Our seas of islands

Epeli Hau‘ofa’s essay on a ‘sea of islands’ was intended to offer a bottom-up, corrective, and holistic view of Oceania. Instead of colonial images of the Pacific as a vast ocean with tiny isolated islands in it, he included the sea as part of what can constitute a home and reimagined Oceania as historically inflected ‘networks . . . integrated by trading and cultural exchange systems’ (Hau‘ofa, 1993, 7–9). From a perspective on the sea, a large landmass can be a haven, danger, or obstruction. Smaller islands might not only block travel, but they can also offer the interactive space of a shore combined with a more accessible interior. Islands may also reticulate in a variety of forms, sometimes presenting series of lands that offer waystations for sea travel. Seas additionally narrow and transition to rivers that can lead far inland. Although an idealistic strain in Hau’ofa’s and others’ visions of Pacific and other maritime networks has been criticized, the point remains that while some oceanic expanses can present a barrier, they tend instead to facilitate travel.¹

These general observations about islands and sea suggest some beginning points for thinking about late-medieval archipelagism, but describing a specifically British area and what it contains remains problematic. We look therefore to the political historian J.G.A. Pocock who adopted the term ‘Atlantic archipelago’ to denote ‘a large – dare I say a sub-continental? – island group lying off the northwest coasts of geographic Europe, partly within and partly without the oceanic limits of the Roman empire and of what is usually called “Europe” in the sense of the latter’s successor states’ (Pocock, 2005, 29). Pocock identifies himself in part as a New Zealand historian, and his antipodean perspective suggests a view of ‘the world as an archipelago of histories rather than a tectonic of continents’ (Pocock, 2005, 19), ‘histories both transplanted by voyagings and generated by settlements and contacts, and consequently as never quite at home’ (Pocock, 2005, 19). British and New Zealand histories, he argues, share myths and narratives of ‘peoples in motion’ and ‘histories traversing distance’ (Pocock, 2005, 23).

From a late-medieval perspective, every locational and spatial term is problematic: England, Britain, British Isles, Atlantic, North Sea, Great Britain, ‘oure occian’ and ‘oure wilde see,’ and so on. In her essay ‘The trouble with Britain,’ Patricia Ingham points out the frequent conflation of one place, region, or name with another, England for Britain most commonly, whereas whatever the entity is, ‘multi-geographic’ might be the most apt term. People forget, for instance, that ‘Great Britain’ and ‘the more Britain’ were the terms to distinguish the island or islands from Brittany. Even Pocock’s adopted term, ‘Atlantic archipelago,’ for all its destabilizing and multiplying, still posits a cohesive sociological unit and is only vaguely located. The term archipelago is itself somewhat anachronistic when applied to medieval geography; it does not appear in English before 1500, at which time it refers to the Aegean Sea before being applied more broadly. Pocock’s ‘Atlantic archipelago’ is a term nevertheless with a past that confirms his idea of archipelagic history. The term does not seem to have been used to refer to Britain, Ireland, and so on until the 1980s and may in part be attributable to the Falklands War. At that point in time, ‘South Atlantic Archipelago,’ or just the ‘Atlantic Archipelago,’ referred to the Falkland Islands 300 miles east of Argentina. It would appear that the reference to British, Irish, and other islands as the ‘Atlantic archipelago’ therefore is transferred from the South Atlantic to the North during that time of martial and political upheaval. This would confirm Pocock’s observation that modern British history is ‘Formed partly in an archipelago of the Southern Ocean’ and ‘presents the islands including Britain as another archipelago . . . not the promontory of a continent; it presupposes histories “not in narrow seas”’ (Pocock, 2005, 23).² We might therefore adopt and adapt Pocock’s term, and call our area of study the ‘Northeast Atlantic archipelago.’

This introduction offers a partial overview to the volume and indicates one possible path to thinking about late-medieval seas of islands in the Northeast Atlantic. Geography, genealogy, law, politics, ethnography, linguistics, economics, literature, and other disciplines afford distinct historical ways to examine the region in the thirteenth, fourteenth, and fifteenth centuries. In the space available, we will address the first four – geography, genealogy, law, and politics – with the understanding that there are of course intersections and overlap among these discourses as well as limitations and challenges within each. It may indeed be said that our approach is under the sign of the number four: four disciplines, four winds and so the cardinal directions, four corners of the world, four seas, and England, Ireland, Scotland, and Wales. This quadrivial venture emphasizes geographical inconsistencies alongside a growing legal and political coherence to the Northeast Atlantic archipelago.

Despite a temptation to collapse disciplines and approaches, geographical considerations of the archipelago are distinguishable from political, ethnographic, and other lines of inquiry, including a proto-English nationalism that has gained a considerable amount of focus in recent years. The insular and maritime entities that the term ‘North Atlantic archipelago’ encompasses are complicated in many ways: what they are, where they are, and the relations among them. Islands are firstly geographical in late-medieval European writings, not in the sense that geography is a distinct, well-defined discipline, but Greek, Roman, and medieval authorities, after describing the three continents of the northern hemisphere, include islands as a separate group. Britain is often prominent in this group; Isidore of Seville, for example, is typical in this regard in Book 14 of the Etymologies when he groups chapters on islands, beginning with Britain, then briefly describes Thanet, Ultima Thule, the Orkneys, Ireland, Cadiz (considered an island), the Fortunate Islands, the Gorgades (traditionally off the African west coast), and so on (Isidore, 1911). Bartholomaeus Anglicus’ thirteenth-century De proprietatibus rerum, his encyclopedic compilation of sources about the world, has a chapter on the earth’s regions, and its introduction states the world has three parts before it resorts to an approximately alphabetical treatment of its various places (Bartholomaeus, 1483, 15.1). The following uses Bartholomaeus, Ranulph Higden’s Polychronicon, and John Trevisa’s late-fourteenth-century translations of both to examine geographical discussions of the Northeast Atlantic archipelago.

First we might ask where the entities are. Christine Chism’s ‘Britain and the Sea of Darkness: Islandology in al-Idrīsī’s Nuzhat al-Mushtaq’ suggests that the answer in part depends on the location from where one looks. From Norman Sicily, the North Atlantic archipelago is in the ‘Sea of Darkness,’ also a place of ‘epistemological breakdown.’ The traditional distinction in both Christian and Muslim geography between Asia-Africa-Europe and the rest may account for Higden’s citation of authorities that say that ‘Anglia Britannica alter orbis appellatur’ and that the edge of the French cliff is the end of the world, so Britain merits being called another world (Higden, 1869, 2:1.39).³ Bartholomaeus likewise describes the following: ‘Anglia occeani est insula maxima que circumfusa mari a toto orbe undique divisa.’ Trevisa renders the passage, ‘Inglond is þe moste ilond of Occean and is biclippid alle aboute with see and departed from þe roundenesse of þe worlde’ (Bartholomaeus, 1975, 15.14). Both Higden and Bartholomaeus claim that Isidore says that Anglia has its name from being ‘angulus’ because it is at the end of the world or, as it were, makes up a corner of the world. The allusion seems to be to Isaiah 11:12 and the ‘four quarters of the earth,’ although why Anglia is in one corner appears to be a traditional addition. In Bartholomaeus, it is the most fruitful and fertile ‘angulus orbis’ (Bartholomaeus, 1483, 15.13; Higden, 1869, 2:1.39). The names are complicated, but Isidore (cited incorrectly as ‘Alfridus’ in the Trevisa translation of the Polychronicon) says further that England is an island, and islands are called such because they are ‘in salo’ [‘in the sea’] (Isidore, 1911, 14.6; Bartholomaeus, 1975, 6).

On the topic of names and the location of the entities within the North Atlantic archipelago, Bede, Geoffrey of Monmouth, Wace, Gerald of Wales, and others offered their versions of genealogical history in earlier centuries, many of which were picked up by later authors. Other less-known genealogies were added. Their contemporary Gervase of Tilbury, for example, suggested the name England from the Angles, who came from a Saxon island of ‘Engla’ (Gervase, 2002, 306–307). Again, both Bartholomaeus and Higden are interesting. They not only repeat the stories from Bede, Geoffrey, and others of Brutus traveling to an island coast during his exile, but they also posit less-known origins for the names. Bartholomaeus and Higden have an abbreviated version of the story of Brutus and then the Saxons, but they also add that the latter named the land after the daughter of one of the Saxon ‘dukes,’ ‘Engela regina.’ Both repeat the story of Gregory the Great and angels, but they also say that the island has its name of Albion geologically because of its white rock cliffs above the sea (Bartholomaeus, 1483, 15.13; Higden, 1869, 2:1.39).

Was the archipelago considered one geographical unit? Was it an it, or were the islands a they? Singular or plural? These are the kinds of questions Alfred Hiatt asks in his essay ‘From Pliny to Brexit: Spatial representation of the British Isles,’ in which he follows Patrick Gautier Dalché in warning that ‘it should not be assumed too readily that the concepts of “island,” “continent,” and even “ocean” remain identical across different cultures.’ The Polychronicon attributes to Henry Huntington the idea that England is ‘floure of londes al aboute,’ and it is unambiguous in stating that the ‘ilond of Bretayne’ in Brutus’ time ‘bygan for to have þ[r]e principal parties’: England, Wales, and Scotland, which also have three main islands near them – Anglesey, Wight, and Man – as well as smaller ones (Higden, 1865, 1869, 1–2:1.37–43). Ireland is not included as part of the Polychronicon’s chapters on England-Britain (although its earlier chapter on Hibernia says it ‘was of olde tyme incorporat in to þe lordschippe of Bretayne’) (Higden, 1865, 1:1.32). Bartholomaeus’ entry on Ireland says it is near Britain, but he does not say it is part of it (Bartholomaeus, 1975, 15.74). Bartholomaeus is in fact interestingly contradictory in his entries on Anglia, Britania, Europa, Hibernia, and so on. First, it is curious that he has entries both for Anglia and Britania. Second, neither corresponds with today’s England or Britain. Trevisa’s translation begins the entry on Anglia by saying ‘Inglond is þe moste ilond of Occean,’ while the entry on Britania simply begins ‘Briteyne is an ilonde of Occean in Europa’ (‘in Europa’ in the Latin) (Bartholomaeus, 1975, 15.14, 15.28). Third, nowhere does Bartholomaeus say that England is part of Britain; they are instead chronological terms: Albion, Britannia, then Anglia.

Even whether Britain/England and Ireland were considered insular in any significant way is a valid question. In the geographical writings, it is not always clear whether Britain and Ireland are islands, as though their greater size tips them over into the category of merely a land. Trevisa often substitutes ‘londe’ for the Latin text’s ‘insula’ when he translates observations about Britain and Ireland, or else he mixes kinds of lands together. In the entry for Europe in Bartholomaeus, for instance, he says ‘And þereaftir [Barbaria] is Gothia, and þanne Dacia, Dennemark, þan Germania, þanne Gallia and þe londe of Breteyne, Orcades, and many ilondes’ (Bartholomaeus, 1975, 15.50). The entry for Ireland similarly says, ‘Irlonde hat Hibernia and is an ilonde of Occean in Europa, and is nȝe to þe londe of Bretayne, and is more narowe and streyte þanne Bretayne and more plenteuouse place’ (Bartholomaeus, 1975, 15.151). Scotland is quite confusedly described as ‘a lon[g]e strecchinge cuntrey, as it were a forlonde in þe erþe⁴ of Bretaigne. And is departede fro norþe Englonde with ryvers and armes of þe see, and is yclosede aboute with Occean in e[i]þere syde, and is also departede fro Irlonde wiþ þe see of Occean’ (Bartholomaeus, 1975, 15.151).

What kinds of relationships obtain among the lands? Three entities are important: other islands or lands, Europe, and the sea. References to other islands are as frequent as they are to continental lands. For Higden, Ireland ‘is þe laste of alle þe west ilondes,’ Spain is three days’ sailing to the southeast of it, ‘þe more Bretayne’ is to the east one day’s sail, and Iceland is three days’ sail to the north (Higden, 1865, 1:1.32). Elsewhere, he claims that Bede ‘clepeþ Hibernia propurliche i-nempned þat west ilond þat is an hundred myle from every Britayne, and departed wiþ þe see bitwene’ (Higden, 1865, 1:1.37). For Britain, he says, ‘Þe meeres [“limites”] and þe marke were þerof somtyme þe Frensche see boþe by est and by souþ’ (Higden, 1869, 2:1.43). Bartholomaeus’ authorities say ‘Brytayne þat now hat Anglia is an ilonde ysette aforne Fraunce and Spayne’ and ‘haþ Fraunce in þe souþe syde,’ and we have already read of it and Ireland being ‘in Europe’ (Bartholomaeus, 1975, 15.14, 15.28). Looking elsewhere for evidence, Flanders is said to have Germania on the east side, ‘þe ilonde of Bretayne in þe north, and þe Frenssge see in þe weste’ (Bartholomaeus, 1975, 15.58).

Various relationships to the sea also appear. The islands are usually described in terms of distance, often including Thule and Iceland, as we have seen. An instance of this sort of measurement in addition to Scotland above is the Orkneys, which Bartholomaeus calls ‘an ilond of Occean in þe Brittishe see in Europa, þerof many oþer ilondes þat ben nyȝe þerto han þe name’ (Bartholomaeus, 1975, 15.110). Thanet is likewise an ‘ilond of Occean,’ ‘departede from Bretayne wiþ a litil arme of þe see,’ and ‘Vitria,’ similarly to the Orkneys, ‘a litil ilonde in þe Bryttisshe see […] departede from þe more Bretayne with a litil arme of þe see.’ (Trevisa interposes in the description of ‘Vitria’ that no mention is made of which direction it lies from Britain, and he questions whether it is the same as the island of Wight [Bartholomaeus, 1975, 15.154, 15.172]). Thule is ‘þe last ilonde of Occean bytwene norþ cuntrey a south, vi. dayes sailynge byȝonde Bretayne,’ while Iceland is ‘þe laste regioun in Europa in þe norþe biȝonde Norwey, in þe firste parties þerof is alwey ise and glass. And streccheþ uppon þe clyffe of Occean toward þe north, yfrore for grete and stronge coolde. And Islonde hath þe over Scicia in þe eest syde, and Norwey in þe souþe, and þe Irysshe occean in þe weste’ (Bartholomaeus, 1975, 15.160, 15.173).

The conclusion from these geographies and genealogies, and there are many other works of course, is that there is very little consistency in terms of the entities themselves, where they are, and the relationships among them. There is even less correspondence between how they were described and the entities we know today as Britain, England, Ireland, the North Sea, the North Atlantic, and so on. One final observation involves cartographical evidence, which suggests something similar, namely that the handful of surviving late-medieval maps also presents a less than consistent view of the area. One problem with maps is that they vary in terms of what they depict because of framing and scale. The Matthew Paris, Gough, Totius Britanniae Tabula Chorographica map (British Library Manuscript Harley 1808), and even John Hardyng’s map of Scotland constrict the focus, while maps of larger areas, including the so-called mappaemundi and portolan charts, tend to show a variety of relationships among lands.

But if the Northeast Atlantic archipelago was not consistently, or at all, considered one geographical unit, were there other senses in which this archipelago was thought of as one entity? Can, as James Smith asks in ‘Brendan meets Columbus: A more commodious islescape,’ separation paradoxically bring things together? In the second edition of his Principall Navigations, Voiages, Traffiques and Discoueries of the English Nation, printed in three volumes between 1598 and 1600, Richard Hakluyt gathered some seventy medieval texts, in addition to hundreds of later works, to build an implied case for English imperial expansion (Hakluyt, 1598–1600).⁵ Hakluyt’s quasi-legal case itself rests on precedents, ancient trade links, insular conquests, and control over the sea surrounding the archipelago. The medieval texts he collected for his volumes range from Ohthere’s northern voyages, over diplomatic spats with the Hanseatic League, to the first complete printing of the virulently jingoistic fifteenth-century poem The Libelle of Englyshe Polycye, which, in the second and more soberly political edition of the Principall Navigations, replaces Mandeville’s Travels.⁶

Among the economic, cultural, and political arguments for an archipelagic Britain advanced by Hakluyt, the legal claims to England’s maritime dominion over the British Isles assume a central position. Of the seventy medieval case studies that preface the Principall Navigations, King Edgar’s circumnavigation of Britain is the most outspoken about England’s archipelagic ambitions, besides revealing that a residual understanding of the unified British Isles as-located-in-the-British-Ocean existed ever since the twelfth century. Although Hakluyt props up the historicity of this momentous claim by stating that he has sourced the material for this section from Roger Howden’s Chronica and the Worcester Chronicle in addition to John Dee’s General and Rare Memorials, the entire section only reproduces material assembled by Dee, whose ideas, in turn, are given an impressive platform in the Principall Navigations. At the heart of Dee’s, and therefore Hakluyt’s, claim of archipelagic dominion is the belief, grounded in the spurious ninth-century charter Altitonantis (which was actually a twelfth-century forgery), that King Edgar ruled the entire British Isles by virtue of keeping the seas:

Altitonantis Dei largiflua clementia, qui est rex Regum, Ego Ædgarus Anglorum Basileus omniúq[ue] Regum, Insularum, Oceaníq[ue] Britanniam circumiacentis, cunctarúmq[ue] nationum qu[æ] infra eam includuntur, Imperator, & Dominus, gratias ago ipsi Deo omnipotenti, Regi meo, qui meum Imperium sic ampliauit, & exaltauit super regnum patru[m] meorum: qui licet Monarchiam totius Angliæ adepti sunt à tempore Athelstani (qui primus regnum Anglorum, & omnes Nationes, qu[æ] Britanniam incolunt, sibi Armis subegit) nullus tamen eoru[m] vltra eius fines imperium suum dilatare aggressus est. Mihi autem concessit propitia Diuinitas, cum Anglorum Imperio, omnia regna Insularum Oceani, cum suis ferocissimis Regibus, vsq[ue] Noruegiam, maximámq[ue] partem Hyberniæ, cum sua nobilissima Ciuitate Dublinia, Anglorum regno subiugare: Quos etiam omnes, meis Imperijs colla subdere (Dei fauente gratia) coegi. Quapropter & ego Christi gloriam, & laudem exaltare, & eius seruitium amplificare deuotus disposui, & per meos fideles Fautores, Dunstanum viz. Archiepiscopum, Athelwoldum, & Oswaldum episcopos (quos mihi patres spirituales, & Consiliatores elegi) magna ex parte, secundum quod disposui, effeci, &c. (Hakluyt, forthcoming)

[By the plentiful mercy of God thundering on high, who is king of kings, I Edgar, king of the English and emperor and lord over all the kings of the islands of the ocean that surround Britain, and of all the nations that inhabit it, give thanks to the omnipotent God, my king, who has so considerably enlarged my empire and raised it above the realms of my forefathers. Although they held the crown of all England since the time of Æthelstan (who, first among the kings of the English, conquered by force all the nations that inhabit Britain), ultimately none of them has attempted to extend his control beyond its boundaries. Yet divine favour has granted me, together with the realm of the English, to subjugate to the kingdom of the English all the kingdoms of the islands of the ocean with their fierce kings, as far as Norway and most of Ireland, with its excellent city of Dublin. All this I have subjected to my yoke with the help of God’s grace. Therefore I exalt Christ’s glory and praise, and I intend to increase His worship. I have performed much of what I intended with the help of my trusted aides, namely Archbishop Dunstan and bishops Æthelwold and Oswald, who are spiritual fathers to me and whom I have chosen as counselors.

(Hakluyt, forthcoming)]

This charter, as do some others, styles Edgar as basileus, an insular emperor ruling over Britain, Ireland, Norway and ‘the islands of the ocean that surrounds Britain’ – in other words, Pocock’s Northeast Atlantic archipelago and then some.

John Dee makes this twelfth-century forgery the centerpiece of his case for archipelagic dominion, buttressing his argument with chronicle accounts of Edgar’s attempt to enforce his total control of the British Isles by his annual circumnavigation, a claim often repeated from the twelfth century onwards. According to most accounts, every year Edgar would command an Anglo-Saxon fleet – inflated by Dee to thousands of ships and a hundred thousand soldiers – to sail ‘about this Brittish Albion, with all the lesser Isles next adjacent round about it,’ and ‘by such ful and peaceable possession, find himselfe (according to right, and his hearts desire) the true and soveraigne Monarch of all the British Ocean, environing any way his empire of Albion and Ireland, with the lesser Islands next adjacent’ (Hakluyt, forthcoming). Despite the artificiality of Dee’s numbers, this was a very real claim made by his sources, perhaps articulated best, again, by Higden, as transmitted by Hakluyt’s Dee:

dem quoque Ædgarus, 4000. naues congregauit, ex quibus omni anno, post festum Paschale, 1000. naues ad quamlibet Angliæ partem statuit, sic, æstate Insulam circumnauigauit: hyeme verò, iudicia in Prouincia exercuit: & hæc omnia ad sui exercitium, & ad hostium fecit terrorem.

(Hakluyt, forthcoming)⁷

[This Edgar gathered 4000 ships, of which each year, after Easter, he placed 1000 on each side of England. Thus, he circumnavigated the Island in summer. In winter, however, he administered justice in his territory. And he did all this for his own navy and the deterrence of his enemies.

(Hakluyt, forthcoming)]

Ever since the twelfth century, Edgar’s imperial fiction, loosely supported by the legal status of the Altitonantis charter, was used to uphold claims to Ireland (especially Dublin), the Isle of Man, and other islands located in the seas surrounding Britain (Sobecki, 2011a).

The second tradition that lent force to the idea of the British archipelago (united under an English monarch) was the alleged subjection of between six and eight insular kings on the River Dee near Chester shortly after Edgar’s coronation on 11 May 973. This event is reported in the various recensions of the Anglo-Saxon Chronicle (Anonymous, 1996–2007) as well as in many other texts (Thornton, 2001):

7 sona æfter þam se cyning gelædde ealle his scipfyrde to Leiceastre, 7 þær him comon ongean .vi. cyningas, 7 ealle wið hine getreowsodon þæt hi woldon efenwyrhtan beon on sæ 7 on lande.⁸

[And immediately after that the king took his whole fleet to Chester, and there six kings came to him, and all gave him pledges that they would be his allies (lit. fellow workers) on sea and on land.

(Thornton, 2001, 50)]

Such accounts of Edgar’s alleged rule over all the islands surrounding Britain, by virtue of controlling the seas, filtered into numerous chronicles and historiographies, from the Melrose chronicle to Aelred of Rievaulx’s Genealogia regum Anglorum, until they eventually reached the fifteenth-century proto-mercantilist Libelle of Englyshe Polycye, a poem most likely associated with the highest ranks of Henry VI’s government and, in particular, with William Lyndwood, Keeper of the Privy Seal, and one of medieval England’s most formidable canon lawyers (Sobecki, 2011b).

By the sixteenth century, with help from the Libelle, Dee, and Hakluyt, the Northeast Atlantic archipelago had become Edgar’s archipelago, a distinctly English and imperial construct.⁹ Lynn Staley’s essay ‘Fictions of the island: Girdling the sea’ similarly posits the ‘off-shore calculus’ of mercantile perspectives of the time that sought to turn an image of seas of islands into a commercial ‘highway of profit.’ In the seventeenth century, John Selden equipped his defense of territorial waters, Mare clausum, with Edgarian support, and Samuel Pepys, clerk of the acts of the Navy Board, owned a copy of Marchamont Nedham’s 1652 translation of Selden’s work. A small panegyric poem addressed to an apostrophized Britannia, and prefaced to Nedham’s translation, spells out in English what Edgar’s archipelago would begin to mean for those who, over the coming centuries, came into contact with an English-ruled Northeast Atlantic archipelago:

New Seas thou gain’st; & to the antient FOUR

By Edgar left, thou addest many more. (Selden, 1652, frontispiece, ll. 23–24)

The extent of the post-Edgarian annexation of oceans to the Four Seas became part of the legal fabric of the British Empire – even the future president of the United States, John Adams, had to fight off Edgar and his archipelago (Sobecki, 2011a, 29–30). The de jure incorporation of other seas is showcased in a trial of 1834. After the captain of a Royal Navy ship was accused of falsely imprisoning a sailor off the coast of New Zealand, the attorney general for New South Wales stated that

[b]y the law of England the King of England is Lord of the four seas and his power over the ocean has extended from time to time by the growth of the Navy and the maritime ascendancy of Great Britain.¹⁰

But before this Northeast Atlantic archipelago – Edgar’s archipelago – could start adding new seas to its small insular empire, it needed to assert Britain’s proprietorship of its adjacent waters.

The essentially paradoxical term ‘territorial waters’ is precisely that: an attempt to turn sea into land, wilderness into legal property. For medieval and post-medieval readers, Edgar’s circumnavigation had created the precedent of an English-centered British basileus who divided the sea he owned into four administrative regions: the Channel, the North Sea, the Irish Sea, and the Western Sea or the southwestern approaches of the Atlantic.¹¹ In a legal context, these four seas became synonymous with proximity, familiarity, reach, and, essentially therefore, Britain itself. Steve Mentz’s essay ‘The Bermuda assemblage: Toward a posthuman globalization’ suggests something of the same accumulatory aspect of Bermudan identity, an accretion of elements that – only retrospectively and from a limited human point of view – achieves a form of coherence.

The legal appropriation was gradual yet lasting and explains why the Edgarian fiction, when it was invoked in the later Middle Ages, was so successful. The groundwork had been prepared in legal discourses since at least the late thirteenth century. The following case from 1290, known as the Modus Levandi Fines, invokes the ‘binding force of the collusive conveyance of land known as the Fine (or final concord)’ on persons who do not object to it within a year:

And the cause wherefore such solemnity ought to be done is because a fine is so high a bar, of so great force, and of so strong nature in itself, that it concludeth not only such as be parties and privies thereto and their heirs but all other people of the world, being of full age, out of prison, of good memory and within the four seas (dedeinz les quaters meers) the day of the fine levied. (Kiralfy, 1989, 380)

Albert Kiralfy has demonstrated that the regular occurrence of the phrase dedeinz les quaters meers, or its Latin translation, ensured that the ‘four seas’ became interchangeable with ‘beyond the seas,’ that is, beyond Britain: ‘“Beyond the seas” was a later equivalent of the “four seas”’ (Kiralfy, 1989, 382). From a legal point of view, that which was thought of as lying within the seas circumscribed jurisdiction; conversely, regions beyond the (four) seas denoted the end of English jurisdiction. So when Edgar’s circumnavigation returned to the forefront of political discourses with the fifteenth-century poem The Libelle of Englyshe Polycye, the legal fiction of the Four Seas had already been absorbed as an expression of implied territorial waters. As such, Edgar’s archipelago had been established long before Selden, Cromwell’s agent Nedham or the Stuart navy could speak of Edgar as ‘Quatuor Maria vindicare’ (Waterhouse, 1663, 408).

Yet insular connectivity is not just about water. Peregrine Horden’s afterword draws attention to land routes in the ongoing archipelagism or ‘Philippinization’ of Britain. Such a focus reminds us of other spatial and motile networks, ‘terrestrial connectivity,’ besides maritime seas of islands. Thinking along these lines might at least serve to remind us that oceans are one of the four principal water ecosystems. The others are rivers, lakes, and wetlands, and each deserves its own particular attention because of its defining characteristics: rivers dividing and connecting, striating Britain on maps, and subject to control of a different kind from seas; lakes in some ways similar to forests in their wildness and as resources; fens, mires, and sloughs with all their ambiguities and challenges. Nor should we forget about islands within islands, such as the Isle of Ely: six hundred years before Dutch hydro engineers drained the Fens of East Anglia, Hereward the Wake blockaded the island of Ely, delaying William the Conqueror’s attempt to connect England to the Duchy of Normandy. Or consider status-changing geographies, such as with the Isle of Thanet, whose separating waters meant that it was a main route of transport from London to the channel for hundreds of years, vessels passing between the mainland and the island until the fifteenth century when Thanet became un-islanded as the rivers silted up. If, as Horden argues, ‘connectivity’ stands for ‘high connectivity,’ then an inward focus on riverine transport, mobility, and the narratives such movement generates can only enhance the idea of the archipelago as such: by linking the shores of islands with their hinterlands, these internal regions are, in turn, connected to the other islands of the archipelago.