Explorative exposure : Media in and of Mark Z. Danielewski’s House of Leaves

This is an electronic reprint of the original article. This reprint may differ from the original in pagination and typographic detail. Author(s): Title: Year: Version: Please cite the original version: All material supplied via JYX is protected by copyright and other intellectual property rights, and duplication or sale of all or part of any of the repository collections is not permitted, except that material may be duplicated by you for your research use or educational purposes in electronic or print form. You must obtain permission for any other use. Electronic or print copies may not be offered, whether for sale or otherwise to anyone who is not an authorised user. Explorative exposure : Media in and of Mark Z. Danielewski’s House of Leaves Kilpiö, Juha-Pekka Kilpiö, J.-P. (2018). Explorative exposure : Media in and of Mark Z. Danielewski’s House of Leaves. In H. Pyrhönen, & J. Kantola (Eds.), Reading Today (pp. 57-70). UCL Press. Comparative Literature and Culture. https://doi.org/10.14324/111.9781787351950 2018 5 7 57 4 Explorative exposure : media in and of Mark Z.Danielewski’s House ofLeaves JuhaPekka  Kilpi ö Even though there has been considerable interest in the medial and material features of literature in recent years, the diff erent research fi elds may not have converged in quite the same way as media themselves are said to have done. There is a distinct line of enquiry devoted to media as interface (with print versus digital as the most hackneyed example) and quite another to media as subject matter (ekphrasis, the musicalization of fi ction and so on). Certain works, however, call for a joint approach. In this chapter, Iexamine mediality as it cuts across Mark Z.Danielewski’s House of Leaves (2000), from cumulative representations in the storyworld to the physical presence of the book itself. Espen Aarseth’s study Cybertext makes a useful distinction between textonomy , the study of textual media, and textology , the study of textual meaning (15). In analysing House of Leaves , it is worthwhile to determine its media position according to Aarseth’s typology, but also to carry the notion of media over to the textological analysis. The former reveals how the book actually functions, and the latter how diff erent media are discussed in the discourse and what kinds of meanings they are assigned. By now Danielewski’s novel has generated almost as much academic commentary as the fi ctitious documentary fi lm The Navidson Record in it. But the ways in which the two medialities, both within the storyworld and on the textual surface, inform and aff ect one another have not been thoroughly analysed. Reminiscent of but also going beyond Vladimir Nabokov’s Pale Fire (1962), it consists of several narrative layers where narrators mediate, comment upon and obviously distort the texts of the preceding narrators, thus throwing into relief all manner of epistemological instabilities. The layered commentaries and metatexts READINGTODAY 58 58 are realised by way of diff erent fonts and their disposition on the page, but the salient typographic variation also refl ects the volatile ontology of the storyworld. The novel’s discourse is mainly based on the representation of the fi ctitious fi lm. Thus, it uses a form of intermediality that Isuggest could be called kinekphrasis : building on the traditional rhetorical term ekphrasis , I use this neologism to designate verbal representations of cinema or other forms of moving image. In addition, the visuality of the novel’s text (layout, fonts, colours, footnotes) is foregrounded so that the characters’ explorations in the storyworld are iconically transposed to the act of actual reading, navigating the forking text and operating the book. This makes the novel ergodic literature , where ‘nontrivial eff ort is required to allow the reader to traverse the text’. The term is composed of the Greek words ergon , meaning ‘work’, and hodos , meaning ‘path’. The required eff ort is physical, or ‘extranoematic’, and entails selection. Thus, the reader’s movement through the text involves something other than the standard practice of turning pages when reading a traditional novel (Aarseth 1– 2). My reading, then, deals with the confl uence of kinekphrasis and ergodics. This enables us to address the activity of reading on more than one level. House of Leaves ’s kinekphrasis is a critical reading of The Navidson Record by a character in the storyworld. This embedded reading is in turn set up against the actions of the actual reader, who must in eff ect become a user:manoeuvre the book, turn it around, leaf back and forth to follow the notes, use a mirror to decipher a piece of mirrorwriting, andsoon. House of Leaves is about a photojournalist named Will Navidson and his family, who discover that their new house opens a portal to an immense labyrinthine underground space that defi es the laws of physics and proves impossible to map. Navidson and his companions search the space and record their explorations on fi lm and video. This material he compiles into a featurelength documentary, The Navidson Record . The most extensive diegetic level is a scholarly exegesis written about thefi lm by an old man named Zampan ò living in Los Angeles. The work is left unfi nished at his death. The material is discovered, transcribed and compiled into a book by one Johnny Truant, a young man working in a tattoo shop. Truant also provides an extensive commentary of his own and reveals, among other things, that Zampan ò was blind – an unlikely candidate to do fi lm studies – and that The Navidson Record is itself fi ctitious, Zampan ò ’s makebelieve. The outermost level is that of unnamed ‘editors’: an amended version of Truant’s manuscript, ostensibly making the book the reader is actually holding House of Leaves by Zampan ò with 59 EXPLORATIVE EXPOSURE: HOUSE OF LEAVES 59 Introduction and Notes by Johnny Truant, 2nd Edition . In short, a reading of a reading of a reading. In the text the diegetic levels are separated by diff erent fonts. For the most part Zampan ò ’s discourse occupies the body text and his sources are given in the footnotes (the other studies about the fi lm obviously being fi ctitious, while most of the other theoretical and literary sources are real and cited correctly according to academic practice). Truant’s discourse is placed mainly in the footnotes. Quantitatively, the editor’s role is the least signifi - cant and is limited to the paratexts and some amendments to Truant’snotes. Zampan ò ’s discourse forms the kinekphrastic core. Also titled The Navidson Record , it purports to be an academic thesis complete with the appropriate references and stylistic decorum. As is the custom, it goes through and synthesizes the previous academic and critical reception of the fi lm and claims superiority over earlier eff orts because of its meticulous close reading and the most apposite interpretations. The analysis follows the fi lm chronologically, scene by scene and even shot by shot, and recounts the unfolding narrative in overt detail. In a sense this gives away that the fi lm probably does not exist in the storyworld either, since the narratee clearly is not expected to have seen it. This does not, of course, diminish its signifying potential in theleast. Cybertextual analysis should be fi rst grounded in textonomy and proceed from there (Aarseth 15), so Ibegin by addressing the work’s media position and then move on to the intermediality within the storyworld. Cybertext theory and House of Leaves ’s media position While Cybertext can be seen as a critique of and improvement on the hypertext theory that prevailed in the 1980s and 1990s conducted by George Landow and Jay David Bolter among others, Aarseth calls for a new understanding of textuality in general. Cybertext theory signifi es a perspective on all textuality and resists any essentialist distinctions between digital and print media (Aarseth 17– 18). In addition to introducing ergodic literature, Aarseth makes a terminological distinction between scriptons and textons . Scriptons form the output, the signs that are actually presented to the reader. Textons comprise the reservoir from which they are created (Aarseth 62). Even though Danielewski’s novel is routinely referred to as ergodic literature – it says so in Wikipedia (s.v. ‘ House of Leaves ’) – so far it has not been placed in Aarseth’s textonomical typology. Generally, it has been READINGTODAY 60 60 analysed with reference to hypertext and digital media (e.g. Chanen, Hansen, Pressman). Oddly enough, its engagement with the cinema has not received particular attention either. Although Paul McCormick mentions its ‘typographic montage’ (64), he discusses primarily the ways in which House of Leaves confronts cinema as a cultural form in the media environment of the 1990s. My analysis, in turn, zooms in on the textual and material devices used in representing the moving image. Cybertext theory conceptualizes the text as a triad consisting of the operator, the verbal signs and the medium, which are interdependent (Aarseth 20– 1). Instead of describing the alleged essence of separate media, the theory presents a typology where individual works can be placed and where they can be observed in relation to each other. It is built on seven variables (Aarseth 62– 4): (1) Dynamics:both textons and scriptons can change, or scriptons alone can change while textons stay the same, or both can be static. (2) Determinability: in a determinate text the adjacent scriptons are always the same; in an indeterminate text they arenot. (3) Transiency: scriptons may be either transient or intransient; in other words, they either have a temporal dimension or they do not. (4) Perspective:if the user can become a person in the fi ctional world, the perspective is personal. Otherwise it is impersonal. (5) Access:the user’s access to scriptons may be either controlled or random. Aarseth mentions the codex, or the bound book, as a typical example of the latter. (6) Linking:if there are links in the text, they can be either explicit or conditional. The conditional links do not function always but only under certain conditions. (7) User functions:as this variable is the most important to my analysis, Iquote it directly: Besides the interpretative function of the user, which is present in all texts, the use of some texts may be described in terms of additional functions:the explorative function, in which the user must decide which path to take, and the confi gurative function, in which scriptons are in part chosen or created by the user. If textons or traversal functions can be (permanently) added to the text, the user function is textonic. (Aarseth 64) The typology has later been expanded with new variables (see Eskelinen). I deal with some of them in the end section of the chapter, but for House of Leaves ’s actual media position the original ones suffi ce. 61 EXPLORATIVE EXPOSURE: HOUSE OF LEAVES 61 When placed in the typology, House of Leaves yields the following values:it (1)is static, (2)is determinate, (3)is intransient, (4)is impersonal by perspective, (5)can be accessed at random, (6)has explicit links and (7)has an explorative user function. Some of these categorizations are due to House of Leaves being a printed book. For example, transience– that is, scriptons appearing or changing over time– cannot be realized in print, for the time being at least (Eskelinen 37). Aprinted text need not necessarily be static, since in a typical gamebook the textons stay the same but the scriptons change because the reader may realize a diff erent string of signs on each reading (Aarseth 68– 9, table3.1; Eskelinen 390n19). At least in part the question comes down to reading conventions. House of Leaves is meant to be read in full, or at least there is no indication to do otherwise, which makes it static. Random access, on the other hand, is intrinsic to the codex in general, but in this case it is called to mind more distinctly. Because of its framing as a mock scholarly volume, House of Leaves includes an index at the end. The index is a strange one, in that it presents not so much the signifi cant concepts and proper names but a kind of frequency analysis on even such words as ‘and’, ‘for’ and ‘out’, and signifi cantly some that do not exist, designated ‘DNE’ (except that, by way of the liar’s paradox, they do exist in the index). The index in general, and the obvious insinuation that the ‘nonexistent’ words are merely concealed somewhere (for clues, see for example Hayles, ‘Mapping Time’), encourage the reader to take advantage of random access and dip into the book more freely. In theloop Aarseth’s examples include a text with the same media position as House of Leaves , Julio Cort á zar’s Rayuela (1963), where the reader is allowed a choice between two paths through the chapters. House of Leaves ’s ergodics, however, is more complex. The explorative user function is mainly brought about by links and the forking layout of the text. At its most simple, a link leads the user from one place in the text to another. Terminologically, link signifi es the connection between source and destination. What appears at the scriptonic level is more precisely called an anchor . The way the anchor is made to stand out– for instance the colour blue designating hyperlinks – is called a cue (Gunder 212–13, 222). Mark B. N. Hansen suggests that the word ‘house’ printed in blue throughout House of Leaves makes a ‘pseudoserious READINGTODAY 62 62 reference to the blue highlighting of hyperlinks on Web pages’ (598). Links indeed fi gure conspicuously in the novel, but the word ‘house’ is not one of them. The links in House of Leaves , to apply Anna Gunder’s (213–16) terminology, are all analogue, but there are both internal and external links. They are mostly realized as footnotes modelled on academic references. A typical typographical link in an academic text has two phases: fi rst from the superscript number in the body text to the corresponding number at the foot of the page (internal) and then to the material outside the book (external). In House of Leaves , it is mainly the internal ones that are used for ergodic purposes, although the principles of both are problematized. In Gunder’s (217–19) terms, an ordinary typographical footnote is bidirectional and homoanchoral both at source and destination. In other words, the superscript number (or similar) in the body text leads to that same number in the footnotes, and that same number from the footnotes back to the body text again – two directions, one anchor at each end. A footnote amounts barely to a minimum degree of ergodics, but in House of Leaves annotation coincides with ‘concrete prose’, to use Brian McHale’s (184) complementary term to the betterknown concrete poetry. Both the link source and the destination are coded to match, but their location in the book goes against the convention. In chapter 9 , or the ‘Labyrinth’ chapter, the typographical balance and hierarchy between the body text and footnotes is disrupted. The notes are distinguished typographically and attributed to narrators on diff erent diegetic levels, each commenting on the preceding ones. However, since ergodics is a particular level distinct from narration (Aarseth 92– 5), I will fi rst deal with the actual use of the text and return to the narrative issues below. Since the text in chapter 9 is laid out in multiple columns, both vertical and horizontal, in geometric patterns varying in shape and size, in mirror image, upside down, and so on, locating the link destination becomes laborious. Source and destination may also be several pages apart. Even though Gunder (222–3) quite correctly points out that in a printed text there cannot be any hidden links, since the link must be indicated by a cue for it to be a link at all, fi nding the link destination is left to the reader (unlike in a digital text, where clicking on the anchor brings forward a new lexia). In House of Leaves the search for the destination is particularly foregrounded, and by the same token an inherent medial feature of the codex, namely random access, is charged with meaning. The bidirectional linking is shortcircuited, however, when there are subordinate notes leading to more notes instead of back to the source. In chapter 9 the superscript number 135 leads from the ‘body text’ to 63 EXPLORATIVE EXPOSURE: HOUSE OF LEAVES 63 footnote 135. This footnote leads both to an external destination– Daniel Hortz’s fi ctitious book Understanding The Self: The Maze of You  – and internally to footnote 129, located three pages back. Note 129 aptly quotes Derrida on the centre not being the centre and diverges to two notes, 130 and 137. Note 130 leads to successive but not forking notes, so fi nally there is an endpoint. Note 137 leads, by way of some intermediate stages, to note 135 where the reader began, thus creating a feedback loop. To be able to continue the reader must step outside the loop, retrace her steps and try a diff erent path. What is more, two mismatched systems have been used to code the anchor cues:one based on consecutive numbering consistent with the rest of the book, the other on the international groundtoair emergency signals, or more precisely their typographical renderings, since the actual signs are meant to be built from natural materials on a scale big enough to catch the attention of an aircraft. (A chart explaining the symbols can be found in a photograph in the Appendix II C.) In the numbered links the source and the destination may be pages apart and they require constant fl ipping back and forth, but they at least conform, by and large, to the numerical order. Page numbers lose their signifi cance– only to be semanticized later– but one can assume that footnote 133 is in the relative vicinity of 132 and 134. The emergency signals have no such order, so the path from source to destination is more challenging, indeed more ergodic. If a typical gamebook produces a particular sequence of scriptons on each reading and so allows or urges the reader to ignore parts of the text, House of Leaves works diff erently. Rather, the implicit urge to go through all the textual material predominates. Lexias containing several links do not in fact off er the reader alternative paths. If a link appears in the middle of a lexia instead of the end, and the reader follows it, she must also make sure to come back and read the rest of that lexia. If a lexia contains more than one link, the reader is not so much allowed a choice as expected to follow all of them as far as they lead, so as not to miss anything. Here ergodic work is related to work ethic, for it appeals to the reader’s sense of completing the task, reading absolutely everything there is in the book. All speculations on authorial intention aside, Iclaim that cybertextuality and the explorative user function have been inscribed in the novel’s discourse as well. Among Aarseth’s strict nomenclature that eschews metaphor, naming a user function ‘explorative’ may seem almost fl ightily poetic. When characters in House of Leaves enter the strange labyrinthine space, these journeys are labelled ‘Explorations’ in the discourse:fi rst ‘Exploration A’ and then numbered from one to fi ve. ‘Exploring’ has been READINGTODAY 64 64 frequently transposed from the literary object language to the critical metalanguage ( sic ) but its cybertextual implications have been overlooked. Furthermore, Cybertext and the scholarly thesis in the novel, The Navidson Record , share some of their theoretical framework. The labyrinth chapter draws heavily on Penelope Reed Doob’s The Idea of the Labyrinth from Classical Antiquity through the Middle Ages , an actual book published by Cornell University Press. Doob’s study and its analysis on diff erent labyrinthine structures such as the unicursal and the multicursal maze have likewise inspired Aarseth’s (5– 9) distinction between texts with only one possible path and those with several. The examples of printed ergodic literature in Aarseth mostly exploit the spatial dimension. Ergodics is based on the layout (Apollinaire’s Calligrammes ) or the disposition and linking of lexias (Cort á zar’s Rayuela , the gamebooks). This is of course apparent already from the defi nition of ergodic literature as requiring nontrivial eff ort ‘to traverse the text’. The emphasis lies on physically passing through the text. This type of ergodics does dominate in House of Leaves as well, but there is in addition a brief passage that requires a diff erent kind of eff ort. At the end of the book, Appendix II E compiles letters written by one of the characters in a mental institution. Paranoid about the staff reading and manipulating her letters, she encrypts one with an acrostic. The key is given in a previous letter, which obviously makes little sense cryptographically, but in the absence of the narratee’s responses, the reader is left to decipher the message. Aarseth does not specifi cally discuss cryptographic techniques, but his most general defi nition of cybertext as those ‘texts that involve calculation in their production of scriptons’ (75) would seem to accord with linguistic steganography, the practice of concealing texts within other texts. Even though deciphering the acrostic does not pose an enormous challenge– it is after all a traditional form of constrained writing, and there are no glitches or further complications– it obviously diff ers from reading straight through and requires nontrivial eff ort. Having determined the novel’s media position, Izoom in on the discourse and the storyworld. This does not mean disconnecting it from the media ecology, however, for here the notion of media is realized as intermediality. Kinekphrastic intermediality According to Irina Rajewsky, the three basic types of intermediality for the analysis of particular texts or media products (instead of more