Hypertext as a social practice

Greco's 1996 "Hypertext With Consequences" argues for a critical approach to hypertext as a social practice.

I'd suggest that appropriating literature and cinema theory's analyses of the role of realism as a style which conceals its own manufacture, and in so doing concealing various other ideological assumptions (gender, colour, race, for instance) would provide one mechanism for this analysis. For instance, simply having the theoretical vocabularly to be able to identify syntagmatic hypertexts as 'realist' provides a context from which it is possible to argue that the hypostatisation of usability, link transparency, and structure embody instrumental conceptions of information that ought to be recognised as one style (or genre) of knowledge production, rather than an ideal.

Adrian Miles: Hypertext syntagmas: cinematic narration with links

A performative hypertext presented by Journal of Digital Information

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

Reading competencies

Since syntagmatic series are largely reader determined it is incumbent on hypertext writers, and developers, to articulate those reading practices necessary to identify and contextualise paradigms of reading.

Successful reading requires the recognition of paradigmatic choices, not only at the simple grammatical level of the sentence, but also at the larger level of narrative episode and generic convention. When readers read poetry, or literature, or even the newspaper, a set of reading competencies are utilised that include an understanding of the genre in relation to other possible genres or styles, that is, that one style exists in a paradigmatic relation to others, and this is largely where the signficance of a particular work is determined. Within hypertext such readerly competencies are much less developed, resulting in a misunderstanding of hypertext pattern, reading, and writing, simply because the paradigm against which hypertext is read and interpreted consists of a normative and potentially singular notion of structure as effiency and economy.

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

It is clear that link source plays a fundamental role to the context of any particular pattern.

For example Joyce's cycle could as easily describe the experience of using the Internet Movie Database, as Afternoon: a story, yet clearly in each case the effect of the cycle is radically different. Similarly a tangle could describe Simpson's Memex project, parts of Amerika's Grammatron, or Miles' 1996 Hyperweb, yet each are fundamentally different hypertexts.

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

The consideration of hypertext structure as consisting of 'whole' syntagmatic series, which in turn become the paradigmatic sets available, has complex implications which require further research.

It would appear that the relation of one node to another, via a link, while significant, gains greater currency when considered from the point of view of a discrete syntagmatic segment (which could, in principle, consist of two nodes). However, such an approach discounts the origin of the link in a manner that is probably untenable - in cinema this is irrelevant as the point of connection is the end of one shot and this is, currently, the only point of connection. Obviously a hypertext provides a point within a node from which connection is possible, and thus for a syntagmatic series to be formed, and so the relation of the link source to its destination also forms a syntagmatic series.

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Sequential movement through time

It has been pointed out that the mode of language is fundamentally one of sequential movement through time.

It follows from this that each word will have a linear or 'horizontal' relationship with the words that precede and succeed it, and a good deal of its capacity to 'mean' various things derives from this pattern of positioning. In the sentence 'the boy kicked the girl', the meaning 'unrolls' as each word follows its predecessor and is not complete until the final word comes into place. This constitutes language's syntagmatic aspect, and it could also be thought of as its 'diachronic' aspect because of its commitment to the passage of time.

But each word will also have relationships with other words in the language that do not occur at this point in time, but are capable of doing so. The word, that is, has 'formulaic' assocations with those other words from among which it has, so to speak, been chosen. And these other words, 'part of the inner storehouse that makes up the language of each speaker' (p. 123) - they might be synonyms, antonyms, words of similar sound or of the same grammatical function - help, by not being chosen, to define the meaning of the word which has. (Hawkes, pp. 26-7.)

----------


Download ebooks on http://www.frenchtheory.com/See that post with different algorithms in metaboleSee the journal French Metablog with today different posts

Real collage

Upon hearing the assertion that hypertext should be thought of as collage-writing, Lars Hubrich, a student in hypertext and literary theory course, remarked that he thought "montage" might be a better term than "collage."

He had in mind something like the first OED definition of montage as the "selection and arrangement of separate cinematographic shots as a consecutive whole; the blending (by superimposition) of separate shots to form a single picture; the sequence or picture resulting from such a process." Hubrich is correct in that whereas collage emphasizes the stage effect of a multiple-windowed hypertext system on a computer screen at any particular moment, montage, at least in its original cinematic meaning, places important emphasis upon sequence, and in hypertext one has to take into account the fact that one reads - or constructs - one's reading of a hypertext in time. (Landow 1999, pp. 169-70.)

---------------------

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Structure in absence

Linked text within a node gains significance by virtue of what is not linked.

This is perhaps the most literal expression of the paradgimatic in terms of link node hypertext. This is different to Bernstein's missing link pattern, where it is the absence of an expected or presumed link that generates a structure-in-absence. Here it is simply the decision to link from one word, phrase, or image, rather than another, that is actually a paradigmatic decision. That the majority of hypertexts privilege nouns as link origins simply demonstrates the paucity of hypertext design and writing.

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

Adrian Miles: Hypertext syntagmas: cinematic narration with links

Metaphor and metonymy

Jakobson sees metaphor and metonymy as the characteristic modes of binarily opposed polarities which between them underpin the two-fold process of selection and combination by which linguistic signs are formed . . . .

Thus messages are constructed, as Saussure said, by a combination of a 'horizontal' movement, which combines words together, and a 'vertical' movement, which selects the particular words from the available inventory or 'inner storehouse' of the language. The combinative (or syntagmatic) process manifests itself in contiguity (one word being places next to another) and its mode is metonymic. The selective (or associative) process manifests itself in similarity (one word or concept being 'like' another) and its mode is metaphoric. (Hawkes, pp. 77-8.)

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

Marked links

In hypertext one also has to decide where within a node a link is to appear, and it is not unreasonable to think of a densely linked node as being 'granularised'.

In some ways this is the experience of such work where links are visually marked - a page with many links is a different cognitive experience to a page with no visible links.


Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Hypertextual minimal unit

This is not particularly suprising as it is apparent that hypertextual fragments (nodes) can, like cinematic shots, be arranged in numerous sequences and that there is no intrinsic and necessary grammar to this (Miles, 1999).
An earlier emphasis on the cinematic nature of the connection between hypertext nodes (Miles, 1999), while important, does not provide an adequate account of the production of narrative (fiction or nonfiction) sequences in hypertext. Like the cinema, it is apparent that hypertext requires a minimal unit constituted by the set of nodes conjoined. That this has been the case in much extant hypertext fiction is clear. Such a process has been much less common in what has been routinely described as hypertext nonfiction.

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Textual temporal dimension

Texts have essentially a temporal dimension.

You take in a painting in one glance, but you read a text over time; film in that sense is closer to text than painting, and the filmic term "montage" would be better for what happens when a text makes use of disparate, found, randomly combined elements. The only true "collage" effects in literature, i.e. the presentation in the same moment of perception of disparate materials would be certain "simultaneities," such as Dada and Merz and other, later sound-poets presented. (Landow 1999, p. 166.)

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Linkable items

One canot make connections without having things to connect.

Those linkable items not only must have some qualities that make the writer want to connect them, they also must exist in separation, apart, divided. As Terence Harpold has pointed out, most writers on hypertext concentrate on the link, but all links simultaneously both bridge and maintain separation. This double effect of linking appears in the way it inevitably produces juxtaposition, concatenation, and assemblage. If part of the pleasure of linking arises in the act of joining two different things, then this aesthetic of juxaposition inevitably tends towards catachresis and difference for their own end, for the effect of surprise, and sometimes surprised pleasure, they produce. (Landow 1999, p. 159.)

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Desire to link

There is some expression of force outside of the shot in the cinema's desire to cut the shot into fragments, and that a similar force is present in hypertext and expressed as a desire to link.

This 'force' is always outside of the shot, or the node, or the link. It is sought by the link, the link is an expression of appropriation of the force that writing and the link is a response to, but it is never adequate to this outside. Outsides remain, always, outside. Harpold's work on links and diversions - the manner in which links are always detours around what is sought, is extremely relevant here.

Adrian Miles: Hypertext syntagmas: cinematic narration with links

A performative hypertext presented by Journal of Digital Information

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

assertive image

Image is almost always assertive - and assertion is one of the great "modalities" of actualization, of the semic act.

It appears therefore that the paradigmatic category in film is condemned to remain partial and fragmentary, at least as long as one tries to isolate it on the level of the image. This is naturally derived from the fact that creation plays a larger role in cinematographic language that it does in the handling of idioms: To "speak" a language is to use it, but to "speak" cinematographic language is to a certain extent to invent it.


Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Syntagmatically oriented hypertext

Hypertext writing and systems that emphasise 'usability' (where usability seems to assume ease of use as a positive attribute for any hypertext), place an emphasis on the syntagmatic.

This emphasis on the smooth flow of links into nodes describes a highly linear reading experience. Interestingly, in Bernstein's examples most of the patterns that would probably relate to 'usability' are represented by highly linear images, the sieve for instance. However, the emphasis in the syntagmatically oriented hypertext is not, as might be thought, on simple patterns (any variety of pattern could be formed by the reader) but on linear continuity.

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

What is grammar

In the context of this essay grammar is understood to be that system of rules, what semiotics describes as langue, that one necessarily ascribes to in a language community. Without this grammar we are condemned to a series of idiolects. To not use grammar (whether a formal gramamr or a colloquial grammar) is to risk intelligibility and comprehension in quite fundamental ways, and it is this that makes a grammar a formal system. Hypertext (and classical cinema) may have commonly accepted styles but these are not grammars in the sense of langue. They do not constitute the possibility of utterance in advance, as is the case in our usual experience of language.

Adrian Miles: Hypertext syntagmas: cinematic narration with links

A performative hypertext presented by Journal of Digital Information

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

Link use

The issue is not one of predetermining what links mean but of recognising the contingent and contextual nature of link use, as readers and writers.

This is common in poststructural approaches to discourse, which no longer place much store in hierarchies determined by formal categories, and is particularly the case in hypertext where it seems reasonably clear that readers find connection by virtue of the link.

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

What is a node

Should be obvious that nodes are not reducible like words can. Of course a node can be reduced to its component parts (if they are thought of as writing) but that is hardly what constitutes the node. A node could just as easily contain a sound, a film, an image, a letter, or any combination of these.

Adrian Miles: Hypertext syntagmas: cinematic narration with links A performative hypertext presented by Journal of Digital Information

Units, shots, words

The cinema, like language, has much to say, but, like sign boards, it actually escapes the first articulation. It proceeds by "sentence," like sign boards, but, like verbal language, its sentences are unlimited in number. The difference is that then sentences of verbal language eventually break down into words, whereas, in the cinema, they do not: A film may be segmented into large units ("shots"), but these shots are not reducible (in Jakobson's sense) into small, basic and specific units. (Metz, p. 88.)

Adrian Miles: Hypertext syntagmas: cinematic narration with links A performative hypertext presented by Journal of Digital Information

Autonomous segments

Hypertext, unlike traditional cinema, provides for nodes that can be reused, or reappear, in any particular pattern, and this practice of reuse or repetition is one of the principal methodologies employed in hypertext writing (and reading). This means, obviously, that the autonomous segments that can be formed in hypertext, while falling into many different types, also have the feature of a particular node or even minor series forming a significant part of a completely different autonomous segment.

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Context and connections

An important implication, as Metz notes, is that if it is the development or articulation of autonomous segments that are fundamental to narration in cinema, then the meaning that accrues to these segments is highly contextual and not inherent within the connections themselves.

Download ebooks on http://www.frenchtheory.com/
See that post with different algorithms in metabole
See the journal French Metablog with today different posts

Adrian Miles: Hypertext syntagmas: cinematic narration with links A performative hypertext presented by Journal of Digital Information

Choice of words

Any element of a film is already linguistically meaningful.

They are more like statements (the cliche that "a picture is worth a thousand words" comes to mind), and the paradigmatic process - choosing which from the paradigmatic series to place in the syntagmatic chain - is considerably diluted in film as the paradigmatic field is made virtually trivial, and the syntagmatic ordering is similarly weakened. In other words, in a sentence the choice of words is constrained to some extent by what choices are available within a vocabulary, and then by the grammar which constrains what can appear where.
When composing a film image there is no substantive paradigmatic series that the sequence most be chosen from within, and then once a shot is composed, its location within a syntagmatic series (a sequence) is similarly much more open than that offered by a linguistic grammar.

Download ebooks on http://www.frenchtheory.com/

See that post with different algorithms in metabole

See the journal French Metablog with today different posts

Rules of combination

As Metz makes abundantly clear, there is no real paradigmatic axis in the cinema, and this is also so in hypertext - when I write hypertext I can link more or less to anywhere and there are no necessary rules of combination that exclude or promote particular destinations. This suggests that it is within the series of combinations formed by links, what Metz in the cinematic case characterises as autonomous syntagmatic units, that hypertext structure resides.

Adrian Miles: Hypertext syntagmas: cinematic narration with links A performative hypertext presented by Journal of Digital Information