The island as a political interstice

Islands are often defined with respect to their physicality, namely small pieces of land surrounded by water. One inherent assumption is that islands can be defined distinctly from other geographic entities, such as the often-presumed dichotomies of island-mainland and land-water. This vocabulary is imbued with political meaning, especially that the opposite of an island is apparently the “ main land ” . Island studies challenges these notions, reinterpreting them in contemporary political domains, with one framing being interstitiality. From this baseline and drawing on some political geography work, this paper argues that the interstitial island is principally a political construction. Islands are, or at minimum can be, multiple forms of interstices, but they are very much created as such — whether inadvertently, deliberately, or a combination — making the island a political interstice. This paper follows this line of reasoning by selecting two characteristics discussed in island studies and geography with respect to islandness: separation and connection. The result is to explore separation and connection as interstitial, demonstrating that politics infuses the discussions, conceptualisations, and practicalities of the interstitial island, although this situation is not necessarily detrimental. Philosophically and practically, many advantages result from constructing the island as a political interstice, suggesting that island interstitiality has far more political than physical value.

Rather than seeking to expand the physicality of the "island" definition, others pursue metaphorical and philosophical realms.Cities with transportation routes cut, perhaps by an earthquake, are "islanded", as in apparently having become islands.When the COVID-19 pandemic started in 2020 leading to lockdowns and border closures, countries and subnational jurisdictions became similarly islanded because physical movement between some locations was banned-a concept extended to households when social contact beyond one's "bubble" was outlawed in countries such as the UK.From Donne's (1624) "No man is an island" to Simon and Garfunkel's "I am a rock/I am an island … I touch no one and no one touches me" (Simon, 1965), individuals have been examined as islands, ascribing certain assumptions about the meaning of being an island, which is termed "islandness".Again, no endorsement is given to these conceptualisations, merely noting that they exist with some accepting and others refuting them.Potential challenges and further perspectives from political geography include from feminism (Jackman et al., 2020), relationality (Pugh, 2016), sovereignty (Dodds, 2012), and tourism (Graci & Dodds, 2010).
All these definitional machinations are not merely intellectual exercises.They can be steeped in practical governance and geographies of governance, especially regarding legislated or otherwise formally designated categories.In Rio de Janeiro, the United Nations Conference on Environment and Development was held from 3-14 June 1992 which first recognised a group of countries and territories as Small Island Developing States (SIDS) (UNCED, 1992, paragraph 17.124).Member numbers have ranged from the thirties to the fifties, always with countries such as Belize, Guyana, and Guinea-Bissau which are firmly connected by land to their continents.The European Union (EU) for statistical purposes differentiates between "island regions" for which data are collated and "island units" for which the data are part of wider regions, while considering islands to be relevant within salt water, but not fresh water (Haase & Maier, 2021).
No matter what one's opinion for answering the question "What is an island?", replies (rightly or wrongly) have moved far beyond "a small piece of land surrounded by water".Discussions interface the physical with the social, making choices about culture, politics, human and physical geographies, and philosophies to symbolize, represent, label, and define islands in many different ways.Since these responses are choices, they are fundamentally human and so are done as social constructions with different levels of politicisation, whether based in law, data collection, ideology, belief, culture, or perception.
Based on this background, this paper argues that the interstitial island is principally a political construction.Interstitiality is, in effect, inbetweenness (after Oxford English Dictionary, 2022), referring to an area, entity, zone, or something more abstract which sits between parts.Islands are, or at minimum can be, multiple forms of interstices (Zhang & Grydehøj, 2021), but they are very much created as such-whether inadvertently, deliberately, or a combination-making the island a political interstice.This statement does not deny the physicality of some island as some interstices, but rather suggests that island interstitiality has far more political than physical value.As this paper shows, notions from island studies support this contention providing an offering to political geography-and vice versa.The island as an interstice, therefore, forms an interstice between island studies and political geography, bringing them closer together so that they could learn from each other.
To do so, this paper selects two characteristics discussed in island studies and political geography with high relevance to islandness: separation and connection.First, the positioning of this manuscript as an interstice between island studies and political geography is further theorised.Then, a section on "Island as interstitial separation" is followed by a section on "Island as interstitial connection".Conclusions explore separation and connection as interstitial.The theorisation and examples demonstrate that politics, political geography, and island studies infuse all the discussions, conceptualisations, and practicalities of the interstitial island.This situation brings strengths rather than detriments to understandings of islands and to island understandings.

Further theoretical positioning
The apparent definition of "island" indicates the need to explore "small" in more detail.Naturally, the concept is subjective and comparative, which is the main conclusion from political geography on smallness, such as when examining jurisdictions and territories (Anckar, 2006;Baldacchino, 2015;Selwyn, 1980;Streeten, 1993).Criteria and rankings have included parameters such as land area, combined land and water area, population numbers, population density, economic or resource metrics, and influence on the physical environment by creating a climate or an ecology.The alleged smallness together with other presumed island characteristics such as marginalisation and isolation has constructed islands as laboratories for experimentation (Weatherly, 1923)-which later included testing the effects of nuclear bombs on people, infrastructure, and ecosystems (DeLoughrey, 2007).This "island as laboratory" viewpoint has been overturned (Greenhough, 2006), as has the notion that smallness must confer disadvantages (Schumacher, 1973).
In fact, political geography and island studies work has led alternative perspectives, considering especially as forms of separation and connection: • Inclusivity and intersectionality.
Standard discourses of islands as small, isolated, and marginalised can themselves be marginalising, compared to the movements toward inclusivity and intersectionality (Crenshaw, 1989) in order to better involve and connect people or geographic entities, by considering their combinations of characteristics.In this regard, Jackman et al. (2020) offer multiple, intersecting characteristics of territory and terrain, indicating gender biases in their formulation and connecting different approaches to better understand the meanings, conceptualisations, and experiences of territory and terrain.Both interstices and islands are much more than their physicality (of terrain and territory; see also Kuus, 2010), further offering Naylor and Thayer's (2022) diverse economies and liveable worlds, as implemented operationally for islands by Baldacchino (2018) and countering the standard discourses that marginalise.In fact, interstitial geographies (notably of islands) can be used to negate standard worldviews and offer more inclusive perspectives (Chandler & Pugh, 2022).Inclusivity and intersectionality thus separate from previous assumptions of characteristics and framings that should dominate in order to connect others for diversity.
Going beyond typically dominant viewpoints, approaches, and articulations involves decolonising politically by supporting selfdetermination (UN, 1992) and intellectually by bringing in different geographical knowledges and understandings (Bonilla, 2015(Bonilla, , 2020;;Ó Tuathail, 1996).Even though many island jurisdictions have opted to remain more colonial as interstices than as sovereign (Baldacchino and Milne, 2006), and within such circumstances even challenge usual understandings of "non-sovereign" (Bonilla, 2015), the key is having a choice to freely make one's own informed decisions, of which resourcefulness is one way of doing so (Derickson & MacKinnon, 2015).Decolonial approaches to political geography can connect and create interstices among postcolonialism, non-representational theory, relational ontologies, cultural geographies, transmodernity, design, planning, border thinking, feminist geopolitics, and geopolitics of knowledge (Dittmer & Gray, 2010;Dodds & Sidaway, 1994;Grove, 2019;Naylor, 2021;Naylor et al., 2018;Scobie, 2019Scobie, , 2021)).Decolonisation thus can separate previously marginalised entities and knowledges from dominant forces while connecting those that were previously subservient or discounted.
• Environmental and social changes.
For the island as a political interstice, the focus on interstitial separation (section 3) and interstitial connection (section 4) thus has a broad and deep theoretical positioning.Geography research, including political geography, offers plenty regarding ontologies of space and spatiality interlaced with underlying epistemologies that inform understandings and knowledges of space, relationality, and territorialism, physically and metaphorically (see more in Dittmer, 2014;Hayward, 2012;Ó Tuathail, 1998;Pugh, 2016Pugh, , 2018)).The next two sections embrace this work to explore further aspects of islands and interstices for further enmeshing contributions from political geography and island studies.

"Island" as interstitial separation
In exploring the interstitial island, part of the attempt at defining "island" involves considering the distinctiveness of the term and the concept, within the assumption that an island can be separated, classified, and identified as different from something which is not an island.This structure sets up a dichotomy of islands and non-islands.This dichotomy is sometimes presumed to be entirely mutually exclusive and is sometimes accepted as overlapping in the structure of a two-circle Venn diagram-in effect creating an interstice, the size and characteristics of which depend on the definition of "island" and/or "non-island".Then, distinctions are developed, particularly with regards to separating (i) island and mainland (sometimes referred to as "continent") and (ii) land and water.
The island-mainland dichotomy introduces interstice possibilities.Taken literally in modern English, an island "is land", which implies "not water", although the word's etymology is actually from "water-land" or "river-land" (Oxford English Dictionary, 2022).Hence, "is land" is a misleading interpretation that might imbue misapprehensions about the origin or placement of islands.By contrast, the etymology of "mainland" is indeed "main land" (Oxford English Dictionary, 2022) connoting that it is the principal area within which outliers, such as islands, are excluded.Conflation occurs in places such as Shetland and Orkney in Scotland where the largest island within each archipelago is called the Mainland.The bias toward English, notable in geographical discussions where it is termed "anglophone squint" (Whitehand, 2005), must also be noted.Ronström (2009) surveys the word "island" in several languages of Europe and then focuses on Swedish, explaining that the ideas of water-land connections and interactions remain prominent across these languages-including for the non-Indo-European languages in Europe of Estonian, Finnish, and Hungarian.Yet the Chinese for "island" is "岛" which comes from "bird mountain" rather than water and land, while another one "屿" means relatively small pieces of land separated from a larger land mass at high tide while linked at low tide.
If the island-mainland dichotomy exists conceptually (not just linguistically) to separate these geographic entities, then where is the interstice, especially with "island" meaning "water-land"?The interstice might be water nearer the mainland until the island's water is reached, which imposes a separation on the waters despite them being a continuum.Alternatively, the interstice might be a boundary line rather than an interval.In this case, the island is not interstitial, but rather at the edge of or being the marginal area of a mainland, a trope found by island studies to be deeply problematic (Freitas and Kitson, 2018;Pugh, 2013a).Interestingly, the mainland would then fill one interstice role-namely, an interstice between islands, within the model that the mainland is the hub, so reaching the "peripheral" locations (the islands) must transit through the centre (the mainland).The rejection of this characterisation within island studies thus rejects the island as a peripheral node looking towards the main(land) hub, thereby leaving open the possibility for the interstitial island enveloping inclusive and intersectional notions from political geography (see section 2).This discussion is principally definitional, so aside from having considered mainly one dictionary of one language, namely the Oxford English Dictionary (2022), slightly tweaking or re-interpreting definitions can be completed according to opinions or circumstances, such as SIDS and the EU.As such, a large element is political of the interstitial island in regard to separating islands and mainlands.
The land-water duality nonetheless displays tangibility, given that it has a physicality to it through states of matter, with land being solid and water being liquid.Solid-liquid is an overlapping Venn diagram rather than a mutually exclusive dichotomy, as known from rheology which examines how matter deforms and flows (Goodwin & Hughes, 2008).The difficulty in separating the two completely appears in everyday life from glass being between a liquid and solid (Shelby, 2005) to geomorphological slides and flows sitting midway between solid and liquid (Pierson & Costa, 1987), e.g., mudflows, lahars, slush avalanches, sludges, and debris flows.Around coastlines, mudflats and quicksands bear similar properties, so the land-water interstice is not straightforward, even if the water part of islands would be defined as the molecule H 2 O as a liquid forming the chemical matrix.The latter point is a needed conditional since salt water by definition is not pure H 2 O while freshwater, too, contains plenty of dissolved and suspended components.Irrespective of sorting out the land-water distinction, the question from the island-mainland discussion repeats: where and what is the interstice, especially with "island" meaning "water-land"?Could the interstitial island be the connector between water and land?But then an island is not quite land surrounded by water.
In addition to the physicality, the attempted island-mainland and land-water splits are imbued with political meaning, because they represent conceptualisations and constructions which might not always withstand scrutiny across all cases.Illustrative examples are provided in the rest of this section through the island-mainland contrast being used to represent separation in space and the land-water contrast being used to represent separation in time.Because the examples come from a variety of sources, they are not robustly comparable.Consequently, aiming to quantify which ones might be more common or more prominent would be hard to defend.The key is demonstrating the variety of the representations, not ranking or tallying them.
The island-mainland separation represents islands as being the minor "other" to mainlands, especially in need of more help due to vulnerability, marginality, isolation, and remoteness-stereotypical island characteristics which are disputed (Conkling, 2007;Ronström, 2021;Selwyn, 1980).Thus, islands are grouped, almost homogenously, to provide them with more aid and support than the mainlands.This reasoning led to the SIDS group being founded with several formal plans of action since then (UN, 1994(UN, , 2005(UN, , 2014)).For statistical compilations, the EU defines island regions as (Eurostat, 2019, p. 104) "regions that are entirely composed of one or more islands" and then "islands are defined as territories having: • a minimum surface of 1 km 2 ; • a minimum distance between the island and the mainland of 1 km; • a resident population of more than 50 inhabitants; • no fixed link (for example, a bridge, a tunnel, or a dyke) between the island(s) and the mainland." Many of these regions are seen as needing specific EU measures due to their characteristics, especially the "outermost regions" which are assumed to have trouble on the basis of being so far away from the EU "mainland".
The focus by the political entities of SIDS and the EU is islandmainland separation through space.Whereas the island is the centre for those living there, the othering creates a spatiality of being marginal, peripheral, and outermost, despite extensive island studies work interrogating these characterisations in the context of islandness, with the evidence providing some verifiability and some counterexamples (Freitas and Kitson, 2018;Grydehøj & Hayward, 2014;Pugh, 2013a).
That is, it is not necessarily the case that being spatially distant confers more needs or exceptional requirements.Sometimes it occurs and sometimes it does not occur, with factors such as governance, resource allocation, livelihood choices, and dependencies tending to impact needs more than spatial distance from a neighbour while noting that spatial distance correlates in some ways with some characteristics (Baldacchino, 2020b;Canfield, 2020;Gillis, 2007;Ronström, 2021;Vézina, 2017).As an example, many island overseas territories (that is, the remaining colonies) eschew independence precisely because they believe themselves to be better off when governed by a larger state, irrespective of distance and irrespective of some evidence to the contrary (Baldacchino and Milne, 2006).Decolonial approaches espoused by political geography (see section 2) are explicitly evaded by those with power and by those voting in the overseas territories.Another example is travel time for an island product to reach markets being more important that physical distance (Danson & Burnett, 2014).
To group together all island geographic entities, with and without similarities, into SIDS or EU island regions and to accord identical needs to them within each group is a political choice.It is telling that, for instance, for human-caused climate change, the most vulnerable locations are reported to be: • SIDS: "Small island communities are the most vulnerable populations susceptible to the effects of climate change" (Akpinar-Elci & Sealy, 2014, p. 279).• Coasts: "Coastal areas have also been identified as the most vulnerable to climate change" (Moreno & Becken, 2009, p. 473).• Sub-Saharan Africa: "This group of people is considered the most vulnerable to climate change" (Adhikari et al., 2015, p. 111).• The Arctic: "Remote northern communities that continue to pursue subsistence-based ways of life are considered most vulnerable" to climate change (Haalboom and Natcher, 2012, p. 320), a point that is cited to another paper which nuances the articulation of this point.
These four geographic categories are not mutually exclusive, but are partitioned with interests trying to capture for them the crown of "most vulnerable to climate change".As with the discussions swirling around the "Anthropocene" (section 2), labelling and characterising are intensely political with respect to environmental and social changes.Even for ecosystems only, Li et al. (2018, pp. 4102-4103) list the variety deemed to be the "the most vulnerable environments to climate change impacts".They do not once use the word "island" whereas Veron et al. (2019, p. 1) begin their abstract with "Island systems are among the most vulnerable to climate change".
Spatially, then, the island-mainland separation and hence the role of interstitiality are both principally political.They are political in terms of defining "island" and "mainland"; they are political in terms of deciding what fixed or non-fixed links (see the discussion in section 4) to provide or not to provide to compress the physical separation among different pieces of land; and they are political in terms of analysing and labelling the expected consequences of being an "island" or a "mainland".Within this political geography spatialisation, where is the interstitial island?Nothing contradictory emerges in being both interstitial and outermost, particularly in considering a gamut of dimensions such as spatial distance, temporal distance, governance, culture, livelihood, and politics.
Physical distance for remoteness can be a boon for interstitiality, depending on definitions and interpretations, yet there are inevitably political undertones to using the island as such an interstice.Some examples: • For prisons (Nethery, 2012), islands can be used as interstices between criminality and law-abiding (or not caught) society.Incarceration on an island is seen as both punishment (a societal value) and impeding escape (assumed physicality, irrespective of its reality) for those who are sentenced to serve their debt to society.For prisoners expected to serve a sentence and then be released, the island prison is in-between the criminal and non-criminal worlds.• For military bases (Flint, 2021), islands can be interstices between peaceful society and military-based security and defence.Placing military bases on islands not only makes use of pieces of land labelled as being "strategic"-such as Diego Garcia in the Indian Ocean, with the Chagossians forcibly displaced to permit the base (Evers and Kooy, 2011)-but also uses the assumed isolation to "protect" military personnel and equipment.Here, the political and physical intertwine for the interstice.When the military withdraws and the land needs to be redeveloped, these islands are frequently seen as interstices between a militarised and peaceful region (Iwama, 2021), including from a post-colonial to decolonised jurisdiction (Grydehøj et al., 2021).• For quarantine (Baldacchino, 2020a), islands can be used as interstices between healthy and infected or possibly infected people.
Quarantine on an island is seen to keep infected (or potentially infected) and uninfected people apart (both a societal value and assumed physicality).Both needs are served by using water, rather than other mechanisms such as fences and guards, for the spatial separation of the groups alongside the temporal in-betweenness allowing people to overcome their infection or to show that they are not infected.• For people fleeing their home, islands can be used as interstices between dangers at the point of origin and a new life in a safer locale.
Examples, examined by Loyd and Mountz (2014), include Lampedusa and Sicily, Italy; Christmas Island, Australia; and Guam and Saipan, affiliated with the USA.In these cases, the island is used more as a political than physical interstice in space and time, to place people in limbo and out of sight with no option for moving on until a decision is made regarding in which direction the people are sent (Mountz, 2011).• For religious sites, including monasteries and hermit abodes, islands become interstices between sinful mainlands and the paradisical afterlife, as well as places 'in between' to meditate and learn (Royle, 2001).
All these examples are political choices, often under the assumption that islands serve the purposes better, whether or not they do.Prisons and military bases exist inside non-island cities; Toronto and Tokyo each has both (while noting that Tokyo is on a large island of an archipelago country).For much of the COVID-19 pandemic, New Zealand and Australia implemented strict, forced isolation regulations for arrivals to their (island?)countries, using hotels (typically near cities) as interstices in which people had to quarantine, rather than any of the these countries' smaller islands.One exception was Rottnest Island, Western Australia, used for a period for a 14-day quarantine for returning Western Australian residents (Moloney & Moloney, 2020) with suggestions later that it be used again, although it did not happen.
The interstitial island exists for the island-mainland spatial separation, but it is not just islands as interstices.The choice of what and where is interstitial, peripheral, or both-and why-is a political choice, sometimes referring to and accounting for alleged islandness and sometimes not.
The political meaning of the second separation, that of land-water, represents islands as land separated from other land by water, although no timeframe is typically assigned to this separation.Some pieces of land become islands with the daily astronomical high tide (most commonly diurnally); others with the high tide on other time scales, such as spring tides and king tides; and others when water levels exceed their typical height due to flooding, such as from wind, waves, precipitation, or solid water melting.
Bodies of water have average levels which rise and fall on other cycles, such as every 2-7 years when the El Niño Southern Oscillation (ENSO) raises or lowers different parts of the Pacific Ocean by up to 0.2 m (Glantz, 2001).Global mean sea level is influenced by the amount of water stored in the planet's glaciers and ice sheets, varying by dozens of metres over recent millennia (Dickinson, 2009).Therefore, in addition to local processes, the world's land-water separation is influenced by large-scale processes, notably: • The Earth's orbit around the Sun varying to move the planet in and out of ice changes (Hodell, 2019).• Human activities leading to a rise in global average temperature which increases sea-level through snow, ice, and permafrost melt and through the ocean water's thermal expansion (IPCC, 2021-2022).• Plate tectonics, moving and reshaping continents.
No island is permanent and thus it perhaps represents an interstice in time, with the same argument applying equally to mainlands.The interstitial island is real on multiple temporal scales, but is neither unique nor exceptional compared to mainlands or within any specific time period or timeframe.Should all timeframes be considered?Is orogeny as important for land-water separation as the diurnal tide or a flash flood?Depending on the purpose of understanding the island and depending on priorities regarding impacts considered, the answers might be different.Temporally, then, the land-water separation and hence the role of interstitiality are both principally political, connecting to political geography even with the above-mentioned physical geography influences.
Given this diversity of island interstices, different forms and degrees of separating out islands-of choosing how much overlaps from the two Venn diagram circles-can be developed and pursued according to particular interests.The separation or the interstice becomes principally artificial.It is possible to construct more separation or less separation of different forms through conceptualisations, definitions, interpretations, and actions, all of which are choices.How much of an interstice results, or if the point is to avoid interstitiality, becomes a political decision to be justified or not, drawing on the political geography perspectives from section 2 or not.

"Island" as interstitial connection
Components of connection as well as separation are inculcated into the interstitial island.The connection aspects are sometimes interpreted as pushback against or counter to the separation.They could alternatively be indicated as complementarities-a continuum or spectrum between separation and connection (see also Conkling, 2007)-especially regarding interstitiality.SIDS again exemplify.In addition to not all members being islands as noted above, not all have the same degree of smallness, such as Tuvalu's 26 km 2 compared to Papua New Guinea's 462,840 km 2 making it equivalent to 1.5 New Zealands (compare also with the analysis from Anckar, 2006).Singapore and Bahrain are SIDS despite the "developing" moniker while the status of being "states" does not apply to non-sovereign territories often listed as SIDS including Cook Islands (which is self-governing in free association with New Zealand, meaning that New Zealand retains responsibility for external affairs in consultation with the Cook Islands) and Montserrat (a UK Overseas Territory).The constructed identity of SIDS nevertheless yields advantages as an interstice between each island or archipelago jurisdiction and international development (Betzold, 2010;Klöck, 2020).This interstice creates the opportunity to pool resources for power and influence in the world despite small population numbers and sometimes limited influence as individual countries or territories (Fry, 2019 for the Pacific and Scobie, 2019 for the Caribbean).Pooling by islands goes beyond the realm of geopolitics, including sports, e.g., the West Indies cricket team, and education, e.g., the University of the West Indies, the University of the South Pacific, and the University of the Highlands and Islands.
Still, the notion of SIDS did not sit well with some parties in this group, so they have sought to redefine themselves as "Large Ocean States".As Chan (2018) details, the idea started perhaps around 2012 and was definitely on agendas by the time of Jumeau (2013).Other forms of the phrase include "Large Ocean Island States" and "Large Ocean Developing States" (LODS).Of note in some phrases is the shift away from the terms "small", "developing", and even "island".Here, the pieces of land within each "Large Ocean" area are seen as interstitial between the ocean resources leading to opportunities (see The Blue Economy discussion below) and the world.This interstitiality is an artificial construction, conducted to emphasise the island countries' roles-or perceived or desired roles-in the world, notably in relation to decolonial approaches from political geography (see section 2) by aiming to empower themselves on their own terms.Explicitly highlighting a state's water-land interactions is an important statement (irrespective of legalities which do so, e.g., UNCLOS, 1982).Here, it is being used in conjunction with "large" to convey power, authority, influence, and affluence-in effect, a political statement.The interstitial Large Ocean State, as with the interstitial SIDS, is political.
Another approach to building this form of the interstitial island through connections emerges from the neologism "aquapelago" (Hayward, 2012, p. 5) which is defined as "an assemblage of the marine and land spaces of a group of islands and their adjacent waters".Whereas an archipelago is seen as a collection of islands such that land is highlighted, an aquapelago is seen as the land of an island group and its waters.Baldacchino (2012) responds that "archipelago" already covers the concept and reality of land-plus-water, although "aquapelago" using a different label to repackage the same contents can have marketing advantages (Buckley, 2002).Hayward (2015) examines New York City, U.S.A. as an island city (Grydehøj, 2015) to analyse the archipelago and the aquapelago, indicating the interstitialities.Hayward (2015) implies Manhattan's interstitiality through phrases such as "the interaction of layers and types of water in the estuarine environment" (p.83), "the encircling estuarine waters aggregated and regulated the various ecological neighbourhoods" (p.83), "multiple nodes of interaction between estuarine and terrestrial environments" (pp.85-86), "an industrial buffer zone" (p.86), and "the fundamental interconnectivity of their land areas with the marine environment that surrounded them" (p.87).Here, interstitiality connects the island's ecology, industry, and living spaces, especially as they transform through time with land and water intermingling-corroborated via a similar analysis for Zhoushan, China (Zhang & Grydehøj, 2021).The aquapelago makes the island an interstice through connecting land and water, provided that the aquapelago rendering is accepted which is an overt choice (see also Gear, 2014) and so can be a political statement.
The same applies to another eternal discussion within island studies, that of fixed links (Baldacchino, 2007).Fixed links refer to structures such as bridges, tunnels, causeways, and dikes connecting two pieces of land across water (compare to Eurostat, 2019, p. 104 referring to "a bridge, a tunnel, or a dyke").They do not necessarily have to be solid constructions, as air routes and communications infrastructure can serve similar connectedness purposes.Then, they might be termed "non-fixed links", although ambiguity emerges in that airports and communications towers are fixed while aircraft and electromagnetic waves are not.Fixed and non-fixed links are debated as to whether they support or detract from island living and how they influence islandness (Baldacchino, 2007;Grydehøj & Zhang, 2020;Lee et al., 2017).As a less common example, a cable car was built between Dursey Island, Ireland and the mainland in 1969, partly to support farmers by improving mainland access (Royle, 1999).Many farmers moved to the mainland using the cable car to commute, so the island's resident population declined.
These links are certainly interstices themselves, being intervals and connectors between locations.Many enjoy aesthetics of and views from a fixed link, but they are generally built to be mainly functional, as efficient connectors so that the journey across is transitional rather than being the destination.
Concerns about structural fixed links include their impacts on water currents and ecology-for instance, as suggested for the archipelago of Smøla, Norway (Thomassen et al., 2008)-as well as their changes to the perception, feeling, and experience of the interstice to and from an island, notably where it connects to a non-island (Baldacchino, 2007;Lee et al., 2017).The idea, reasonable or not, is that ferries and airplanes have timetables and so require waiting and planning, inducing separation, compared to bridges, tunnels, causeways, and dikes which are, in effect, "interstice on demand", inducing connection-as long as people have private vehicles or can walk or cycle on their own.Much of this interpretation is psychological and political, especially to represent and promote assumed meanings of a fixed link.Before Denmark's largest island Sjaelland (with the capital Copenhagen) was connected to Jylland (Jutland) by bridges via the island of Fyn, ferries would operate continually at frequent intervals-although a regular ferry service still connects Sjaelland and Jylland on a route north of Fyn.Disruptions and delays from weather, traffic volume, and staffing occurred, recognising that bridges also have disruptions and delays due to weather, traffic volume, and crashes.The Netherlands and Norway use on-demand, self-operated and self-driven ferries respectively, for a few short crossings.Any short delays in waiting for the ferry are paralleled by traffic delays and crowds on structural links.Conversely, aircraft require expertise to operate and self-flying ones are a long way from being commercially available.Communication links do not entail travel, but also require expertise to set up and operate while they cannot be 100% reliable.Fixed and non-fixed links present different types of strongly overlapping interstices, with the options selected according to political decisions and the many impacts subject to political interpretation.
In fact, creating political viewpoints of these interstices is inescapably evident.Confederation Bridge connecting the Canadian provinces of Prince Edward Island (PEI) and New Brunswick (not considered to be an island) is a toll bridge.Users are charged on a round-trip basis when they leave PEI, not upon arriving.The psychology is making it more pleasant to arrive on the island than to leave it, which could have particular appeal for visitors, forming a warm welcome while making it feel annoying to leave.When the bridge opened connecting the Isle of Skye to much of the rest of Scotland, it had significant support, but was built and run by a private company charging a large crossing toll.A strong protest movement emerged and, in 2004, the Scottish Government purchased the bridge and abolished the toll.The Hong Kong-Zhuhai-Macau Bridge which officially opened in 2018 is resented by many in Hong Kong and Macau due to the symbolism of having a prominent fixed linked with China-even though both territories already had fixed links with China.
The question thus is raised about fixed and non-fixed links as "A connector to/from what?" which, for this paper, would be reframed as "An interstice between what?" Within the context of political geography's takes on inclusivity and intersectionality (see section 2), what is connected, included, and intersected, by whom, and why?Aquapelagos, fixed links, and non-fixed links aim to connect, include, and intersect, yet can be presented or framed in such a way that they do the opposite.If the island-mainland division is accepted (although it might not be), does a fixed link connect the island to the mainland, the mainland to the island, or both to each other?This question is important for examining the interstitial island-and the inclusive island and intersectional island-because the answer to the question is one of perspective; i.e., politics.Articulating that Confederation Bridge connects PEI to New Brunswick, or vice versa, imposes views of the start and end points-which makes the island and the non-island a start and/or end point rather than primarily an interstice.Conversely, Confederation Bridge connecting PEI and New Brunswick (or New Brunswick and PEI) makes the bridge an interstice without judging either province.
Other connectors within island studies raise similar questions with similar answers based on political choices.The main such connector is water.Naidu et al. (1993) and Hau'ofa (1998) articulate the ocean as a connector rather than as a barrier for islanders, seeing it as their transport route to others and to the world.Moving people across the water is accepted as the norm, so migration is part of islander life, livelihood, and culture-covering permanent relocation in addition to temporary and circular movements for education, healthcare, and livelihoods among other reasons (see also DeLoughrey, 2007;Fiddian--Qasmiyeh, 2020).Livelihoods are often highlighted by low-paid jobs, such as fruit and vegetable pickers, fishers, nannies, construction workers, and cleaners.Plenty of higher-waged employment lures islanders to migrate temporarily or permanently: science, consulting (frequently international development careers), natural resource exploration and extraction, and business development.
With respect to the fixed links discussion, this formulation then expresses two conceptual issues.First, the ocean is definitely a link, but how fixed is it?Second, what exactly is interstitial?Naidu et al. (1993), Hau'ofa (1998), and their "sea of islands" could be interpreted as avoiding separation between journeys and destinations: No boat launch is a beginning and no landfall is an end.Instead, the time on an island's land and the travel between pieces of land via the islands' waters are amalgamated journeys and destinations.Oceans are interstices for islands as much as islands are interstices for oceans; islanders live on the waters and lands together with the journey and destination being the same (see also DeLoughrey, 2007).
Perhaps the delineation between interstitial and non-interstitial fades as much as between fixed and non-fixed links.The label becomes, as with "archipelago" or "aquapelago", a political choice to use, providing helpful perspectives and inspirations, but debatable regarding novelty and pragmatism.The interstitial island is not so much an object (compared to an object of representation; Baldacchino, 2005) as part of the interstice of islander life and livelihood.
This understanding, grounded in islander knowledges (Hau'ofa, 1998;Naidu et al., 1993;Nakashima et al., 2018;Pugh, 2013b) is then politicised (which might be fair and needed) by morphing it into rhetoric for the global policy realm.SIDS, Large Ocean States, and EU island regions aim to connect land and water through the island by transforming the "Green Economy" (Pearce et al., 1989) focusing on the environment into the "Blue Economy" (Behnam, 2013) focusing on water.The Blue Economy is deliberate political marketing, with possible advantages in reconnecting islanders with their aqua environments, beyond external influences on tourism and fishing, including selling fishing rights to external bidders.The Blue Economy thus could be an island interstice spurring more local, long-term, lucrative livelihoods over the short-termism and dependency on external choices personified by tourism and resource extraction for living and non-living resources.Overcoming the rhetoric is not straightforward, with variations such as "The New Blue Economy" (Spalding, 2016) bearing remarkable similarities to "The Blue Economy" particularly since some authors conflate the two phrases.
Implications emerge in how the (New) Blue Economy is applied, some of which might be the antithesis of the Green Economy and some of which definitely promote the interstitial island as being between other global players: • Vote selling and vote trading at the International Whaling Commission (Dippel, 2015), since cetaceans are a Blue Economy resource.• Coastal resorts and recreational facilities exemplified by golf courses, promoting luxury and exclusivity, often typically to the exclusion of locals who are forced into servile, underpaid jobs with much of the facilities' takings going elsewhere (Graci & Dodds, 2010).• Large cruise ships, frequently epitomising consumption, superficiality, and high costs for the island ports where the ships stop (Dowling and Weeden, 2017).• S3 (sun-sea-sand) and S4 (sun-sea-sand-sex) tourism (Graci & Dodds, 2010), which can involve underage locals, denigrating local environments and peoples.US Spring Breaks and UK stag/hen parties are also notorious for this behaviour (Bell, 2008) while using the Blue Economy through beaches.
• Resource exploitation and extraction, with examples being offshore fossil fuels in Trinidad and Tobago (Hosein, 2021), seabed and deep-sea mining around the Caribbean (Scobie, 2019), and bioprospecting to patent genetic resources which is opposed in Hawai'i as a form of neo-colonialism (Kanehe, 2014).Nauru, through a combination of internal errors and external colonialism, squandered its rich phosphate resources leaving itself with no savings and a moonscape (Weeramantry, 1992), leading to further exploitation as Australia forces potential migrants into squalid conditions there (McAdam, 2017).Norway's Svalbard in the High Arctic began as a resource economy.As coal, its latest resource, loses popularity, the archipelago has been shifting toward a tourism economy with an overlapping science economy (Holmgaard et al., 2019)-all with an undercurrent of Norway's geostrategic interest in keeping primarily Norwegian settlements there (Government of Norway, 2015).• Payment to permit trans-shipment or dumping of toxic materials, such as e-waste into Indonesia (Rochman et al., 2017).• Rents for island military bases, such as Bermuda and Okinawa (Flint, 2021;Iwama, 2021).
Other specific Blue Economy approaches are much more in line with Naidu et al. (1993) andHau'ofa (1998), and represent the Green Economy too, demonstrating how much Blue re-markets Green: • Heritage-based livelihoods focused on supporting the heritage, especially combining natural and cultural heritages, some of which are accorded World Heritage status.Examples are Vega in Norway (Kaltenborn et al., 2013), Vallée de Mai Nature Reserve in Seychelles (Usher, 1993), and Svalbard (mentioned in the previous list).This approach might include patenting local products to bring money to the islanders, as was done in the Cook Islands for medicines (Das, 2020) and deliberate island branding as pursued by King Island, Tasmania for tourism (Khamis, 2007).• Marine Protected Areas, such as in Bonaire and Palau which have banned types of sunscreen to protect the ecosystems (Tovar-Sánchez et al., 2020).• Balanced aquaculture, such as for Singapore (Shen et al., 2021), including farming and harvesting of seaweed, sea grass, fish, crustaceans, and others for mainly local consumption with some export and trade.• Local electricity generation-e.g., offshore solar, offshore wind, osmosis, tidal, currents, and thermal conversion-aiming for island self-sufficiency in electricity (Surroop et al., 2018), while ensuring that expertise and parts for repair and maintenance are available from the island.• Rent and expenditure from scientists and students for research and education centres, such as the Bellairs Research Institute in Barbados and the Marine Research and High Education Centre in Maldives.
Many of these livelihoods, from international development and business development to resort tourism and electricity generation, are linked to the political geographies of environmental and social changes (see section 2).Whether aiming to address adverse impacts or using beneficial impacts, local-to-global changes interplay with and often create the Green and Blue Economies.
The interstices are hard to identify exactly, but descriptions could be mustered, if desirable (i.e., politically expedient) and (hopefully) useful.For the tourism livelihoods, the island and/or its heritage could be the interstice between non-locals and locals, further supported by Butler's (2012) analysis of trying to define "island tourism".Same with harvesting kelp and producing products from it.The level of interstitiality could be an academic discussion, even to the extent of seeking quantification such as through indices or rankings.Defining interstices or not, though, yields little overall impact on the livelihood.
Consequently, as with "island as interstitial separation", "island as interstitial connection" produces interstices, many of which are arbitrarily defined, being constructed for practical and political purposes, with or without culturally grounded or island-exclusive meaning and application, and drawing or not drawing on the political geography perspectives from section 2. The SIDS group followed by Large Ocean States and EU island regions exemplify."Sea of islands" is a powerful metaphor, yet the meaning and application depend primarily on defining "sea" and "island", just as with the difficulties of lucidly and consistently separating land and water as well as island and mainland.Then, different coloured economies are political constructions, repackaging what exists already, partly for motivational purposes and partly to garner attention and funding.The interstitial island emerges as political at many levels, even when practical aspects remain.

Conclusion: separation and connection as interstitial
The previous two sections show how separation and connection together help to indicate the island as an interstice or not, with specific interstitial traits and representations, thus connecting island studies and political geography.As noted and evidenced, separation and connection can be expounded as a false dichotomy since both typically exist simultaneously, even at different levels for a similar characteristic.Consequently, islanders and others purporting to represent islands often seek to construct and present islands as interstitial in different ways, some of which involve separation, some of which involve connection, and some of which involve connection and separation together.This pattern is further revealed by the definition of "island" in section 1 as a small piece of land surrounded by water, or variations thereof.In line with the theorisations and questions in section 2, further examination could be made of whether or not islands should be considered as being unique to planet Earth or to worlds with liquid H 2 O.Both political geography (e.g., Dunnett, 2017) and island studies (e.g., Webb, 2021) have examined interplanetary aspects, also referred to as 'astropolitics'.
The notion of small pieces of land surrounded by solid water raises the question of small pieces of land surrounded by gaseous water or water vapour, implying settlements floating in the sky as islands.Flying islands are noted in both fiction (Swift, 1726) and science (Riffat et al., 2016).Underwater cities, too, either floating or fixed are depicted as islands (Dawson & Hayward, 2016), as are space stations and lunar settlements (Schmidt & Zubrin, 1996), despite being surrounded by a vacuum.The definition of "island" is being stretched, perhaps beyond appropriateness, to connect with a variety of contexts.If a piece of land surrounded by solid water would be an island, then other solids ought to be considered, such as frozen carbon dioxide at Mars' South Pole (Guo et al., 2010).Since solid lava is rock, this definitional slippery slope leads to an island potentially being one type of land surrounded by another type of land!This phrasing of "island" is common in ecology: "sky islands" are high-elevation areas separated from each other and so harbouring a land population isolated from similar niches (Robin et al., 2010).Then, what relevance would there be for ponds-small areas of water surrounded by land-to be islands?Due to the overlaps of island types mixing different materials for the "land" and the land's surroundings, this "fluid" definition of "island" provides interstitial connections among diverse forms of real and fictional islands involving different matter.Numerous disciplines illustrate this point from molecular physics for building integrated circuits (germanium "islands" surrounded by silicon from Kamins et al., 1997) to the political geography of exclaves (examined as islands; Robinson, 1959) to encounters among different knowledges (examined as shoals; King, 2019) to filmmaking that mixes cultures in an island context (Zalipour, 2016).
False islands, too, demonstrate interstitiality in various forms.Islands which never existed or which no longer exist have appeared on maps for centuries (Stommel, 2017), leading to discussions melding these islands' political and physical geographies (Nunn, 2008).Sometimes they are placed inadvertently, through belief that an island was sighted or through misinterpretation of observations.Sometimes, they are placed deliberately to confuse others claiming similar territory, for outright fraud, or to identity copyright theft by people reproducing a map, and hence its errors, without credit.False islands appear in remote sensing through misidentifying cloud shadows on the ocean surface (Dai et al., 2019) and through shading topography so that low-lying areas appear as water, thereby apparently separating pieces of land (Slater et al., 2011).These diverse forms of false islands become interstitial between the real and imaginary worlds of islands and islandness, in effect separating islands which exist and islands which do not exist.In parallel, Lopesi (2018) details "false divides", referring to artificial separations of Pacific island peoples by colonial powers.This attitude isolated peoples from islands and the ocean rather than adopting the islanders' approach of the ocean as a connector (see also Section 4's discussion of Naidu et al., 1993;Hau'ofa, 1998).These separations do not preclude changes over time, since islands appear due to underwater volcanism (e.g., Surtsey, Iceland in 1963;Thorarinsson, 1967) and disappear due to erosion (e.g., Bangladesh's chars; Sarker et al., 2003), with concomitant political geography consequences regarding territories, knowledges, and identities (Baldacchino & Tsai, 2014;Grydehøj, 2016;Mountz, 2014b;Roper, 2007).This separation, instead, is interstitial separation for real and fictional islands-as well as interstitial connection between the observable and the imaginary.
Much of this joint interstitial separation and connection for islands is indeed from the physical nature of being a small piece of land surrounded by water while simultaneously being morphed and moulded, across interpretations and purposes, through and for political geographies.Other interstitial aspects are entirely constructed politically to manufacture and provide roles for the island-in livelihoods, with peoples, and for the world stage.Philosophically and practically, many advantages result from constructing the island as a political interstice, suggesting that the interstitial island has far more political than physical value, without obviating the physicality in many circumstances.

Declaration of competing interest
The author declares that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.