The largest theoretical difference between presenting history online and in traditional forms is the ability to create hypertextual documents by inserting links into the text.  These links then allow for reader participation in the structuring of a text, which has the effect of decentering the author.  They also enable the reader to read according to what  Vannevar Bush  has termed associational thinking.  Furthermore it is possible to link to other material online, which can break down the borders and 'completeness' of a work.  All of these transformations are, however, contingent on the insertion of links into documents, and this section of the analysis will assess the extent to which the two sites can be said to be hypertextual.

Both the Lost Museum and Uncle Tom's Cabin have been classified by their designers as archives, and despite the doubts about this classification raised in Section One, it is worth considering how linking theoretically affects the archive.  According to Mike Featherstone , 'hypertext works on a less hierarchical, more lateral view of knowledge' than traditional archiving which 'is a kind of reason concerned with detail, [that] directs us constantly away from the big generalization, down into the particularity and singularity of the event'.  Restated, the difference is that hypertext allows one to move horizontally amongst the archive to find points of comparison and resonance, whereas the traditional archive is structured so that one must always 'descend' from the general down into the specific.  It is worth considering where the 'archives' being analysed fit between these two competing models.

Interestingly, both sites link to the archive with the phrase 'browse the archive'.  This rhetoric implies that it is possible to skim parts of the archive and follow the associative links that one would make when one was browsing a shelf for example.  It therefore implies that all the material is visible from any point, and that one can browse vertically or laterally.  In reality, however, it is not possible to do this in either site, a lack of links means that one must descend up and down a hierarchy to view material and cannot browse across organisational categories.

1) Linking Within the Site

The structure of Uncle Tom's Cabin seems to reflect traditional archive reason.  From the browse site map, one selects a category.  From the category 'homepage', (and  the rhetoric of 'the homepage' reveals how discrete each section is), one can choose from any number of categories.  This process of picking from lists that become ever more specific continues until one has reached the lowest level of the archive and is presented with a solitary source.  For example, one could pick the African-American Responses category.  From its  homepage  one could select  newspaper commentaries  and then pick  Frederick Douglass' paper .  Here one is presented with a list of articles and when one is selected it is presented and the 'bottom' of the archive has been reached.  The point is that at each of the five stages of this process, with the slight exception of the final level, discussed below, one never has the opportunity to move across the archive to a section of equivalent specificity.  It is impossible to get to another African American paper from Frederick Douglas', for example, without ascending to the required level of generality and then descending down another route.  This structure clearly reflects traditional archive reason and not hypertextuality.

This lack of links is perhaps most evident when one looks at the way that individual sources are presented.  Returning to the first article  from Frederick Douglas' Paper, one can see the template for presenting textual sources that is used throughout the site.  There are six links on the page.  One is at the bottom, and provides a link to bibliographic information.  Of the five that are in the side bar, one takes you back to the site map, one takes you to the 'home' of the source, in this case back to the Frederick Douglass' paper site, one provides a search function (see below) and one takes you to a searchable list of relevant names that provides very short biogarphical details for a few persons.  The final link is to the 'next' article in the archive, and it therefore allows horizontal movement across the archive to an extent.  However, if one hits the next button whilst looking at the  last article  from Frederick Douglass' paper, one is returned to the Frederick Douglass' paper page again.  The analogy is to a folder in an archive, one descends until they receive the folder of paper clippings, which can be moved through laterally because they are sufficiently specific, but cannot move horizontally beyond the bounds of the folder.

The lack of links provided from actual sources is very disappointing and hampers the utility of the site in making sense of Uncle Tom's Cabin.  The site contains playable versions of the hymns in the text to give the reader an 'equivalent experience' to the contemporary reader who could have 'heard' the songs as they read.  Given this aim, it is frustrating that it is impossible to link to the hymns from the text itself.  The lack of links appears to reflect a lack of thought regarding their potential uses.  When Railton is attempting to explain the  Bible references  in the novel, he has a table that allows the reader to find the origins of the bible references by chapter.  To see the original source, however, one cannot simply click on the reference, but must go to a second table that contains the relvant chapters of the bible in their entirety.  This clumsy interface reflects a lack of thought as to the potential of linking: it would have been possible to link to the exact quotation from the chapter table.

There are two exceptions to the general rule that there are no lateral links within the site.  The first is that there are links from a small number of images to the texts from which they are quoted ( an example ).  The second is that from some of the contents pages, for example the Frederick Douglass' paper page, there are links to parts of the source that are contained within a seperate part of the site, most often to the reviews sections.  Whilst these do allow for some movement across the site, they really only avoid the need for putting more than one version of the same document up and do not allow the reader to follow associational trails of thought.

The archive in the Lost Museum has a similar structure to that presented above, albeit on a much smaller scale (see Section Four).  From the archive homepage one chooses a particular archive, and here the rhetoric of seperate archives casts them as more distinct entities than Uncle Tom's Cabin.  From the archive mainpage one is generally presented with a list of links to sources, but occassionally it is necessary to descend to another level by choosing between image and textual sources.  The point remains however that even in the two or three tiered archive, one cannot move laterally but must ascend or descend through the various degrees of specificity.

It is worth briefly considering the way that the particular archives are linked to the other parts of their respective sites.  It is possible to link to the Lost Museum Archive from the various exhibitions within the museum but it is not possible to link from the archive into the exhibit.  This may partially be a product of technical concerns, but is still establishes a structural hierarchy in the site.  It is clear that the archive is meant to support the museum, and not the other way around.

The linking between the intepretive mode and the archive on Uncle Tom's Cabin is similarly unilinear.  It is impossible to move from the archive to the interpretation, which reveals the primacy of the 'pure' archive (see Section One) in the site.  It is also worth considering the links from the interpretive 'exhibits', (in reality they are no different to articles or essays), into the archived material.  Generally this is very disappointing, Jim O'Loughlin's essay, for example, contains only 5 links to the archive, and Stephen Railton tends to link to only source excerpts which could easily be quoted or footnoted.  Whilst the essays do contain a number of image thumbnails, there is no theoretical reason why these could not be included as images in a traditional article.  The only wholesale linking that these articles use is in a list of the sections of the archive they used for research, but these are spatially seperated from the text as they are located on a seperate web-page.

  Note: the interpretive mode opens in a new page that doesn't display the URL so to look at these exhibits you will need to follow the links from the  site map
 
 

2) Search Engines

Even though the sites do not move beyond traditional archive reasoning in terms of their linking, the ability to perform full text searches through the site allows for the potential to transcend the hierarchical structure of the site.  Although the Lost Museum does not yet have a search function, it is interesting to look at how the search function on Uncle Tom's Cabin in fact preserves rather than destabilises the hierarchical structure of the site.

According to Railton, the search mode can use or 'cut across' the organisational categories of the site.  When one looks at the search screen it is worth noting how it is possible to limit the search to maintain the categories by searching within particular sections or subsections only.  Furthermore, the default search screen only searches the primary documents.  To perform a full search of the entire site, including the intepretive essays, one must click on the small, and apparently insignificant 'HTML search' button on the sidebar and be taken to a different screen.  It is obvious then, that the search engine is intended to maintain the organisational categories and divisions that have been established by the links on the site.

However even when one attempts to 'cut across' the organisational categories of the site by conducting a full search of the primary documents, one cannot escape the divisions that Railton has established.  Typing 'Jenny Lind' into the search window results in 33 hits.  Yet these 33 hits are broken down into the broad categories of the site, we are told that 20 of them are from the pre-texts section, 8 are from the responses section 3 are from the other media section and 2 are from the Stowe section.  To actually get to the sources themselves one must descend down a skeletal version of the same hierarchical structure as the browse mode, by moving into the minstrelsy section and then into a particular song book and so on.  The fact that the search findings page is called the 'Collection Top Level' reinforces the hierarchical structure of the site and the search function.

     Note: the search mode opens in a new page that doesn't display the URL so to look at these pages you will need to follow the links from the site map

3) Linking Outside the Site

To fully allow for associational reading and to deconstruct the boundaries of a work it is necessary to link to material outside of the site.  There are no links on Uncle Tom's Cabin to external sites except to the institutions and individuals responsible for designing it.  The Lost Museum contains two links to external material that are contained in a  bibliography  to text based reading.  By placing the links within the 'further reading' rhetoric of a list of offline reading, the designers do a great deal to limit even the minor impact to the integrity of the site posed by these external links.

It is unfortunate that neither site links beyond their own site given that there is a great deal of material online that could be useful for studying 19th century culture.  Most obviously both sites could have linked to the wealth of 19th century source material at American Memory.  Similarly, it would make sense for the two sites, which are both looking at mid century cultural forms, to link to each other.

Conclusion

It is clear that both sites feature a relatively paltry implementation of linking.  As such both archives reflect a traditional mode of  hierarchical archive structure as opposed to a lateral, hypertextual form.  The lack of links, both internal and external, also maintains the integrity of the boundaries of the sites and the centrality of the designer.  As such, it is clear that both sites are very far from an ideal form of hypertextual history.

Introduction
Section One: What Are The Sites Trying to Do?
Section Two: Hypertextuality
Section Three: The Use of New Media
Section Four: Taking Advantage of the Scale of the Internet
Conclusion