's responses to flags holds the potential to constantly (re)produce and (re)define the space around them. Accordingly, future research into vexillgeography would benefit from methodological approaches sensitive to "bare life" reactions to flag performances, incorporating the "fleeting space of the moment … utterly wrapped up with its context, and most especially the object world" (Thrift, 2004, p. 152). This suggests an imperative for ethnographic data capture and analysis, which could involve the researcher's observations and vignette-based examinations (Militz & Schurr, 2016) of actors' embodied and discursive interactions with flag-bearing events and occurrences, and consideration of how these both produce and reflect Cartesian and relational spatialities.
Third, future vexillgeographic enquiry should explore the significance of flags representing non-national spaces at subnational and supra-national levels. This can also inform current debates about the nature of scale. For example, some English counties have recently developed and flown their own flags to assert local identities (British County Flags, 2013). Flags representing supra-national organisations may also be politically charged, such as those of the EU, NATO or the Association of Southeast Asian Nations (and see also consideration of the ISIS and Rainbow flags above). Furthermore, other nonnational flagssuch as those of transnational corporations and businessestranscend scale in complex ways, signifying corporate power. In summary, more needs to be understood about how different flags relate to expressions of localism versus state power, nationalism versus supra-national power, or how corporate power might be "softly" asserted over states through flag deployment.
Fourth, although flags have long been recognised as powerful political symbols, we have illustrated how they are rarely deployed in isolation, but are instead actants within complex materialities, associated rituals, performances, discourses, representations, affects and emotions. In particular, there is a need to consider the role(s) of flags as an assemblage of these multiple elements through which space is (re)produced and maintained. Conceptualising flags as assemblages can also develop insight into how they may (re)produce scale in a territorial and relational sense (see Marston, 2000). It is time, therefore, for geographers to examine flags from a more critical perspective, considering their performance(s) as a form of non-linguistic perlocutionary act, which can both maintain and change spatial understanding, and ultimately the nature and feel of places as social and lived constructs.