Maps are wonderful tools that can help us find our way and divide up our surroundings: into our land and theirs, into safe places and unsafe, and, ultimately, into the known world and the unknown. Exploring has become tantamount to mapping, turning the empty margins and blank areas of terra incognita into familiar terrain. In a literary genre as concerned with exploring new worlds as fantasy is, it is hardly surprising that the map is a frequent complement to the texts, a companion on the reader’s journey through the alien landscape. The inclusion of maps is not restricted to fantasy novels, however, and it has been done for almost as long as there have been printed books. As a result of the Genevan reform, the second half of the sixteenth century saw a widespread use of maps in printed Bibles1 and already during the late fifteenth century, numerous illustrations charting Dante’s Hell had been produced, based on descriptions in the text.2 Other well-known works of fiction that are furnished with maps include Thomas More’s Utopia (1516), Jonathan Swift’s Gulliver’s Travels (1726), and Robert Louis Stevenson’s Treasure Island (1883)—of the last, it has even been said that the map was not just produced to be published together with the novel; the novel was actually written to go with the map.3
In modern fantasy, especially high fantasy, maps are considered common enough to be almost obligatory, mainly because of the maps J. R. R. Tolkien included in The Lord of the Rings (1954–55).4 According to Roz Kaveney, the map has come to be used as an authenticating device and a means to facilitate understanding, but she also suggests that Tolkien supplied maps “much in the same spirit that he provided endless glossaries and appendices.”5 What this spirit is, she omits to mention; but it is clear that to Tolkien, the map aids in the construction of an internally consistent world. Furthermore, the maps in The Lord of the Rings, just like the many casual references to Arda’s historical events and people, serve to provide the secondary world with the width, depth, and height that Tolkien sought in the realm of fairy stories. Whether provided for authentication, understanding, inner consistency, or world expansion, maps are expected to be supplied in high-fantasy novels today.
PREVIOUS EXPLORATIONS OF FANTASY MAPS
Fantasy maps have long fascinated readers. To my mind, the most impressive collection is still J. B. Post’s An Atlas of Fantasy, first published in 1973 (with a revised second edition in 1979).6 Post included more than a hundred maps from modern fantasy worlds as well as from a large range of imaginary places. Few scholars have discussed fantasy maps in writing, however. In 1976, Diane Duane wrote a short piece entitled “Cartography for Other Worlds: A Short Look at a Neglected Subject,” wherein she acknowledged the value of maps and insisted that fantasy and science-fiction maps should be created by author and mapmaker in collaboration.7 A year later, Julian May, writing as Lee N. Falconer, examined the history and cartography of the maps of the fictional world of Conan, including details about map projections, in the preface to A Gazet[t]eer of the Hyborian World of Conan.8 The most in-depth description of fantasy maps can be found in Diana Wynne Jones’s satirical commentary on the typical worlds of Tolkien’s epigones, The Tough Guide to Fantasyland, in which she declares that the first thing to do on any tour of Fantasyland is to “[f]ind the map. It will be there. No Tour of Fantasyland is complete without one.”9 The reader is then encouraged to examine the map, and a number of common features of fantasy maps are noted. Apart from Jones’s satire, few studies of fantasy maps have been more than cursory, despite the alleged prevalence and uniformity of the phenomenon.
The importance of maps has not passed the scholarly community by completely, however. In 1979, at the beginning of the mass-market fantasy boom that started in the late 1970s, Frank W. Day points out how the increased popularity of science fiction and fantasy, and the use of maps in these genres, “[makes] analysis of the map communication process in such literature more important than ever.”10 His own study focuses on maps as communicative devices, from the perspective of readers and authors. Coming to maps from a somewhat different angle, Clare Ranson ends her 1996 paper “Cartography in Children’s Literature” by expressing the hope that she has demonstrated how maps are “an interesting branch of illustration and worthy of more critical attention.”11 Yet despite Day’s and Ranson’s exhortations, few critics have devoted any notable scholarly efforts to fictional maps—and in all fairness, the fact that Day’s study is an unpublished master’s thesis and Ranson’s paper was presented at a conference on school librarianship probably did not add to their impact among fantasy scholars. Ranson only incorporates minimal readings of maps, including brief comments on the map in Ursula K. Le Guin’s A Wizard of Earthsea (1968). Instead, she proposes a taxonomy of maps in children’s books, with three categories, comprising maps of real places, real-place maps adjusted to fit the plot, and “imaginary maps.”12 Regrettably, this last category includes both maps of secondary worlds and maps of fictional places in the primary world, and thus Ranson’s taxonomy provides little help in advancing a closer investigation of secondary-world maps. R. C. Walker similarly devotes much of his Mythlore piece from 1981, “The Cartography of Fantasy,” to a general taxonomy of fantasy settings, noting the importance of settings to the genre.13 The article proceeds to stress the need for maps in fantasy books, proposing that maps be drawn for books that do not have them.14
The past three decades have seen a few articles that treat fantasy maps as cartographic objects. Peter Hunt connects maps with landscapes and journeys, bringing up the significance of maps in English fantasy—maps that, he argues, “are both reductive and suggestive.” Fantasy maps, according to Hunt, “stabilize the fantasy, while releasing greater imaginative potential.”15 His discussion does not differentiate between maps of primary and secondary worlds; in fact, his focus is on primary-world maps, and he notes that the maps might be said to “symbolize the tension that exists for the writer between the real landscape and the fantasy which inhabits it.”16 That tension is important to Hunt’s readings of the fictionalized versions of the English landscape, and the map becomes, to him, a tool for tapping into “landscapes of profound national symbolism.”17 The Middle-earth map is thus discussed in terms of how the secondary world it describes can be matched with the English landscape. The problem is that Hunt glides between map and setting in his discussion, so while many of the landscapes described in The Lord of the Rings might fit Hunt’s vision of Englishness, the map of Middle-earth itself does not.
Where Hunt focuses on maps in English—primarily low—fantasy, Myles Balfe looks at genre fantasy from an Orientalist perspective. He uses the map from Robert Jordan’s Wheel of Time series (1990–2013) to demonstrate how, rather than portraying a completely imaginary landscape, it “continues a Western historical convention of representing the ‘Orient’ and the ‘Orientals’ who exist there as both opposite and inferior to the West, and its Westland heroes.”18 Like Hunt, however, Balfe discusses both setting and its (re)presentation, somewhat undermining any cartographic point he tries to make, and he leaves maps out of his discussions of Feist’s Magician (1982) and the Dungeons & Dragons setting Al Qadim (1992).
Pierre Jourde dedicates a chapter of his study of imaginary geographies to cartographic representations. After a brief discussion of how map relations can be read (using the allegorical “Carte de Tendre” from Madeleine de Scudéry’s Clélie [1654–60] and the map of Utopia from More’s work as examples), Jourde focuses on the maps from The Hobbit (1937) and The Lord of the Rings.19 He sees two basic divisions in Middle-earth’s geography. The first is that between east and west, roughly along the central mountain range. The western regions are, according to Jourde, the more civilized, with a great number of “microcosms” and innumerable rivers, a land open to the western sea. The regions to the east are more corrupted by the forces of evil, with vast, undefined areas, open to an unknown continent from which Sauron’s troops come.20 Jourde also sees a second division between northern and southern Middle-earth, observing how the northern communities are more “rudimentary” than their southern counterparts and comparing the elven realms of Mirkwood and Lothlórien, the human lands of Esgaroth, Rohan, and Gondor, and even Sauron’s two guises, as the “diffuse necromancer” of Dol Goldur and the “quasi-omniscient potentate” of Barad-Dur.21 Jourde’s map readings offer some valuable insights, although his search for patterns sometimes makes him ignore those map features that do not fit, such as the uncivilized interior of Eriador; his desire to support his argument also leads him to overstate some map features in his outlines.22
The most extensive study to date of fantasy maps as cartographic objects is presented by Deirdre F. Baker. She carries out what she refers to as “a casual survey” based on a convenience sample of fantasy maps.23 Her findings confirm the similarities Jones satirizes in the Tough Guide. Regrettably, given that Baker’s stated purpose promises much more, her actual reading of the maps turns out to be rather shallow, and she forces the Middle-earth map from The Lord of the Rings into an allegory based on “what we know of Tolkien.”24 To suggest that the map illustrates the threat of Nazi Germany by virtue of its physical layout feels like a somewhat outdated interpretation of Tolkien. Also regrettable is that when Baker identifies the Earthsea map as intriguingly different from the other maps she has examined, she turns from the physical map to what she terms “metaphysical” maps. The unique vision of Le Guin’s map is coupled with a “mapping” (in the text of the novels) of Earthsea’s spiritual world, according to Baker, and she eventually suggests that “the sameness of geographical layout determines a sameness in simplistic moral or metaphysical vision.”25 Unconventional maps and originality in plot and metaphysics go together, she argues, yet while her discussions of a number of fantasy works are not without merit, they do not necessarily substantiate that conclusion.
Ricardo Padrón investigates fantasy maps together with other types of maps of imaginary worlds, offering a wider consideration of cartographic objects that map fictional geographies. He discusses the functions of a wide array of maps, from Dante’s Hell to William Faulkner’s Yoknapatawpha County, even including examples from the visual arts, such as Adrian Leskiw’s road maps of imaginary places. Given such a comprehensive collection of examples, it is not surprising that Padrón includes Tolkien’s map of Middle-earth along with user-created maps of the online computer game World of Warcraft (2004–present) and maps of the land of fairy tales by Jaro Hess and Bernard Sleigh. As opposed to the maps of Utopia and Gulliver’s travel locations, these various fantasy maps all invite us to revel in imaginative travel.26 Padrón’s analysis, while on the whole interesting, is occasionally so brief as to suggest that he bases his discussion on a personal view of fantasy as escapist fiction, an impression made especially strong when he fails to support his arguments.27 His conclusion is that the maps he has investigated are not radically different from other maps as maps, and that they work because of their similarity with those other maps. It is the imaginary worlds of the maps that trigger our imagination, but that is something any map can do—if we let it.28
Among the many recent online essays and blog posts that wax lyrical about fantasy and imaginary maps, I have found one example worth mentioning here. In his nicely argued and beautifully illustrated piece “Here Be Cartographers: Reading the Fantasy Map,”29 Nicholas Tam sketches out a theory of how to analyze fantasy maps. He proposes a number of angles from which to approach a map, based on ways in which the fantasy map can relate to the story and the fictive world. Tam’s discussion ranges from material conditions of mapmaking to worldviews encoded in maps, and his examples make clear why the various angles are worth considering in theoretical investigations of the genre’s numerous maps. This is a brief essay rather than a scholarly article, lacking theoretical depth, but it provides a useful starting point for anyone who wants to think critically about fantasy maps.
Over the past four decades, many scholars have argued that fantasy maps deserve more critical attention, but what little scholarship has been published has either offered only a cursory exploration or too narrow a scope to provide any deeper insight into the subject. A greater understanding of the genre’s maps requires both a more comprehensive study of a large number of maps and more thorough examinations of particular maps. It is also important to bear in mind that Padrón’s point—that maps of imaginary worlds are similar to other maps—implies that fantasy maps can be analyzed in the same way as maps of our world. However, any textbook in cartography will quickly reveal a difference between the two types of maps that is significant enough to call into doubt whether fantasy maps are maps at all, a topic next up for consideration.
WHAT IS A FANTASY MAP?
A map is a symbolized representation of geographical reality, representing selected features or characteristics, resulting from the creative effort of its author’s execution of choices, and is designed for use when spatial relationships are of primary relevance.30
If this is how a map is to be defined, the maplike illustrations found in a vast number of fantasy works present a problem. Although they undeniably “[result] from the creative effort of [their authors’] execution of choices,” an overwhelming majority of them are not representations of “geographical reality.” Their “features or characteristics” bear little or no relation to anything in the actual world. They are simply not maps, as they violate the deeply ingrained notion that a map must in some way represent the world of the cartographer. Even the most concise definition I have found, that of American cartographers Arthur H. Robinson and Barbara Bartz Petchenik, ultimately falls back on this notion. To them a “map is a graphic representation of the milieu,” wherein the word milieu “connotes one’s surroundings or environment in addition to its meaning of place.”31 Cartographically, a fantasy map seems to be a contradiction in terms.
Actually, Robinson and Petchenik seem to have no problems with maps of imaginary places, as they include them among their examples.32 Historian Jeremy Black brings up the problem of mapping politics in maps of imaginary worlds, but he never questions their status as maps.33 In his influential book The Power of Maps, map scholar Denis Wood similarly acknowledges the existence of fictional and fantastic maps.34 In a later book, Wood discusses maps of imaginary places (mentioning, for instance, the maps of Middle-earth, Dungeons & Dragons, and the Marvel Universe) at some length, concluding that they illustrate how maps need not represent a part of the Earth’s surface.35 Consequently, fantasy maps are treated as maps here, with one important caveat. The difference between the map that graphically represents the milieu and a map of an imaginary place is one of priority: a map re-presents what is already there; a fictional map is often primary—to create the map means, largely, to create the world of the map. (This is the case even though maps that are made to fit an existing literary work are by no means uncommon. Padrón includes the many maps of Hell that are based on Dante’s Divina Commedia as examples of this phenomenon.36) Some points, then, on terminology before proceeding: First, the maps in the fantasy novels to be examined are not maps in the sense that they necessarily correspond to anything in the actual world. Drawing inspiration from Wood’s terms, I refer to fictional maps and fantasy maps, where the former category includes any map that does not represent the actual world and the latter is a map of a fantasy world, generally found in a fantasy novel (although the term would fit maps from fantasy role-playing games equally well). Maps of the actual world are consequently actual maps. Second, to avoid suggesting that fictional maps in any way correspond to a position in the actual world, I say that they portray rather than represent something.
The status of the fictional map in relation to the text is in no way clear-cut, and various points of view yield different insights. In Paratexts: Thresholds of Interpretation, Gérard Genette uses the term paratext to refer to the various “verbal or other productions, such as an author’s name, a title, a preface, illustrations” that accompany a text.37 Whether they belong to the text or not, he explains, these productions surround and extend it. Genette mainly considers textual paratexts, but in his conclusion, he mentions some paratextual elements that he has not examined. Among these are “certain elements of the documentary paratext that are characteristic of didactic works” but that sometimes appear in works of fiction.38 Genette’s examples of such elements include fictional maps, for instance Faulkner’s map of Yoknapatawpha County and Umberto Eco’s plan of the abbey in The Name of the Rose (1980).
It makes sense to regard a fantasy map as something that extends the fantasy text. The maps are generally not part of the narrative, in that they do not refer directly to any object in the text (with a few exceptions, such as Thror’s Map in The Hobbit, which is also parenthetically referred to by the narrator39). Instead, they and the text both refer to the fictional world or to a part of it—that is, to the story’s setting. Thus, they become, in Genette’s terminology, a threshold,40 a liminal space between the actual world of the reader and the fictional (generally secondary; see Table 2.2) world of the fantasy story. The maps blur the distinction between representation and imagination, suggesting that the places portrayed are in fact representations of existing places. This suggestion would explain why fictional maps are usually considered to be maps, even though they do not share the actual maps’ representing of the “milieu.”
Alternatively, fantasy maps can be interpreted as docemes. A doceme is defined by documentation-studies scholar Niels Windfeld Lund as “any part of a document, anything that can be identified and isolated analytically as part of the documentation process or the resulting document.” Moreover, Lund explains that although, for instance, a photograph can be a document in itself, if it is part of a newspaper article, it (as well as the article text) is only a doceme—“[a] doceme can never be something in itself.”41
Thinking of the fantasy map as a doceme puts a stronger emphasis on the relationship between narrative (text) and map. Rather than offering a threshold between fiction and reader, the map is part of the total fantasy document. Lund’s observation that the doceme is part of the documentation process or the resulting document is also highly relevant. The map is not only one of several parts of the finished document; it can, as in the case of Stevenson’s Treasure Island, be a central part of the creation process: according to Orson Scott Card, maps are basic to Card’s world creation and provide him with story ideas;42 Poul Anderson explains how, “[w]hen a story has an imaginary setting, I draw a map as part of the planning”;43 and in a letter, Tolkien relates how he “wisely started with a map, and made the story fit (generally with meticulous care for distances).”44 Three authors do not, of course, represent a genre, but they all point in the same direction, and Tolkien proceeds to explain why the map’s priority is important: “The other way about lands one in confusions and impossibilities, and in any case it is weary work to compose a map from a story.” In the same letter, he also apologizes for “the Geography” and acknowledges how difficult it must have been to read The Lord of the Rings proofs without maps. In other words, Tolkien felt the map doceme to be necessary both during the documentation process and when reading the resulting document.
In the two discussions on maps that follow, we can see how fantasy maps can be fruitfully interpreted as both paratexts and docemes. The two perspectives are not mutually exclusive but relate the map differently to the text. Constituting thresholds between the actual and fictional worlds, they are also an essential part of the document that is the fantasy book.
A SURVEY OF FANTASY MAPS
To take a quantitative approach to the fantasy map, I carried out a survey of two hundred randomly selected fantasy works and looked at the maps I found in them. This section is devoted to a presentation of the survey results, examining the maps’ general features as well as the types of map elements found, and discussing what these findings can tell us about specific settings and the genre as a whole. The small number of works in the sample, and the low proportion of maps among those works, resulted in fairly large margins of error, but the results still indicate some interesting features of fantasy maps. (A more thorough presentation of the survey method and the assumptions on which it is based, along with the statistical method used for calculating margins of error, can be found in appendix A; a list of works in the sample can be found in appendix B.)
The Prevalence of Maps
Here is the most basic question: how common is it for fantasy novels to contain at least one map? Of the two hundred novels in the sample, sixty-seven (34 percent) contained one or more maps. In terms of the entire genre, this means that no more than 40 percent and possibly as few as 27 percent of all fantasy novels actually contain any maps. In other words, maps are not the compulsory ingredients they are widely held to be. The main explanation for this apparent lack of maps has to do with setting. The sampling frame (and thus the sample) comprises the entire genre, high fantasy as well as low, but maps are much more common in fantasy set in a secondary world. In the sample, of the sixty-seven novels that have maps, only six contain maps portraying the primary world (corresponding to 3 to 18 percent of all novels with maps), and these are all set in historic or prehistoric times. While there are low-fantasy works in contemporary settings that have maps, their numbers are small enough not to crop up in the sample (constituting fewer than 5 percent of all maps and less than 2 percent of the genre). Examples include Alan Garner’s The Weirdstone of Brisingamen (1960), with a map of the area around Macclesfield in Cheshire; Neil Gaiman’s Neverwhere (1996), with a map of the London Underground; and Charlie Fletcher’s Stoneheart (2006), with a map of central London.45 As I had no data on the general distribution of high fantasy to low in the sample, I was unable to pursue this matter further (see also appendix A).
2.1. HOW MANY MAPS DO FANTASY NOVELS CONTAIN?
N = 67
Related to the question of how prevalent maps are in fantasy novels is the question of how many maps a fantasy novel contains. While three quarters of the novels with maps contained only a single map (roughly between 20 and 30 percent of the genre), slightly over one fifth had two maps, and novels with three, four, and six maps also appeared (see Table 2.1). In most cases (fourteen out of eighteen), the additional maps provided one or more large-scale views of one or more areas (in four cases, this included a city map). Of the remaining four cases, two had floor plans for buildings, one had maps of two different continents, and one mapped the area’s political and physical features on two separate maps. The main reason for including more than one map, in other words, seems to be to provide a general, small-scale map and a larger-scale map of an important setting: a country, province, or city, for instance. The maps in The Lord of the Rings provide an example of this tendency. A large-scale map of the Shire, a small-scale map of the entire western Middle-earth, and a medium-scale map of the area around Gondor and Mordor are included, illustrating how the story’s quest-narrative is built around long journeys across the world, but also requires detailed maps for locales where central events are set. (Tolkien’s large-and small-scale maps are discussed in detail later in this chapter.)
When examining the features of the fantasy map, I use as my sample the ninety-two maps found in the survey. In five cases, the same map (or similar versions of the same map, when, for instance, different artists have drawn maps for books with the same setting) appears in two or more books. Such maps have been counted as separate instances, however, partly because the efforts of different mapmakers may result in very different renditions of the same location (as can be observed from the maps in the Conan books in the sample46), but mainly because they constitute docemes in different documents.
General Map Features:
Subject, Orientation, Surround Elements
Only six of the novels with maps portray a primary-world setting. These six novels contain thirteen of the ninety-two maps, and even though there are fantasy books that have maps of primary-world cities, no such maps can be found in the sample, confirming how rare such maps are (corresponding to less than 4 percent of all fantasy maps). The vast majority of maps portray a secondary world or an area in a secondary world (almost four fifths of the sample, or between 68 and 86 percent of all fantasy maps). Of the remaining maps, about 5 percent (2 to 12 percent) portray an imaginary city, either in a secondary world or set in the primary world (an example of the latter is the city of Ys, a map of which appears in Poul and Karen Anderson’s Dahut [1988]), and 2 percent (0.3 to 8 percent) are plans of buildings or building complexes.
The portrayal by at least two thirds of all fantasy maps of secondary worlds suggests a need in fantasy novels to provide a visual image of the imaginary setting, but also to provide the setting with some sort of structure. The map can help the reader understand complex spatial relationships that the text alone may fail to convey. Conversely, the absence of a map in a fantasy novel in which movements, positions, and spatial relationships in the imaginary world are central to the story may prove bewildering to the reader.47
With maps of secondary worlds clearly dominating, the question arises to what extent these maps of alien worlds also reflect alien forms of mapmaking. In The Hobbit, Tolkien includes two maps, one of which has THROR’S MAP written in the lower left corner and includes some text in dwarvish runes. It is obviously meant to refer to a map in the story, and the brief preface points out that the map has “East at the top, as usual in dwarf-maps.”48 The other Hobbit map, that of the Wilderlands, follows the convention of having north at the top. This convention is comparatively modern, however. According to historian P. D. A. Harvey, almost all world maps before the fifteenth century were either zonal (or climatic) maps or T-O (orbis terrarium) maps, circular maps with east at the top where the world is divided into three continents (Asia, Europe, Africa) by a T-shape.49 In fact, of the extant medieval maps from the eighth through the fifteenth centuries, a clear majority are basic T-O maps.50
2.2. MAIN SUBJECT OF MAPS
| % of Maps in Sample (n) | % of All Fantasy Maps | |
| Primary World | 14.1 (13) | 7.7–23.0 |
| Secondary World | 78.3 (72) | 68.4–86.2 |
| Imaginary City | 5.4 (5) | 1.8–12.2 |
| Building/s | 2.2 (2) | 0.3–7.6 |
N = 92
Despite the dominance of secondary worlds or historical settings in the sample, the maps largely follow the modern convention of placing north at the top. Ann Swinfen remarks on how (northern-hemisphere) fantasy writers maintain primary-world compass directions in their worlds.51 A majority of all fantasy maps—at least 58 percent—come with a compass rose or similar design indicating which way north is. Other maps signal their orientation in other ways; for example, Southern Ithania is located below Northern Ithania in Trudy Canavan’s Last of the Wilds (2005). For only nine maps in the sample can orientation not be determined from the map alone. Of the eighty-three maps for which orientation can be determined from the map, nine are not oriented with north at the top, but all are oriented so as to have a direction somewhere between northeast and northwest at the top. In the nine cases for which information about orientation is completely absent from the map, re-lated maps or the texts in question have been used to work out how they are facing. North is at the top of all these maps except for one, which has north-northeast at the top.
2.3. MAP ORIENTATION
| % of Maps in Sample (n) | % of All Fantasy Maps | |
| N | 80.4(74) | 70.9–88.0 |
| NE to NW | 9.8(9) | 4.6–17.8 |
| No Orientation Given | 9.8(9) | 4.6–17.8 |
| Compass Rose | 68.5(63) | 58.0–77.8 |
N = 92
So although Thror’s Map demonstrates the existence of fantasy maps that are oriented differently, such maps constitute less than 4 percent of all maps. Only very few fantasy writers or mapmakers avail themselves of the freedom to turn the map whichever way according to the conventions of imaginary societies in secondary worlds, or to create completely new directions. Instead, just as Swinfen suggests, the actual-world convention of orienting the map with north at the top dominates almost completely.
A typical feature of the T-O maps, a feature that, according to John Noble Wilford, goes back to the earliest extant world map (a Babylonian map from the sixth century B.C.),52 is that the “whole [world] is surrounded by a circumfluent ocean.”53 The surrounding water is where the world ends, where even the possibility of knowledge ends. It frames the known world, establishing that what is on the map is all there is. Similar circumfluent oceans can be found on fantasy maps, providing the worlds with what John Clute, in The Encyclopedia of Fantasy, calls water margins. According to Clute, these margins surround the central land or reality and fade away beyond the edges of any map. He adds that secondary worlds often have maps whose edges are water margins.54
Such “ultimately unmappable regions”55 that completely enclose the known world of a fantasy map within regions of the unknown are not, in fact, particularly common; only a small proportion of the fantasy maps are completely surrounded by water margins (9 to 24 percent; see Table 2.4). On at least three maps out of four, land stretches to the edge of the map, suggesting that the world not only continues but is accessible. The term water margin is also slightly misleading; a water margin need not consist of water, but can be a region of any “endless” or “impassable” terrain type. Of the fourteen cases in the sample, two maps have water margins that are not water. In Terry Brooks’s The Tangle Box (1994), the land of Landover is surrounded by a mountain range beyond which are “Mists and the Fairy World.” Brooks’s fairy world, it is explained in the first Landover novel, Magic Kingdom for Sale/Sold! (1986), is a numinous place that borders on all worlds. It can be traversed but only with the help of magic; it cannot be mapped, it is unknowable.56 In Martin Gardner’s Visitors from Oz (1998), one of many late additions to L. Frank Baum’s classic Oz books, the land of Oz is surrounded by an “impassable desert” (also described as “shifting sands,” “great sandy waste,” and “deadly desert”).57 While obviously meant to emphasize the futility of any attempt to leave Oz by nonmagical means, other Oz books (for instance, the third book in the sequence, Ozma of Oz [1907]) allow for such journeys and open up a world beyond the land of Oz, which illustrates how water margins can be breached, the unknowable made knowable—and known.
The notion that a water margin must enclose the world completely is not unproblematic. On many maps from the sample, there is some land at the edge of the map, providing the possibility of larger landmasses unaccounted for by the mapped area. An example from the sample is Michelle M. Welch’s Chasing Fire (2005), where a continent surrounded by water takes up most of the map but where a small part of another landmass (Ikinda) can be found along a section of its southern edge. It is obviously the continent that is the map’s focus: it features mountains, rivers, and a lake, as well as political borders and various locations (mostly towns, presumably, but names such as Mt. Alaz, Seven Oaks, and Naniantemple suggest that other types of places may also be included). What can be seen of Ikinda, on the other hand, is completely empty. This might mean that it is featureless, unexplored, or simply irrelevant to the story. On the map, Ikinda is portrayed as unknown but knowable, not quite part of the water margin but almost. Welch’s Ikinda is hence truly marginalized—not only pushed to (and beyond) the edge of the map but also empty—whereas the map’s central continent has a variety of features. Yet setting out to explore the parts of Ikinda that lie beyond the map edges is certainly possible, and such exploration might reveal a small island or a vast continent. We cannot tell which from the map, yet we know that there is something there.
Even when there is a circumfluent ocean, however, the world can be opened up. Expeditions into the water margin have been undertaken in numerous fantasy works. In, for instance, Stephen R. Donaldson’s Second Chronicle of Thomas Covenant series (1980–83), the Mallorean series (1987–91) by David and Leigh Eddings, and Raymond E. Feist’s The King’s Buccaneer (1992), sea voyages off the map turn up new continents. On the other hand, as the Landover and Oz maps have already demonstrated, terrains other than water can create an unmappable, unknowable region at a map’s edge. The map in Gail Dayton’s The Barbed Rose (2006) places the central land of Adara between sea to the east and west, whereas to the north and south there are mountains that give the impression of being impassable. The mountainous northern isthmus is called “The Devil’s Neck,” a name that emphasizes just how impassable it really is. While this is not a complete margin in a strict sense, the map still makes clear that Adara is where the story takes place, and the reader who ponders what can be found beyond the mountains is left with a vague suspicion that there will only be more mountains. (Mountains as map elements will be discussed further in the text that follows.)
Even if the landmass on the map is fully surrounded by water, this does not necessarily constitute a water margin in Clute’s sense. The map of the Isles of Glory (map 2.1) in Glenda Larke’s Gilfeather (2004) may set its islands in a surrounding ocean, but comments on the map make plain that in the diegesis, there is no unknown world to discover beyond the edges of the map. Instead, the islands themselves have been discovered—and, it is implied, fairly recently at that. Not only does the map tell us that the isles were “[s]urveyed by the 2nd Explor. / Ex. Kells 1782–1784,” but it also notes by whom and when the various parts of the archipelago were discovered. Known and unknown are turned around here, the well-known residing off the map. Wherever the political, financial, and cultural centers are to be found in this secondary world, they belong in the regions beyond the map’s margin—the Isles of Glory, in the middle of the map, are part of the world’s periphery.
The absence of complete water margins, with at least some land reaching all the way to the map’s edge, indicates that a map does not portray an entire world. Even when the map gives the impression of portraying the whole world, such as the map of Earthsea in Le Guin’s A Wizard of Earthsea, an extremely short land border still suggests that there is more to the world than this. It also implies that the region on the map is located somewhere in the larger world.
MAP 2.1. The Isles of Glory from Glenda Larke’s Gilfeather (2004).
Copyright Perdita Phillips, www.perditaphillips.com.
Of the ninety-two maps in the sample, only one third are clearly set in the northern or southern hemisphere; for the rest, this distinction could not be determined from the maps. Obviously, a secondary world does not have to be set in a hemisphere; Terry Pratchett has demonstrated with his Discworld novels that a world shaped like a disc works as a setting. The change in shape also led to his abandoning the traditional compass points. Instead, the Disc has the main directions “hubward” and “rimward,” and the lesser directions “turnwise” and “widdershins.”58 Of the thirty-one maps for which the hemisphere can be determined, twenty-five are set in the northern hemisphere, five in the southern, and one included both hemispheres.
In other words, the northern hemisphere is clearly more common, even when the margin of error is taken into account. One possible explanation is, of course, the preponderance of writers from the northern hemisphere. Despite a growing number of Australian and New Zealand fantasy writers, the vast majority of fantasy writers in English come from the United States and Great Britain. Yet the five maps of a southern hemisphere actually come from three novels by three different writers (Ian Irvine, Sherwood Smith, and Harry Turtledove), with only Irvine hailing from the Antipodes.59 Another possible interpretation would be that the cultural and political bias toward the northern hemisphere that we find in the actual world rubs off on the creation of secondary worlds and the maps that portray them.
2.4. SURROUND ELEMENTS
| % of Maps in Sample (n) | % of All Fantasy Maps | |
| Water Margins | 15.2(14) | 8.6–24.2 |
| Map Projection | 5.4(5) | 1.8–12.2 |
| Legend | 23.9(22) | 15.6–33.9 |
| Scale | 16.3(15) | 9.4–25.5 |
N = 92
Still, at least half—possibly as many as three quarters—of all fantasy maps lack information about where on the globe they are situated. They may not even be situated on a globe at all. To the extent that they are, or are meant to be, most maps also lack any information about what method has been used to project the spherical surface to a plane (see Table 2.4). This can seem like a minor point in relation to maps of imaginary places; the ideological implications of various map projections may carry political relevance and interest in the actual world,60 but do such things matter in a fantasy map? An interesting case is provided by the maps created by author and cartographer Russell Kirkpatrick for his fantasy novels, one of which, The Right Hand of God (2005), is included in the sample. There are four maps in The Right Hand of God: a small-scale map titled “The Sixteen Kingdoms of Faltha”; a medium-scale map of “Westrau, Straux”; an untitled, large-scale map of a mountain pass called The Gap; and a city plan of “Instruere and Environs.” Of these, only the Sixteen Kingdoms and Instruere maps contain any information about their map projections. At the bottom of the Instruere map, a note says, “Nestor’s Equal-Area Projection © University of Instruere.” The Sixteen Kingdoms map lacks explicit information about what projection is used, but has a graticule or “web” of latitude and longitude lines that shows the curvature of the surface (see map 2.2). The scale is small enough to let the map reader perceive how the northern parts bend toward the map’s vertical axis. This bending causes the squares of the graticule to elongate toward the pole, giving an impression of the mapped area’s position between pole and equator. Although no degrees are given, the curves of the longitude lines lead the eye to the projection’s central meridian, the one line of longitude that is running parallel to the map’s vertical axis. Near the map’s center, where the central meridian crosses a latitude line, sits the city of Instruere, its importance emphasized by the map projection.
2.5. HEMISPHERE
| % of Maps in Sample (n) | % of All Fantasy Maps | |
| Northern | 27.2(25) | 18.4–37.4 |
| Southern | 5.4(5) | 1.8–12.2 |
| Both | 1.1(1) | 0.03–5.9 |
| No Data Available | 66.3(61) | 55.7–75.8 |
N = 92
While the importance of Instruere is made clear by the city’s location on, as well as the projection of, the Sixteen Kingdoms map, the note about projection and copyright at the bottom of the Instruere map has a different function. By copyrighting the map to an entity in the secondary world, the map claims itself to be part of the same (fictional) world as that of the city it portrays. The map is no longer an overview of a fictional world; just like Thror’s Map, it becomes part of that world, a product of it. This effect is subtly reinforced by the comment about its projection. Nestor’s Equal-Area Projection, while cartographically valid, is turned into a product of the secondary world as a result of being attributed to a fictional cartographer.61 When the map is commandeered by the fictional world, as in this case, it resists a paratextual perspective, stressing instead how it should be considered a doceme, a part that, together with the text, makes up a greater whole. Examining the map is not a question of entering the story; the fictional world has already been entered.
MAP 2.2. Graticule extracted from the map “The Sixteen Kingdoms of Faltha” in Russell Kirkpatrick’s The Right Hand of God (2005). Adapted from the original by the author.
Of all the maps in the sample, only five have any sort of information about map projection, meaning that of fifty fantasy maps, between one and six maps say anything about projection (Table 2.4). Apart from Kirkpatrick’s Instruere map, only the map of “The World” in Louise Cooper’s Aisling (1994) mentions what projection is used; of the other two maps, one has a graticule while the second has longitudinal degrees running along its left side. On Cooper’s map, the Winkel Tripel projection has been used. Unlike Kirkpatrick’s Nestor projection, the Winkel Tripel projection is an actual map projection, used to represent the three-dimensional actual world in two dimensions; linking it to the map blurs the map’s status as fictional. This emphasizes the map’s nature as paratext rather than, as in the Kirkpatrick case, anchoring the map in the fiction as a doceme.
In his oft-cited discussion of the North Carolina highway map, Denis Wood forcefully denies that legends or keys would be “naturally indispensable to most maps, since they provide the explanations of the various symbols used,”62 and fantasy maps appear to be no exception. No more than a third of all fantasy maps have legends or keys of any sort (see Table 2.4), and of these, none explains all the symbols used. In fact, only one third of the map legends in the sample (seven legends in total) include any symbols for terrain features. Instead, the legends comment on the map and the fictional world. Wood suggests that “the role of the legend is less to elucidate the ‘meaning’ of this or that map element than to function as a sign in its own right,”63 something that is quite obvious from the legends on fantasy maps. These clarify not what separate map symbols mean but what is important about the map; in some cases, they are little more than translation keys that aid in the reader’s understanding of where events take place (for instance, the lists on the maps in Dahut, which offer translations between contemporary place-names and the names of those places in Roman times). Other legends contain symbols for the travel routes of the protagonists, placing the journey through the world at the center (for example, Orson Scott Card’s Seventh Son [1987] and Linda Lay Shuler’s Let the Drum Speak [1996]). Even when the map legend only—or nearly only—contains signs for various terrain features, there is a focus, a comment on the map. Two brief examples:
The legend on Ian Irvine’s Geomancer (2001) map (map 2.3) lists a number of terrain features, including separate symbols for Conifer Forest, Broadleaf Forest, and Tropical Forest. The legend conveys the impression of a world defined by an abundance of terrain types, a world whose climate ranges from temperate or cold (mountains, hills, coniferous forest) to tropical or warm (reef, desert, tropical forest), with landscapes ranging from dry (desert) to wet (marsh/swamp) and with a variety of landforms (from mountains to grassland). Landscape is what this place is all about, the legend says, adding, almost as an afterthought, that there are also people: a symbol for Main Road (a dotted line) crops up below all the terrain elements, although no road can be found on the maps.
On the map in The Burning City (2000), by Larry Niven and Jerry Pournelle, the legend similarly brings terrain features into focus. The first map element (topmost in the left-hand column) is an icon of a coniferous tree that refers to Redwoods. The importance of redwood trees as opposed to other trees is emphasized on at least three levels: their relative position (the Redwood sign precedes the sign for other trees); their being singled out, as opposed to being part of a group (Redwoods versus Other Forest, or all trees that are not redwoods); and the use of an icon rather than a symbol. The legend also stresses the rough terrain: there is a single element for Cultivated Fields and seven that refer to types of wilderness (the legend has no signs for houses or roads, although these signs appear on the map). No fewer than three of the wilderness signs have to do with elevated land: Hills, Mountains, and High Mountains. According to the legend, this is a wild landscape where people are present but marginalized, a landscape dominated by tall trees and mountains.
MAP 2.3. Map from Ian Irvine’s Geomancer (2001).
Copyright Ian Irvine, 2001.
A map element that offers a different kind of comment on the map and what it portrays is the scale. As with legends, this is a relatively rare element, which can be found on fifteen of the sample’s ninety-two maps (between a tenth and a quarter of all maps; see Table 2.4). In general, a scale allows you to calculate the area covered by the map. With such a low incidence of scales, little can be said about the actual areas that fantasy maps cover. The values are spread fairly evenly, without any noticeable clusters that would suggest typical area sizes. Excluding a city map that covers about 70 square kilometers, the maps that have scales range in size from about 1,100 square kilometers to 27 million square kilometers (approximately the size of Hong Kong and North America, respectively). In other words, they are all significantly smaller than Earth, a fact that, like the prevalence of land borders, hints at a larger, unmapped world beyond the map’s edges.
Regardless of map area, the presence of a scale suggests a particular control of the world. It hints at the “meticulous care for distances” that Tolkien mentions in his letter, but even more, it implies that this is a world that can be measured. A scale, like information about projection, offers a way to understand positions in map space in terms of positions in “actual geographical” space; the scale on the fictional map announces that there is another space to which the map positions correspond, strengthening the impression that the map not only portrays but represents, that there is a measurable space to which the map refers. This measurability also suggests precision and control, thus either accentuating the author’s role as creator and the world’s status as fictional or (especially if the map is interpreted as a doceme and refers to a specific or implied map object in the fictional world) requiring a level of scientific knowledge that may be incompatible with the fantasy world as a whole.
Types of Map Elements
Table 2.6 lists the various types of map elements that occur in the sample maps. Significantly more common than any others are topographical elements,64 especially mountains, coastlines, and rivers (see Table 2.7) as well as various population centers, often without distinctions between villages, towns, and cities. In her Tough Guide satirical description of maps, Diana Wynne Jones plays on the perceived prevalence of these elements: “[The map] will show most of a continent (and sometimes part of another) with a large number of bays, offshore islands, an inland sea or so and a sprinkle of towns. There will be scribbly snakes that are probably rivers [.…] [The] empty inland parts will be sporadically peppered with little molehills.”65 While inland seas are not as common as Jones implies—large landlocked bodies of water appear on between one and two fantasy maps out of six—she still describes the most common types of map elements (see also Table 2.7). The occurrence of these types of elements in about nine maps out of ten in the sample (corresponding to 85 to 97 percent [topography] and 78–93 percent [towns] of all fantasy maps) is not surprising; on the majority of maps, the landmasses are at least separated from a surrounding sea by coastlines and also include some description of the landforms (two examples of topography). Rivers and mountains seem to be the basic way to define spatial relations; and—especially in portal–quest fantasies—they are both traversed frequently and with some difficulty. Because of the hardships involved in crossing (or failing to cross) mountains, these are often used to define the limits of the world; as Jones puts it, there will be “a whole line of molehills near the top [of the map] called ‘Great Northern Barrier.’ Above this will be various warnings of danger.”66 Such mountain ranges (with or without names and warnings) fulfill the same function as a water margin, although they suggest rather more strongly that land exists on the other side. Such ranges can be found, although without any “warnings of danger,” on 9 to 24 percent of all fantasy maps. (A number of other maps use a variety of wasteland in a similar fashion, as in the Oz map.)
2.6. TYPES OF MAPS ELEMENTS
| % of Maps in Sample (n)a | % of All Fantasy Maps | |
| Topographical | 92.4(85) | 85.0–96.9 |
| Population Centersb | 87.0(80) | 78.3–93.1 |
| Other Constructionsc | 67.4(62) | 56.8–76.8 |
| Politicald | 46.7(43) | 36.3–57.4 |
| Demographical/Zoologicale | 10.9(10) | 5.3–19.1 |
| Otherf | 10.9(10) | 5.3–19.1 |
N = 92
a Adds up to more than 100 percent, as each map may contain several types of elements
b Cities, towns, villages
c Roads, buildings, bridges, and other artificial constructions
d Political borders, seats of government, historical sites
e Peoples or creatures inhabiting certain areas
f Includes a wide range of elements, generally peculiar to particular works
2.7. THE FIVE MOST COMMON TOPOGRAPHICAL ELEMENTSa
| % of Maps in Sample (n) | % of All Fantasy Maps | |
| Rivers | 87.0(80) | 78.3–93.1 |
| Bays | 84.8(78) | 75.8–91.4 |
| Towns | 79.3(73) | 69.6–87.1 |
| Mountains | 75.0(69) | 64.9–83.4 |
| Islands | 75.0(69) | 64.9–83.4 |
N = 92
a The number of occurrences of an element on each map is not taken into consideration.
The group Other Constructions in Table 2.6 (found on two thirds of the maps) is dominated by roads, with only occasional buildings appearing (other than population centers: towers, fortresses, houses). There are rather fewer Ruins, Towers of Sorcery (or other centers of magic), and Dark Citadels (or other edifices of evil) than is implied by the Tough Guide (on eleven maps in the sample—somewhere between 6 and 20 percent of all fantasy maps), but Jones is right in pointing to the almost total absence of inns and rest stops. Not even when “camps” are included would this category be found on more than one fantasy map in seven (six maps in the sample).67 The roads that crawl all over the maps have two main functions. First, as possible travel routes, they tie together the distant points of the map, telling the reader that journeying through this fictional world is possible, even mandatory—the map is a traveling aid, a tool for exploration. This function is particularly prominent in, for instance, Baum’s The Wonderful Wizard of Oz (1900), in which the yellow brick road is Dorothy’s safe way to navigate the fantasy world; and, differently, in D. M. Cornish’s Monster-Blood Tattoo/The Foundling’s Tale series (2006–2010), in which the protagonist is trained to become a lamplighter, one who lights the lamps along the highways of a dark, monster-infested continent. In both cases, the road is meant to offer safe conduct through a dangerous, fantastic landscape. Second, though, not all fantasy journeys follow the roads; staying away from the road is often more important. The road itself can become a dangerous, unprotected place, open and unsafe. The focus shifts to the wild landscape around the road. The dangers of the road are particularly vivid in The Lord of the Rings; it turns into the realm of the enemy, and capture threatens whenever Frodo sets foot on a road. On some maps, the absence or scarcity of roads implies that the world is wild and uncivilized: the few roads that exist are not enough to tame the world, to tie its places together; instead, the world’s inhabitants are left exposed and vulnerable to the trackless wilderness.
The political elements mainly consist of political borders between countries or smaller administrative units (for instance, counties), and they are rather uncommon. At most, they appear on two thirds as many maps as do topographical element types. Still, they can be expected to occur on at least somewhat over a third of all fantasy maps. So, while the importance of political units to the fantasy world should not be discounted, it is clear that fantasy maps are more topographical than political.
To fantasy cartographers, it is apparently not particularly important to provide information about what people or creatures live where. In the sample, only 11 percent of the maps contained such information, meaning that as few as one in twenty and as many as one in five of all fantasy maps contain such information in some way. This is a significantly smaller proportion than any of the other map elements, apart from the miscellaneous element types in the Other category. Some of this information might actually be found in the Political category—a certain country could be inhabited by a certain people or type of creature, such as the hobbits in the Shire, or the frequent elven realms, such as Tolkien’s Lothlórien and Feist’s Elvandar (in Magician and other books set in Midkemia). Except for maps in tie-in novels to role-playing game settings (where there are races aplenty, spread over the maps), it is rare to find information such as the “Wood Elves” mark on Tolkien’s map of western Middle-earth. The lack of information about what people and creatures live where (“Here Be Elves” or “Here Be Dragons”) implies both that the variety of inhabitants found in a fantasy world should be taken as a matter of course and, simultaneously, that the world is, in fact, a terra incognita full of secrets, especially in the cases of portal–quest fantasies, in which the secondary world is explored by protagonist/s and reader together.68 Rather than being a paratext offering hints to the reader about what to expect once the fictional world is entered, the map in this respect is more of a doceme—a part of the world it describes—for which obvious information is not required, just as we do not often find (actual) general maps that specify “Here Be Germans” in a cursive script over Germany, or have “Here Be Reindeer” printed over northern Scandinavia.
Hill Signs
The signs for mountains offer a useful litmus test in relating the fictional map to maps of actual historical periods. Map projections deal with the transformation of the three-dimensional surface of a sphere to the two-dimensional surface of a plane, but mapmaking also requires another transformation of three-to two-dimensional surfaces. Although it is possible to imagine a world in which the surface is completely flat, most if not all settings tend to have an uneven landscape with hills and mountains as well as plains and valleys. According to geographer Norman J. W. Thrower, “[d]elineation of the continuous three-dimensional form of the land has always been one of the most challenging problems in cartography.”69 The sign for mountains is also among the most ubiquitous topographical elements on fantasy maps—it crops up on three quarters of the maps, which makes it as common as islands but less common than rivers, bays, and towns. (Taking the margin of error into account, it is actually not possible to say which of these five elements is the most prevalent. See Table 2.7.)
The wide variety of signs used for mountains and hills—hill signs—in the sample appears confusing at first. Simple, gray triangles or jagged profiles, pyramids with shadows, even contours and shaded relief can be found. The most common type of hill sign in the sample, appearing on two fifths of the maps (between one third and one half of all fantasy maps), is an oblique (or bird’s-eye view) picture of a mountain, shaded to give it an appearance of volume.70 This kind of hill sign came into use on nonfictional maps during the Renaissance, as copper-engraved, printed maps came to displace hand-painted maps.71 At the 95 percent confidence level, however, the oblique hill sign is not significantly more common than the sign type it succeeded, the profile view of a mountain, sometimes with hints of shade, a type that can be found on at least 17 percent, and on as many as 36 percent, of all fantasy maps. On medieval maps, the profile view looked “rather like cock’s combs” or “serrated bands”72 or had, at best, basic shading on the sides.73 The fantasy map equivalents, with or without shading, tend to be simple upside-down V-shapes with no significant difference in prevalence between profiles with and without shading (although the latter is twice as common in the sample).
Around 1680, vertical shading came to be used, an advance that “enabled the cartographers to show […] the length and breadth of a mountain and also to give some approximate idea of the gradients of its slopes.”74 During the eighteenth century, the hatches of the vertical shading sometimes resulted in mountains that looked like “hairy caterpillars”75 but also led to the development of hachures, “short lines whose thickness indicates steepness of the slope.”76 This type of hill sign is fairly rare on fantasy maps, found in only five instances (or just over 5 percent) in the sample. Given the margin of error, we can expect to find this sign on less than 12 percent of all fantasy maps, possibly on as little as 2 percent. That makes the sign at least 30 percent less common than the profile view and 60 percent less common than the oblique hill signs.
Even less common in the sample, although not significantly so in the fantasy-map population, are the two types of hill signs that are typically used on general survey maps in the actual world today. Contours, which derive from earlier charts of isobaths (lines of equal water depth), appeared on maps as early as 1737 but did not supplant hachures until the second half of the nineteenth century.77 Expression of landform through shaded relief is related to hachuring and was developed in the 1860s.78 Only two instances each of these hill signs can be found in the sample, and one book (Kirkpatrick’s The Right Hand of God) contains one map with shaded relief and one with contours.
Medieval and Renaissance (pre-Enlightenment) hill signs clearly dominate on the maps, however, constituting at least four fifths of all fantasy hill signs. The reason for this is not only that these signs are iconic and self-explanatory but also that they remain highly conventional, part of a cartographic language we acquire along with other cultural knowledge as we grow up.79 Wood persuasively argues that individuals learn the various types of hill signs in the same sequence that these signs developed historically;80 the earlier types of hill signs, therefore, appear obvious, self-explanatory. But the relative absence of post-Renaissance hill signs, as well as the general tendencies found in this survey, also agrees with (high) fantasy’s general proclivity for pseudomedieval settings.
In “‘Fantastic Neomedievalism’: The Image of the Middle Ages in Popular Fantasy,” Kim Selling discusses why a “significant number of fantasy authors persistently locate their stories in environments where the characters wear medieval dress, fight with swords, and live in hierarchical, vaguely feudal, semi-pastoral societies with low levels of technology.”81 Although Selling uses Umberto Eco’s term “fantastic neomedievalism,”82 Eco as well as Kenneth J. Zahorski and Robert H. Boyer also refer to the “pseudomedieval.”83 Indeed, pseudo-rather than neo-is the more suitable prefix, as the Middle Ages are evoked rather than recreated. The “vaguely feudal” setting is, in Brian Attebery’s words, “essentially a simplified version of the Middle Ages”84—simplified in that enough contemporary ideas and sentiments replace their historical counterparts to make the story palatable and comprehensible to a reader of today. The survey indicates that the same goes for the maps.
As the choice of hill signs suggests, the maps pursue a pre-Enlightenment aesthetic. What we generally perceive as late developments, such as map projections and legends,85 are uncommon. On the other hand, truly medieval conventions are rare. No maps in the sample use the hill signs that were prevalent on actual medieval maps: the serrated bands and cockscombs that represented mountains on many pre-Renaissance maps cannot be deciphered by today’s readers. Furthermore, the maps are oriented with north at the top unlike the actual medieval T-O maps, which had east at the top. Rather than appearing medieval, the maps only vaguely suggest the Middle Ages by mixing simplified medieval features with modern conventions.
These modern map conventions are, in fact, only a part of a much larger cluster of social conventions that can be found in fantasy settings, something indicated by the dominance of northern-hemisphere settings. On the whole, the maps indicate a genre-wide conventionality, although some maps, such as the cartographically advanced maps of Kirkpatrick or the landscape-drawing-as-map of Larry and Robert Elmore (as close as any map in the sample gets to what Ptolemy calls chōrographica, a more artistic representation of a small region86), try to escape the pattern. Fantasy, especially high fantasy, offers a chance to break with the conventions of the actual world and invent new rules for mapmaking (or return to previous ones), but such inventiveness is actually very rare. Thror’s Map in The Hobbit, more than any map in the sample, takes advantage of the genre’s cartographic possibilities, with its alternative (medieval) orientation and use of an alien alphabet. The vast majority of the maps follow a basic mold established by the two maps in the first edition of The Fellowship of the Ring. In the close reading that follows, these two maps are examined to demonstrate what we can learn about a fantasy work and the world it creates by paying attention to its maps.
N = 92
a Adds up to more than 100 percent, as some maps use more than one type of hill sign.
b One map uses white shading to represent mountains; another is a perspectival drawing of the landscape.
READING FANTASY MAPS
In Joseph Conrad’s Heart of Darkness, Marlow reflects on the attraction of maps:
Now when I was a little chap I had a passion for maps. I would look for hours at South America, or Africa, or Australia, and lose myself in all the glories of exploration. At that time there were many blank spaces on the earth, and when I saw one that looked particularly inviting on a map (but they all look that) I would put my finger on it and say, When I grow up I will go there.87
The attraction of Marlow’s map apparently lies in its blank spaces, in all the areas left to explore. It is tempting to assume that a fantasy map holds the same attraction—that it opens up an unknown world to explore. Yet, as the ensuing readings of the two Tolkien maps demonstrate, blank spaces on fantasy maps do not necessarily refer to something unknown. On the whole, the fantasy map appears to have a totally opposite function from Marlow’s blank spaces. “[T]he very presence of maps at the front of many fantasies implies,” according to Farah Mendlesohn, “that the destination and its meaning are known.”88 In other words, the map reflects an effort to make known, not to invitingly offer the unknown. The unknown is largely kept outside the map or along its edges. As was pointed out earlier, at least three quarters of all fantasy maps have land borders beyond which the unknown lurks—or entices. For the greater portion of the mapped fantasy genre, only part of the world is a stage, illuminated and clearly visible.
A “close reading” of a fantasy map means an investigation of what the map makes known and how it makes it known. In such a reading, it must be recalled that although the map is a conspicuous part of how the fictional world is understood, it is never a stand-alone portrayal of that world. Nor should the fantasy map’s customary position at the front of the book be taken to mean that the map is to be read without reference to the text. Text and map go together. The map is presumably meant to be consulted during the reading of the text, and it therefore makes sense to read the map in the context of the entire text.
In the reading of the first Fellowship map, “A Part of the Shire,” the main theoretical underpinnings of literary map readings are presented. The larger, more complex map of the western part of Middle-earth is read according to the same basic theoretical ideas. Some theoretical additions are made when warranted by differences between the two maps. The maps used for the readings come from the Houghton Mifflin Fiftieth Anniversary edition,89 where they are printed in red and black ink (like the maps in the first edition of The Lord of the Rings). In the first (and many a subsequent) hardback edition, volume one, The Fellowship of the Ring, contains two maps: the Shire map, a one-page map set just before chapter one in book I; and the general map of western Middle-earth, a foldout map pasted in at the end of the book. The one-volume Fiftieth Anniversary edition has the general map pasted in at the beginning and the larger-scale map of Gondor and Mordor pasted in at the end. For paperback editions, the foldout maps are cut up and printed in black ink only on several pages.90 Although the readings are based on the maps with black and red print, descriptions are added to make the discussion possible to follow on a map with black print only.
Reading “A Part of the Shire”
Every map, Wood informs his readers, has an author, a subject, and a theme;91 and while these three characteristics may seem straightforward, they are not as easily identified as one might at first believe. Although a fantasy map does not point to a location in the actual world, it has a subject, a fictional place to which it refers. The map that comes between the prologue and the first chapter in The Fellowship of the Ring appears to state its subject clearly: a label in the top right corner announces that it is a map of “A Part of the Shire.” Some knowledge of the Shire (gained, for instance, from the preceding prologue) and only the most cursory of glances suggests that this is not completely true, and that “A part of the Shire and some of the land on its eastern border” would be more accurate. Except for the small region of Buckland, carved out between the river and the Old Forest, the land to the east of the Brandywine River is, in fact, not part of the Shire (FR, prologue, 5).92 This small discrepancy might seem irrelevant, but actually emphasizes the insular mentality of the Shire hobbits. Like them, the map does not admit to the presence of anything outside; the Shire is all there is—at least all there is worth mentioning. Any outside world is ignored (this is also the case with the arrows at the map’s edges; no mention is made of whether a destination is in or outside the Shire), and everything that appears on the map is subsumed as “A Part of the Shire.” Outside or inside is all presented within what further examination will prove to be a discourse of safety and control—regardless of whether it is Woody End, Bindbole Wood,93 or the Old Forest. By restricting the map’s explicit subject, the secondary world is divided into the known, on the map, and the unknown, off it. The label makes the map reassuring, as it implies that everything on the map is described in the prologue and therefore known and safe.
Just like the map’s subject, its author appears to be obvious. At closer scrutiny, however, the author’s actual identity becomes uncertain, and the reader begins to wonder whether it is part of the fiction or not. These two aspects tie into each other. The Shire map is located between two sections of text (the prologue and the rest of the book), sandwiched, as it were, between two parts of the narrative. Rather than providing a paratextual threshold, it is evidently part of the narrative document. It portrays an area of the world that has already been verbally described in the prologue. There is thus a suggestion that the map is itself part of the fiction or, rather, that the map doceme refers to a fictional Shire map. The author of this fictional map is equally fictional, a cartographer in the secondary world, not an artist in the actual world, even though the prologue casts some doubt over this fictional status by claiming that an extant copy of the Red Book of Westmarch can, in fact, be found in the primary world.
The Shire map is, in other words, not one map but two: a fictional map from the fictional Red Book, designed by a fictional cartographer; and an actual doceme in the actual The Lord of the Rings, which refers to the fictional map. This map doceme can be taken to have at least two different authors. The obvious author would be J. R. R. Tolkien, the author of the book as a whole. The map is signed with the initials “C.T.,” however, which suggests that someone else has been involved in making the map, if not in creating the secondary world. In a letter, Tolkien relates how he “had to call in the help of my son—the C.T. or C.J.R.T. of the modest initials on the maps”—to help with the maps,94 and Christopher Tolkien recalls that he is “virtually certain that my father allowed me some latitude of invention in that region of the Shire.”95 Thus, searching for a historical author complicates rather than clarifies matters, as at least two people (to father and son Tolkien could arguably be added, for instance, the editor and the engraver) have contributed to the map’s form. Using the concept of an “implied author,” an author encoded in the text, simplifies the discussion somewhat96 and would, for a map, correspond to an “implied cartographer.” Along the same lines, the text’s narrator would parallel the map’s “fictional cartographer.” The Shire map still oscillates between fictional and nonfictional authors, however, and since these two authors are difficult—often impossible—to tell apart, I have made no attempt to distinguish between them in the discussion that follows, unless such distinction is relevant to the interpretation of the map.
What we have is thus a map that slips between authors and fudges its subject. At a first glance, the map does not have a particular theme, either. According to Wood, a map’s theme is its “focus of attention.”97 No such focus presents itself. In fact, the Shire map is more like a general reference map in that it presents numerous themes that, together, establish the map’s general argument.98 The most prominent (but not the only) themes of the Shire map are topography (including vegetation and water courses), the road system, population centers, and administrative regions. These themes, along with the multitude of names found on the map, are part of an overarching discourse of defining, situating, and familiarizing the Shire—indeed, of instilling secondary belief in the country. Through its themes, the Shire map creates the “small, safe, and understood world” that a portal–quest fantasy such as The Lord of the Rings requires for its starting point.99
The topography of the Shire, as portrayed in the Shire map, is quite ordinary, perfectly safe, and totally understandable. Actually, topography is less important to the map than might be supposed, which is evident from the time and effort it takes to work out the terrain from what appears to be a topographical map. The landscape is one of hill country and river valleys. The west-to-east-running valleys of the Water and the Shirebourn rivers meet the north-to-south-running river Brandywine. That the land at the northeastern part of the map is higher is evident from the hills at Brockenborings and Scary, but also from how the Water tributary runs south. The Water and Brandywine valleys are both fairly densely populated, which suggests arable land (a suggestion that is also confirmed by the text). In the southern valley, separated from the valley of the Water by the Green-Hill Country and the forest at Woody End, the Shirebourn and its tributary (the Thistle Brook) have a similar, but shorter, south-then-east course to the Water. This valley is apparently less farmed, with Willowbottom south of Woody End as the only population center.
It is not this rolling landscape that is most striking, though, especially not on the maps printed in black and red. What stands out most is actually the network of roads (although their impact is somewhat reduced in the black-print-only editions). These roads dominate the map, running over most of it and illustrating how easy it is to travel in the Shire. Every place is accessible by a bright-red road: it is a country with infrastructure. It does not matter where you are on the map, there is always a road you can follow to any other place. To further establish the importance of the road system, more than half of the roads leave the map with a fletched arrow informing the reader that the road leads to other, named and thus known, places rather than into the unknown. Off to Bree to the east and Michel Delving in the west, north to Oatbarton and south to Longbottom—the land on the map exists somewhere, anchored by locations outside it. This display of certainty about what exists off the map makes its edges the very opposite to water margins. This land is not afloat in an uncharted sea but instead is clearly a part of something, a message emphasized by the map label: “A Part of the Shire.” It is also a place where you travel, and not necessarily to places on the map. The arrows that anchor this part of the land to the world outside also point out directions of travel. In a story that comes with a map that includes so many roads and so many places to travel off the map, journeys are inevitable. In its way, this map is as clear about what will happen in the book as is the title of The Hobbit, or, There and Back Again. (And the map’s message that the world is mappable and knowable sets readers up for a surprise once the journey takes them outside the Shire.)
All the places that the roads lead to are meticulously named. Names of regions and administrative areas, of rivers and marshes, and in particular of villages of varying sizes are liberally sprinkled over the map. This is not a map of the unknown, it is very much the known, the labeled, the familiar. It is a landscape tamed, not only by the red roads but also by the black names of settlements and topographical features alike, and it is divided into areas even more conspicuously named in red. Through this abundance of names, of labels, the map subjugates the landscape, brings it under control. The names also give “repeated implicit assurances of the existence of the things they label,” Tom Shippey argues, assurances that strengthen our secondary belief in the fantasy world.100 As most of the Shire names are taken from Tolkien’s actual-world surroundings, they do not even sound exotic but seem familiar, believable.101
The control of the landscape is particularly apparent in the division of the land into four administrative areas: the north, east, west, and south farthings. The farthing names, printed in large red letters, are made even more prominent by facing in different directions around the center of the map: when held with north at the top of the map, “North Farthing” is printed upside down, and “West Farthing” and “East Farthing” are turned ninety degrees clockwise and counterclockwise, respectively. The farthing borders (black, dashed-dotted lines) cross the land without appearing to follow or regard any natural borders, except in the southeast, where the border between the South and East Farthing may run along the river Shirebourn (though this is unclear). Straighter than any river or road, such borders appear on actual maps too. On maps of, for instance, Africa, Australia, or North America, straight administrative borders can be found, borders drawn with a ruler without regard for the landscape the map is meant to represent.102 The map, not the physical environment, becomes primary. Secondary-world maps are, as was pointed out earlier in this chapter, often primary to the landscape they portray; but the farthing borders also emphasize the priority of the fictional map over the fictional landscape. A colonization of the land is implied; like the many names, the borders present a landscape subordinated to hobbit culture.
Not only space but time is encoded in the map.103 A map’s tense, according to Wood, is the direction in time in which it points: whether it refers to its own past, present, or future.104 The tense is relative to the map; a map from 1858 that refers to the world of 1858 uses the present tense. A map of tomorrow’s weather will remain in the future tense when we look at it a week from now. The Shire map, at a first glance, appears to refer neither to its past nor to its future—surely, it uses the present tense? There is nothing on the Shire map that suggests any temporal direction: the map features are the same at the beginning of the story and at its end. To the extent that the Middle-earth maps (and other fantasy maps) have a tense, it is fictional, based on the time when they were created and the direction in which they point in the fictional time line. The Shire map, unlike the western Middle-earth map read shortly, lacks any signs of future or past tense, so it must indicate its own present.
Actually, to say that the map uses the present tense is not quite true, as we see when we also take into account the map’s duration, its “temporal thickness” or the time span it embraces: a few hours, a year, a century …105 The Shire map, by including elements that refer only to that which remains unchanged during the story, spans the time of the entire story. It can be used as the hobbits leave the Shire in the beginning as well as when they return to find Saruman’s destruction near the end. Anything transient enough to be changed by story events—for instance, the hobbits’ camps or the cut-down trees—has been left out. The time pointed to is thus as much now in the beginning as in the end—the map’s tense is not just the present, but a constant present. The time referred to is not a specific now but a now that runs all through the story. This constant present is, to some extent, similar to the present tense employed in discussing a text. The map, like the text, spans all the time in the story and refers to it all simultaneously.
Despite the constant present, however, the map also offers a historical perspective.106 To have a forest called Old Forest, for instance, automatically suggests that there is history, that some things are old while others are new or young. From the map we can tell that apparently one forest is older than the other two: Bindbole Wood and Woody End may be forests, but they are not old enough to be old forests, let alone Old Forests. (The Old Forest and its connotations of age are discussed further in relation to the other Tolkien map.) From the text, we learn that the map contains other features that are part of the Shire’s history, in that they have not been there always. The prologue refers to some villages as “older,” implying that others are younger (FR, prologue, 6; further discussion on this topic follows)—these “older villages” are distinguished on the map by their uppercase script. We are also told that the hobbit area of Buckland—plainly marked as special on the map, something I will return to shortly—is a new addition to the Shire, its development being part of Shire history (FR, I, v, 96–107; vi, 108). Indeed, the very number of hobbit communities that dot the map, and the roads that connect them, carry an implication of the passage of time: the Shire was not built in a day but required time, history, to become that which is shown on this map.
The Shire map does not obey the cartographic conventions of any actual historical period. Instead, through its mixing of conventions and signs from different times, it presents that same pseudomedieval aesthetic that was identified from the survey. Even though there is no compass rose or similar device, the location of the North Farthing shows that the map is clearly oriented with north at the top, and this is not the only map feature that tends to the modern rather than the medieval. As was discussed in connection with the hill signs of the survey, the oblique hills that are used in the Shire map date back to pre-Enlightenment times. They are reminiscent of the “gentle rolling downs” that Edward Lynam says were fashionable with some draftsmen of the late sixteenth century, although, intriguingly, the Shire hills (as well as the hills and mountains on the map of western Middle-earth) are shadowed on the west side rather than the conventional east.107 Having the farthing names facing different ways suggests, faintly, that this map did not originally belong in a book but was a loose sheet that could be turned in the hands of the user, a convention that Harvey points out among some loose maps of the fifteenth century.108 The pre-Enlightenment bird’s-eye view is not carried through in the hobbit settlements, however, which are represented as black, angular dots, shaped as if the buildings were seen from straight above. The only exception is the Hill, Bilbo’s and Frodo’s home, which is portrayed with a small, oblique hill sign. This mix of bird’s-eye and plan views is particularly noticeable in the case of Tuckborough, where the houses seem almost glued to the hillside. Many other elements, such as script, borders, and labels, similarly follow post-Renaissance conventions.
Deciding what conventions to follow is only one of many choices that face a mapmaker, and the choices affect much more than the map’s aesthetic qualities. For a variety of reasons, an actual map does not—cannot—include everything; the work of the cartographer is a process of selection.109 The same is true for fictional maps. “It is the attempt to cut [the maps] down and omitting all their color (verbal and otherwise) to reduce them to black and white bareness, on a scale so small that hardly any names can appear, that has stumped me,” Tolkien complains in a letter to his publisher,110 and it is reasonable to suspect that only some elements of his fictional world were encoded in the map to begin with. The choices made mirror the map’s purpose: by including only those features that remain unchanged during the story, it is possible to make the map refer to a story-long now, as observed before. Choosing one hill sign above another means preferring a certain aesthetic over another; it also means alluding to one historical period rather than another. There are at least three selection processes at work in the construction of a fantasy map: an unconscious process that filters out features that are not even considered for inclusion; a conscious selection of what features to include in the map and what to leave out; and a choice of how included features are to be presented. Interpreting a map involves examining the results of these processes: what are the effects of including some features (and excluding others)? Of portraying some elements like this and other elements like that? A number of features of the Shire map invite such attention, for instance: the central location of the Three Farthing Stone and the marginal location of a quarry; the distinct red script used for Buckland; the differences in upper-and lowercase in village names; and how different script is used for the three different forests. Such selections influence what information the map conveys about the secondary world.
Arguably the most privileged location on a map is its center, the spot that, ceteris paribus, tends to be the first focus for the viewer’s attention.111 At the center of the Shire map, we find the Three Farthing Stone, the meeting point of the borders of the East, West, and South Farthings, which sits “as near the centre of the Shire as no matter” (RK, VI, ix, 1000). The map, in other words, makes quite clear what the text mentions almost a thousand pages later, near the very end of the story. At the map’s center are the geographical and administrative centers of the Shire, and it portrays not only “A Part of the Shire” but also the central part of the Shire—the central part of the world, at least as far as the mapmaker is concerned. At this point, it makes sense to distinguish between the implied cartographer and the fictional cartographer: the Shire map has the same implied cartographer as the map of western Middle-earth, but it also has a fictional cartographer. If the Shire map, like the text, is considered to be translated from the (fictional) Red Book of Westmarch (FR, prologue, 1), we can assume a mapmaker from the Shire, an assumption corroborated by the map’s center. This center is clearly the hobbit heartland: close by are found the villages of Tuckborough, By-water, and Hobbiton, the home of the Bagginses. It does not matter that Michel Delving off to the west is actually the “chief township”—we see from the map where the center is located, and the fact that one of the villages is called Hobbiton and that there is a separate mark for the Hill just to its north adds to this center’s importance. Any reader looking to find where the story begins will have little trouble in hitting the right spot, and what is left off the map is merely periphery.
There is a periphery on the map as well, however: a fairly empty landscape with a number of roads leading off to the south and west (and a nameless village on the Michel Delving road), and an eastern fringe of countryside that is not even part of the Shire (although the map does not tell us that). Along the northern edge, there are a handful of communities, all with names that seem somewhat denigrating or belittling: Nobottle, Needlehole, Brockenborings, and Scary.112 Next to the last of these, we find a quarry, the inclusion of which is quite intriguing. This quarry is the only place of production of any kind on the map, even when fields and mills, coppices and breweries would seem to be more relevant to hobbit culture. Geographer J. B. Harley asserts that “maps—just as much as examples of literature or the spoken word—exert a social influence through their omissions as much as by the features they depict and emphasize,”113 and the presence of the quarry underscores the absence of the other places of production. Would it not make more sense to include mills, especially as the text presents mills as important means of production and the destruction (“modernization”) of them is described as villainy (RK, VI, viii, 990)? Quarries, on the other hand, have negative connotations: Gimli assures Legolas that the dwarves “would tend [the Caverns of Helm’s Deep], not quarry them” (TT, III, viii, 535); Saruman’s henchmen turn Bagshot Row into a “yawning sand and gravel quarry” (RK, VI, vii, 993); and the ruffians had hidden stolen goods and food in the “old quarries” at Scary (RK, VI, ix, 999). Even though the quarry is “old” (that is, disused), it is a blot on the Shire’s peaceful, bucolic nature, a warning on the map that not all is, was, or will be well. The negative connotations of the quarry are added to by its equivocal name. Scary may, as Tolkien points out in his commentary to translators, “contain E[nglish] dialectal scar ‘rocky cliff’”114 but more obviously, it means “fearsome.” Scary Quarry is clearly a place to be feared, and although it is located at the periphery, away from the road network and thus disconnected from the rest of the Shire, it prophesies the ruin of the hobbit idyll.
The presence of the quarry and the absence of other places of production parallel a greater omission that is brought into focus by the inclusion of its (negative) opposite. As I mentioned before, the Shire map communicates control and safety; it is a map that demonstrates how wilderness has been tamed: only occasional pockets are left, surrounded by blank space. This is not the blank spaces of Conrad’s maps, the secretive, alluring unknown. It is the white areas of certainty, of that which is so obvious that it need not be included on the map. The few forests, marshes, and hills are exceptions, just like the quarry; the (positive) norm is left out because it is so obvious. Flat, cultivated land is left blank; it has nothing worth referring to, nothing to get excited about. In particular, nothing that is of any relevance to the story.
Another presence that draws attention to itself as well as to the omissions that it signals is Buckland, or more precisely the red capitals used to mark it. No other “folklands” (for instance, Tookland, where nearly all the Tooks live [FR, prologue, 9]) are mentioned on the map, which indicates that Buckland has a distinct status, that its position is of special interest to the reader (and mapmaker). There are actually several grounds for this distinction. First, in the prologue, Buckland is mentioned as a later addition to Shire territory. It is not, in fact, a farthing, but one of the marches, so it constitutes a separate administrative entity, whereas Tookland is part of the West Farthing. Just as the farthing borders denote hobbit control of the landscape, so does Buckland. When this land was made a part of the Shire, the hobbits had to fight for it against the trees of the Old Forest in a battle against the landscape itself. Second, because it was added comparatively recently, Buckland is not fully regarded as a part of the Shire, and the hobbits who live there are considered “peculiar, half foreigners as it were” by the other hobbits (FR, I, v, 96). Finally, whereas Tookland is only mentioned briefly and “off-stage,” Buckland is the goal for the first leg of Frodo’s journey. Except for Bridgefields to the north and the Overbourn Marshes to the south, the areas written in red (Green-Hill Country, Woody End, The Marish, Buckland, and the Old Forest) chart the hobbits’ journey across the Shire. (The Farthings, although also in red script, are written so differently in size and orientation that they obviously indicate a different category.) Any reader who tracks the journey across the map will have little difficulty even if the actual trail has no marks of its own. Again, the story plays a large part in how the map is shaped, and thus how the world is interpreted. Red draws more attention and conveys greater prominence to elements, and it is made clear to the reader that Buckland is worth paying special attention to.
Employing different types of script for different map elements (color, style, size, and so on) helps indicate various dissimilarities in the referents. On the Shire map, differences in case (using either uppercase only or lowercase with an uppercase initial) are used to signal a relevant difference, although the difference itself is not immediately clear. In terms of the watercourses, the larger ones (rivers) have names written in all capitals whereas the brooks’ names are written with only an initial capital. This suggests that the stylistic variations in the script are what Robinson and Petchenik term “mimetic” rather than “arbitrary”: in this case, a more prominent style corresponds to a more prominent (or larger) referent.115 If this correlation is true, uppercase-only villages should be more prominent than the rest. To some extent, the text supports this assumption. Of the five settlements that are all uppercase, three are mentioned in the prologue: the older villages Hobbiton and Tuckborough, and the Shire’s chief township, Michel Delving (FR, prologue, 6). The text’s juxtaposition of the older villages and the chief township implies that these entities have similar social status, a social status reflected as well as reinforced by the uppercase script. Regarding the status of By-water and Stock, nothing is said; the former is close to Hobbiton and is often referred to in the story, and Stock is mentioned but never visited by Frodo’s company. Since social status rather than story relevance appears to be the guiding principle behind the stylistic variation here, it may be assumed that the two villages share the social status of the other three.
The use of upper-and lowercase letters for names, hence, suggests that differences and similarities in script are “mimetically” assigned. A difference in script prominence reflects, in the case of watercourses, a difference in size; in the case of village names, a difference in social status. Consequently, the differences in style, size, and color used to write the forest names should also reflect some difference between the forests. The Old Forest and Woody End have names written in red uppercase but different sizes, with the names placed inside the forest, whereas Bindbole Wood to the north has its name set outside the forest and written in small, black script with uppercase only for the initials. From the map in general, we find that red script is used for areas and regions (the only exception being the Yale, which, on the map, could also refer to the black dot just beneath it116), so Woody End and Old Forest may be considered names for areas or regions rather than for the forests as such. Even so, the difference in script prominence suggests an inexplicable difference in prominence. As both the Old Forest and Bindbole Wood disappear off the map, it cannot be determined whether the question is one of size. Their relative script prominence, from smallest (Bindbole Wood: small, black, lowercase) to largest (Old Forest: large, red, uppercase), corresponds to their respective relevance to the story: Bindbole Wood, while appearing on the map, does not play any part; Woody End is the central setting of most of one chapter; and the Old Forest provides the first encounter with the world outside the Shire and the setting of an entire chapter (named after the forest).
By examining the choices of the mapmaker, it is possible to learn how the map relates to the story it presumably supports, but such an examination also sheds light on some of the social norms and constructions behind the map. In the case of the Shire map, the norms and constructions of the fictional cartographer are braided together with those of the implied cartographer. When it comes to the map of the western Middle-earth, which is more clearly a paratext in that it does not refer to any apparent map in the fictional world, the messages communicated by the map are more clearly those of the implied cartographer (or author). Whereas this map has some traits in common with the Shire map, it offers a worldview different from the controlled safety of the hobbit lands: according to the larger map, Middle-earth is a wilder, older place, and the map is much more explicitly made to serve the story.
Reading “The West of Middle-earth”
The general map of the western parts of Middle-earth was first included on a foldout sheet, printed in black and red, in the first edition of The Fellowship of the Ring and The Two Towers. In paperback editions, this map was printed in black only and divided into four sections.117 Although the Shire map and the general map have in common an abundance of names that strengthen the readers’ secondary belief (as well as allowing them to identify almost any location mentioned in the novel), the story comes across as much more central to the construction of the general map.
Unlike the Shire map, the general map has a compass rose (with north at the top) and a scale bar. For my edition of the map, one inch equals one hundred miles, which corresponds to 1:6,336,000 and yields a map area of about 6.5 million square kilometers, or about twice the size of India. A quarter of the map is covered by water, which gives a land area approximately half that of Europe or the United States. According to the label, the map is of “The West of Middle-earth at the End of the Third Age,” and even though some features are included, much of the eastern and some of the northern side of the map are, in fact, quite empty compared to the western lands. This is a map whose subject is not a nonspecific “west” but a number of regions defined as “The West,” a cultural, political, and historical as well as geographical location, something that is confirmed by how its military leaders are repeatedly referred to as “the Captains of the West” (RK, V, x, 867 et passim). These regions (Eriador, Rohan, Gondor, and Rhovanion) are set against the land of Mordor, which is the only land on the eastern third of the maps to be reasonably detailed. The label is signed CJRT, for Christopher Tolkien, but since it is impossible (and in this discussion irrelevant) to decide which features of the map come from which Tolkien, this map is also considered to have an implied cartographer. However, as the general map is more clearly paratextual than the Shire map and exists only on the threshold to the text, I assume it to have no fictional cartographer; this aspect is therefore left out of the subsequent discussion.
Of all the topography portrayed on the map of the western Middle-earth, the mountain ranges attract the most attention. In the middle of the map, extending from the hills beyond Carn Dûm in the northwest to Minas Tirith in the southeast, runs the sinuous mountain range of the Misty Mountains and Ered Nimrais (I use the English rather than Elvish names if they are given on the map). The inverted S-shape of this range neatly indicates where the story’s action will take place: in the southeast, it points at the mountains around Mordor; and, together with the Blue Mountains, its northern curve embraces Eriador. The shapes of Mordor’s mountain ranges and of the Misty Mountain/Ered Nimrais range give an artificial impression, suggesting that supernatural forces rather than tectonics are behind the very landforms of Middle-earth. This impression is confirmed in The Silmarillion, which describes (in Ainulindalë) how the semidivine Valar worked to create the landscapes of the world and (in Quenta Silmarillion) how the evil Melkor built the Misty Mountains.118 Given their prominence in terms of location, shape, and quantity, it is hardly surprising that negotiating various mountains is central to the plot, such as the Fellowship’s passage through Moria, Aragorn’s passage through the Paths of the Dead, and Sam and Frodo’s passage through Cirith Ungol. The very goal of the quest is a mountain, Mount Doom, inside which the Ring is destroyed.
Another prominent topographical feature is the enormous forest of Mirkwood. Although the great woodlands are never visited by the protagonists of The Lord of the Rings, Mirkwood draws attention to the other forests of the map. Forested areas are far scarcer than would be expected from such a wide stretch of land; apart from Mirkwood, only a few forests are marked on the map, some of them appearing to be quite small. At this scale, however, size becomes tricky to judge. The smallest forest on the map is Chetwood outside Bree, which takes the hobbits and Aragorn more than a full day’s walk to get through (FR, I, xi, 178). There ought to be more forests like Chetwood on the map, for instance Woody End and perhaps even Bindbole Wood in the Shire, or the western parts of Middle-earth would be a bare place indeed. This bareness does not agree with the lack of settlements; if there are no people, what keeps the formerly forested areas119 from turning back into forest? Yet the forests included on the map are not simply those visited in the story, although all forests the hobbits travel through or see outside the Shire are marked, regardless of size. The woodlands along the feet of the Blue Mountains and around the northeast corner of the Sea of Rhûn, which are neither visited nor seen, only add to the impression that most of western Middle-earth is bare, bringing into focus what few forests there are.
Other topographical features reinforce the bareness of the map and consequently bring focus to the few elements that are included. Thus, mostly relevant rivers are found on the map, allowing for easy identification of each river crossed or traveled along. Ultimately, the topography serves the story; that, rather than size, determines what is included and what is left out, with only occasional exceptions, such as the Sea of Rhûn or the Blue Mountain woodlands.
Where the nonverbal map elements serve the story, the verbal ones—the names—serve secondary belief in the world the map portrays.120 In many cases, features are only marked by the linguistic signs; apart from topographical signs, roads and population centers/strongholds are the only features that have nonverbal signs. In other words, the map is mainly about names. Indeed, when Shippey observes how the characters tend to “talk like maps,” he exemplifies this point with characters who tend to list the names and spatial relations of various features.121 Through the red script, the text is highly visible, and it is emphasized how this is a world where every place is known and named, often even with two names. Names are provided in languages from the secondary world (mostly Elvish) as well as in English, sometimes in both (generally with the English translation within parentheses). With the help of all these names, the reader can navigate the world of the story and follow the characters’ journeys; but the names also define the secondary world spatially, by creating a great number of places and spatial relations. Through their multitude and the many translations, these names even allow the reader to puzzle out some of the basic morphemes of the fictional languages, observing, for instance, how mountain ranges are called Ered (Ered Luin/Blue Mountains; Ered Lithui/Ash Mountains; Ered Mithrin/Grey Mountains); how mith-can mean “gray” (Ered Mithrin/Grey Mountains; Mithlond/Grey Havens); and how Emyn Uial/Hills of Evendim, Nenuial/Lake Evendim, and the [Sea of] Núrnen in Nurn give us uial = “evendim,” emyn = “hills,” and nen = “lake/water.” This is not only a map of a world but a key to its languages.
The fact that places and features on the map often have more than one name also suggests a relation between the languages other than the purely linguistic. Generally, an Elvish name is followed by the English version within parentheses. Typical examples are Ered Luin (Blue Mountains), Gwathló (Greyflood), and Baranduin (Brandywine), with Weathertop (Amon Sûl) providing an exception. The English names are often a (nearly always literal) translation—ered = “mountains,” luin = “blue”—but sometimes the English name is only notionally similar, so that gwath-in Gwathló becomes “gray” rather than “shadow” and Amon Sûl (“windhill”) is Weathertop in English. There might be a phonetic similarity only—Brandywine for Baranduin (“brown river”)—or no apparent connection to modern English, as when Angren (“of iron”) is translated as Isen (Old English for “of iron”). Numerous names are also left untranslated, appearing either in English or (more commonly) in Elvish. Some of the English names are straightforward (e.g., North Downs, Dead Marshes, the Brown Lands); others use more obscure or old-fashioned language or roots (e.g., Trollshaws, Entwash, Rivendell).122 The Elvish language is made obviously superior to English, partly through its status as the preferred language, partly because although there are a great number of English names, the major names—labeling larger regions and therefore written in larger script—are in Elvish. Middle-earth, we see, contains both the familiar and the alien, although the latter is more prominent.
The variations in script separate the marks on the map into different classes. Small uppercase script indicates names of mountain ranges; regions are written with larger uppercase script, the size of the script increasing with the size of the region (compare, for instance, Eriador and Minhiriath or Mordor and Nurn), although the names of smaller regions (for example, Forlindon, Lebennin, and Lossarnach) are written in uppercase for the initial letter only—as are all other names as well. The script is curved to roughly indicate what region, mountain range, or river it refers to, but the text is straighter when referring to a population center, stronghold, or tower. Outlined capitals are used for the former realm of Arnor and the three kingdoms into which it was divided. The many variations in size, case, and curvature reinforce the impression that this world is fully explored and fully believable. It is not only littered with names; these names appear to be divided into an abundance of categories. Regions are divided and subdivided, rivers have tributaries and marshes, mountain ranges and hills are everywhere: all over the world, the map tells us, there are places whose names are worth knowing.
This profusion of names and categories obscures the fact that there are also places, and kinds of places, apparently not worth knowing about, however. Just as there are actually very few forests on the general map, there are not many places where people live, at least in terms of towns, villages, cities, castles, fortresses, harbors, towers, and so on. These signs of civilization (human or otherwise) are few and far between on the map, giving it an impression not only of bareness but of desolation. Closer examination reveals that the guiding principle for inclusion is, again, the story, not size or social or political relevance. Nor are the respective marks mimetic in relation to a place’s relevance to the story. Tharbad (mentioned a few times in passing) is more prominently marked than the Ford at Rivendell (where Frodo faces—and escapes—the Black Riders); Lond Daer (which does not feature in the story) is as prominently marked as the Grey Havens (where the last Elven ship leaves, marking the end of the Third Age). The addition of places other than those of the story affects the reader’s perception of the story world, regardless of whether the map is seen as a paratextual element or as a doceme. As the former, the map extends the world of the text; as the latter, the world of the total document is the sum of the world as portrayed on the map and in the text. Both perspectives offer insights into how the many places reinforce secondary belief in the world, implying that Middle-earth is more than the setting of a story.
The west of Middle-earth is not a place that crawls with people, at least if we go by the lack of settlements. This impression is corroborated by the small number of roads, the tiny dotted trails that wind across the land. Compared to the Shire map, where roads are given pride of place, the world outside is clearly not a place to go traveling. It is wilderness, untamed and unsafe (but not unknown). This is stressed even further by the fact that no administrative (political or other) borders are to be found anywhere on the general map. Whatever borders there are coincide with natural borders: the Mountains of Shadow suggest the border to Mordor; the end of the respective forests are the borders of Fangorn and Lothlórien. The Ered Nimrais provides a border between Rohan and Gondor (and the different languages reflected in the names suggest another border between them along the Mering Stream, between Eastfold and Anórien).
Where the Shire map subjects the landscape to its culturally constructed borders, the general map does the opposite. It portrays an internal tension between its natural landscape and cultural control of that landscape. While the profusion of names emphasizes how well known, how defined, the secondary world is, the scarcity of cultural constructions, be they roads, villages, or borders, stresses the world’s wilderness and depopulation. This tension runs through Tolkien’s text, clearly visible, for instance, in the ambivalent stance between tame and wild nature, where parklike Lothlórien is set against primeval Fangorn, the entwives’ horticulture against the ents’ forests, even Shire countryside against Old Forest wilderness.123
The map does not, however, encode this tension in the schematic structure that Pierre Jourde proposes when he divides western Middle-earth into one region of civilization and goodness (Gondor and Eriador) and one of wilderness and evil (Mordor and Rhovanion).124 While the conflict between Gondor and Mordor is plain, not least in the text, neither Eriador nor Rhovanion is a region that can easily be interpreted as either wild or civilized, and even less as good or evil. The former region may contain the Shire, Bree, and Rivendell, but it is also a place of lost realms, both good and evil, where Rangers do battle against evil beings in the wilderness. Rhovanion, on the other hand, contains the evil strongholds of Sauron and Saruman but is also dotted with civilized societies such as Erebor, Dale, Lothlórien, and Rohan. Like the conflict between good and evil, the tension between wild and tame is present all over the map—visible, but never simple. Jeremy Black somewhat misses this point when he claims that the map of western Middle-earth “gives no real sense of the spatial range and potency of wisdom and evil, good and ill, that are important themes in [Tolkien’s] narrative.”125 The age-old conflict between the “evil” side of Morgoth and Sauron and the “good” people who oppose them can be traced all over the map: from the notes about Arnor, Angmar, and major battlefields to the very absence of Beleriand west of the Blue Mountains; from Mordor and the empty lands beyond to Dol Goldur and Mirkwood, even to the icebay of Forochel, a remnant of Morgoth’s icy reign—the “spatial range” of evil is stamped on the map, a part of Middle-earth’s history and development.
Middle-earth’s history and the tension between past and present is a theme as clearly visible in the general map as is the secondary world’s topography. The very label of the map ensures that the reader comes to this map with a historical awareness: “The West of a Middle-earth at the End of the Third Age.” Apart from instilling a sense of finality, it accentuates the fact that Middle-earth has a past (three ages of it, at the very least) as well as a future, a Fourth Age from which it is possible to establish the end of the previous age. Already the map label can thus explain why Ricardo Padrón feels that the suspension of Tolkien’s world between a deep past and an impending apocalypse is encoded in the map.126 Parts of this past are then explicitly marked on the map. In a particular kind of script (outlined capitals), the approximate location of “The Lost Realm of Arnor” is given and, with smaller letters, the regions Arthedain, Cardolan, and Rhudaur are marked. Even if nothing is known about the world, the location of a “Lost Realm” gives the map temporal thickness. Along with the note stating that “Here was of old the Witch-realm of Angmar” in northern Eriador, the references to the old kingdom actually cover the history from the founding of Arnor in the year 3320 of the Second Age, through its division into three realms in the year 861 of the Third Age, the establishing of Angmar (circa 1300), the respective falls of Rhudar, Cardolan, and Arthedain (in 1409, 1636, and 1974), and Gondor’s final defeat of Angmar in 1975 (RK, Appx A, 1014–17; Appx B, 1060–61). References to Arnor’s history do more than add three millennia to the map’s duration, however. Numerous map features are relevant only in connection to the lost realm’s long history, such as the old cities of Annúminas and Fornost and the arrow that points off the map to the icebay of Forochel. The map creates temporal depth through the inclusion of references to Cardolan and Rhudaur in much the same way that the text brings temporal resonance to the story events by including the history of Weathertop and the fall of Rhudaur (cf. FR, I, xi, 181; xii, 196). Similarly, the comment that South Gondor is “now a debatable and desert land” adds to the theme of Middle-earth history with its reference to the Gondor civil war (RK, Appx A, 1022–23; Appx B, 1061), and the “now” emphasizes the diachronic nature of the map.
A handful of other marks similarly draw attention to the map’s diachronicity, but with less focus on a specific time and more focus on how time passes. These marks invoke the past through the word old: Old Forest, Old Ford, and Old Forest Road. In a world where the past is more present than the present, where ancient conflicts and mistakes return to haunt the living, where some of the living actually recall events several millennia in the past, there is, in fact, still a need to refer to a forest, a ford, a road, as old. That which is old has the power to withstand the ravages of time, it seems; the Old Forest Road runs straight through the great Mirkwood forest, obviously remarkable enough not to succumb to the forest. The Old Forest is kept back from the Shire by a hedge and a gate. The forest, in particular, appears to be oldest rather than old. It is associated with Tom Bombadil, who “remembers the first raindrop and the first acorn” (FR, I, vii, 129), but also with Treebeard, “the oldest living thing that still walks beneath the Sun” (TT, III, v, 488); it is a remnant of the primeval woodlands that once covered much of Eriador.127
Despite the declaration of the label, the map does not refer merely to its own present but also to its past; its tense is only overtly present. A closer investigation uncovers traces of history, explicit as well as implicit. To Padrón, the many place-names bear witness to the richness of the world’s past;128 but there is more to the map’s encoded history than names. Features are included to enable the reader to chart Bilbo’s adventures in The Hobbit; battlefields and mass graves of old can be found outside Mordor (Dagorlad and the Dead Marshes); and even the rule of, and war against, Morgoth is hinted at. (The icebay of Forochel is the result of the great cold of Morgoth, and the broken Blue Mountains and absent Beleriand are the outcome of the War of Wrath that finally defeated him.) The map’s duration thus not only includes the end of the Third Age; it is actually thick enough to go back to the First Age and the battle with Morgoth. This temporal thickness is not immediately apparent to a reader who comes to the map for the first time, but if the map is used as a reference tool during the reading, the encoded history becomes apparent. The land is one that has evolved over the ages, and this evolution is clearly present in the map. Although the story has been a guiding principle for the cartographer, the history of Middle-earth has been just as important. Indeed, the meeting of historical and linguistic setting with story that characterizes the novel is evident already from the general map.
The significance of story as well as history is noticeable from spatial positions. The area where explicit historical features are densest is also where we find the beginning of the story. Unlike the Shire map, where the beginning is located near the center of the map, the story begins near the top left corner of the general map. In Western society, top left is a privileged position: it is where we start our reading on a page, and the Fellowship travels across the map rather the way our gaze scans a newspaper (or web) page, more or less diagonally from top left to bottom right. The landscape, and the characters’ journey through it, seems to be set out to make reading the map and finding the characters’ route as easy as possible, emphasizing once more how the story is a guiding principle of the map.
A guiding principle it is, but not an unbreakable law: the location where the story ends is unclear. Certainly, Sam’s final return home and announcement that he is back is the obvious ending of the book, in terms of being the last few words of the final chapter; but there are other ways of considering where the story ends. There are at least three main contenders for what constitutes the end of The Lord of the Rings, each bringing a particular aspect of the story to a close.129 The diagonal journey across the map ends with the disposal of the Ring in Orodruin, a moment that marks the end of the hobbits’ quest and the victory over Sauron’s evil. Spatially, this ending marks the farthest point away from the Shire on the general map, in the same way as the dead and deadly land of Mordor is the total opposite of the fertile and pleasant Shire. The subsequent defeat of Saruman marks another conclusion, with the defeat of an evil that is less cosmic and more human in scale. This ending, as well as Sam’s return home in the book’s final lines, is set in the Shire. The journey has taken the protagonists full circle; the hero has returned, having traveled, as in The Hobbit, there and back again.130 A third possible ending is the departure of Frodo, Gandalf, and the elves from the Grey Havens, an event that symbolizes the departure of magic from the world, not just the defeat of evil. This ending points to the west, off the map and into the unknown. Each ending thus relates differently to the map and the journey that trails across it, allowing the story as a whole to have a threefold ending of arrival, return, and departure into the unknown.
The Shire and the Grey Havens are located at, or very close to, the place where the story begins. The more far-reaching consequences of the story resolution—the disappearance of magic and the arrival of the Age of Men—are worldwide and have no specific location, but they are alluded to at the center of the map, another privileged position. There, the forests of Fangorn and Lothlórien are found next to each other. These forests offer a much subtler connection to the depths of Middle-earth’s history than the verbal signs that refer to Arnor’s location of old,131 but whereas the lost realm of Arnor is restored through the destruction of the Ring and the return of the king, the ancient forest realms come to an end. The end of the Third Age is the end of magic in Middle-earth, and the world changes. The juxtaposition of Lothlórien and Fangorn thus foreshadows Treebeard’s meeting with Galadriel and Celeborn near the end of The Return of the King. “[T]he world is changing,” the old ent says to the elven rulers, “I feel it in the water, I feel it in the earth, and I smell it in the air” (RK, VI, vi, 959). The change he feels is the price for defeating Sauron. “Much fantasy does not have what we could call a ‘happy ending,’” Attebery observes in Strategies of Fantasy. “Indeed, the fantasist often seems to start with the idea of such a resolution and then to qualify it, finding every hidden cost in the victory.”132 Middle-earth is losing its magic: that is what Treebeard feels, and it is on that loss, on the doom of elves and ents and all things magical, that the map is centered. The “End of the Third Age” proclaimed by the map label is, in fact, the end of magic in Middle-earth.133
The middle of the map presents what is at stake, but the periphery warns about the enemy. Ranged around the northern, eastern, and southern borders of the map are names that signal the threats to the people of the West. The Witch-king of Angmar returns as leader of the Ringwraiths, and the peoples of Rhûn, Khand, and Harad—the Southrons and Easterlings—ally themselves to Sauron. Being marginalized also means being primitive. The Forodwaith, or Northmen, became Gondor’s allies but were still considered “lesser Men” (lesser, that is, than the Númenorean descendants in Gondor and Arnor), along with the Southrons and Easterlings. The privileged direction in Middle-earth is west: western Middle-earth is superior to other parts; the humans from Númenor (an island once situated in the sea to the west of Middle-earth) are superior to other humans; and to the west of Númenor is the Blessed Realm, where the semidivine Valar reside. Regardless of whether the periphery is teeming with the enemy or offers the only way to sail to an Elysium off the map, it is the unknown margin outside the relevant middle, beyond the reach of the story.
Much of the relevant middle is conspicuously empty, however. While it is littered with names, there are only a few iconic map elements indicating terrain or buildings, roads or rivers. On the Shire map, the blank spaces correspond to a norm that does not need to be mapped. Here, the white emptiness, portrayed on the map by name only, is not a stand-in for fields too obvious to map, nor is it treeless, flat heaths, or desert, tundra, grassy plains, or any other one type of terrain. It is not even simply “wilderness.” On the general map, the fields of the Shire are just as blank as the desolate Brown Lands; the grasslands of Rohan are empty white, as is the broken wasteland of the Plateau of Gorgoroth. White is not the unknown or unmapped, nor is it a specific type of landscape: it is landscape that it is irrelevant to map. The only relevant features of this unmapped whiteness are the numerous names by which its areas are known. All those names—in English as well as Elvish, and providing translations between the two—communicate the importance of language in itself. The map shows us a world defined by names and created by language, thus confirming that Middle-earth is a creation centered on language rather than nature—on the creation of new language and the translations involved in understanding it. Given Tolkien’s love of languages and philology, it comes as no surprise that one of the principal messages delivered by the map of the western part of Middle-earth is how his fictional world is ultimately a linguistic creation.
• • •
At least since the 1970s, critics have observed the importance of maps to the fantasy genre, although no comprehensive studies have been published before now. Obviously, no study, quantitative or qualitative, can be all-encompassing, and my own contributions to the investigation of fantasy maps in this chapter should be seen as the first tentative mapping of an alien country, not as the definitive exploration of all its blank spaces. As in many studies, time and money came to be unwanted constraints. Maps proved to be more uncommon in the genre as a whole than I originally thought (occurring in at most two fifths of the books rather than in at least half, which a cursory pilot study had led me to believe would be the figure). Their scarcity in the book sample thus led to a smaller map sample that, in turn, resulted in large margins of error. To be certain that the largest margin of error could be at least halved, however, the book sample would need to be considerably larger, requiring almost 360 maps. The advantage of such a large map sample would, of course, be the appearance of more rare features (for instance, oriented any way other than northward; showing a subway system; or portraying modern cities of the primary world); even if a certain phenomenon did not appear in my sample, it might still be found on as many as one fantasy map in twenty-five. (With 360 maps, something not found in the sample would be rare enough only to appear on one map in every hundred in the fantasy-map population.)
My survey made plain that the vast majority of maps mapped settings in secondary worlds. A valid question that remains for a future study would thus be to what extent high-fantasy novels come with maps. Such a study would tell us the proportions of maps in high to low fantasy, but it would require either the books in the sampling frame or (probably more feasibly) those in the sample to be separated into high-and low-fantasy works (something for which there was no time in this survey). Future studies may also want to include more children’s and young adult fiction in the sampling frame. My own sampling frame is biased away from fiction for young readers because such fiction was separated from fantasy fiction in the database I used, with the result that the survey largely concerned fantasy for adult readers. Whether there are any major differences between maps in fantasy for adults and in fantasy for children and young adults remains to be investigated, for instance by using another sampling frame.
Certain typical features of fantasy maps were indicated by the survey, features we can expect to find in at least half of all fantasy maps. In brief, a typical fantasy map portrays a secondary world, a compass rose or similar device showing its orientation with north at the top. It is not set in any given hemisphere (not necessarily in a spherical world at all), although there are reasons to believe that clues in the text would indicate north as the direction of colder climates. Apart from topographical map elements such as rivers, bays, islands, and mountains, such a map would also contain towns and other artificial constructions. The hill signs used are typically pre-Enlightenment (either profile or oblique).
Even this brief list reveals the mixture of modern and historical map features. Like much high fantasy, the secondary-world maps follow a pseudomedieval aesthetic according to which dashes of pre-Enlightenment mapping conventions are rather routinely added to a mostly modern creation. Whether this is because of careless research, genre conformity, lack of imagination, or a desire to give the reader the easiest possible access to the map and the world it portrays is hard to say. If the map is meant as an aid for reading (and writing) the story, as a paratext on the threshold between the actual world and the unknown geography of the secondary world, maybe the map should simply challenge the reader’s map conventions as little as possible.
Whatever the reason, a mixture of cartographic conventions from various time periods is found in the reading of the two maps from The Lord of the Rings as well. The readings also demonstrate that paying close attention to fantasy maps, as maps, can reveal information about the maps beyond the elements that were used in their construction. To find that center and periphery are set up against each other on both maps is hardly surprising; to set the familiar in focus, in the center of the map, is a traditional mapmaking strategy. But whereas the Shire map privileges the familiar over the unknown and communicates as its dominant message the control of the land (and the landscape) and the safety this brings, the map of western Middle-earth defines the world it portrays by naming it. The importance of the world’s language and history is as central to this map’s message as is geography, and both maps contain a link between geography and story. Where the Shire map communicates control and safety, however, the map of western Middle-earth communicates a tension between cultural control and wilderness.
Apart from what the readings of the Tolkien maps tell us about The Lord of the Rings, they demonstrate how much any fantasy map can say about the work it belongs to. Each map, it is clear, relates in one way or another to the text. To ignore what the map communicates and only analyze the text means omitting a significant part of the work. Furthermore, as two maps from the same book, portraying parts of the same world, display such significant differences in the messages they communicate, it seems obvious that not only one map but all maps present in a book should be considered. Although the great majority of all fantasy novels that come with maps only include one map, a not insignificant number include two, and at least some come equipped with three or more. Each of these maps is a doceme that adds something to the document as a whole, and each map is a paratext that offers a particular threshold across which the text can be entered.
Whether the map is alone or one of several, typical or idiosyncratic, referring to a fictive map or situated firmly outside the diegesis, we should not dismiss it lightly. Instead, critics as well as readers should let the map do what is—ultimately—its job: to lead us into the fantastic world of the story.
