Hypertext is a term first used by Theodor H. Nelson in the 1960s to refer to 'non-sequential writing - text that branches and allows choices to the reader, best read at an interactive screen' (Nelson 1980, 0/2). In contemporary parlance it has come to refer to the electronic text used to write, publish and read documents on computers. The terms 'hypertext' and 'hypermedia' are frequently used interchangeably; these terms refer to "texts composed of blocks of texts... and the electronic links that join them, using visual information, sound, animation and other forms of data" (Landow 1992).
Much has been made, in the growing literature about the nature of hypertext, of the potential of electronic writing to destroy the primacy of the single autonomous author (Joyce 1995; Landow 1992; Lanham 1993; Taylor and Saarinen 1994a). Certainly, in my role as the 'weaver' of the Levi Jordan Plantation web site, I have found that my own presence is dissipated and dispersed. In any hypertext, there are few guideposts about beginnings, middles and ends - readers decide which links to use, and in which order. The logic of the Net is not unilinear, nor is it un-linear (though it is, arguably, multilinear).
However, within the context of the Web and CD-ROMs, readers are not able to create their own links for subsequent readers to use. Within these contexts, the arguments made by many hypertext authors about the 'death of the author' (and about the so-called democratic nature of hypertext) (Bolter 1991; Rheingold 1991; Rheingold 1993) are much less convincing. It is this aspect of hypertext linking that I wish to address in this essay.
On the Internet and in CD-ROM environments, readers may indeed choose their own pathways through the links we provide, but the links are 'pre-set' by the time the web site or CD-ROM is published. These links, when examined critically, reveal what the creators of the hypertext document think about the relative importance of different chunks of text, images, kinds of data - and what they think about different points of view. As we make decisions about which links to provide and where to provide them, these links become, in effect, powerful rhetorical, political and social devices.
Some of the material available for the Levi Jordan web site content included transcripts from oral history interviews that were conducted with the white descendants of the owners of the plantation. Another set of material included data from an oral history project conducted amongst descendants of the African-American residents. We also had access to a diary, written by one of Levi Jordan's granddaughters from 1858 to 1874. We had copies of more recent historical research about the people who owned the plantation in 1888, when it appears that the tenant farmers left the site very suddenly - and left their possessions behind them. These plantation owners were very involved in white supremacist activities of the period and seem to have been responsible for the rapid departure of the tenants from the site - and, indirectly, responsible for the archaeology itself (Barnes 1998). Finally, we had the archaeological data, which comes primarily from excavation of those hastily abandoned quarters.
While each set of material contains much information that is particular to it alone, each also includes references to many of the same events, people, and objects. For example, the diary mentions some of the enslaved people by name, and the archaeological data indicates how enslaved people used various aspects of material culture. The oral history data refers in turn to some of the same kinds of activity - and so on. The web site contains many examples of material which, in 'real' space and 'real' time was linked - people had relationships with each other that merged and converged at numerous points in their lives.
We could, if we wished, have no links from the material about the European-Americans who lived on the site and the material about the Africans and African-American residents. We could publish each type of interview data (from both African-American and European-American descendants) but neglect to provide internal links from each set of material to different kinds of material that mention the same or similar things. We could provide only one link into and out of each set of information. We could also 'hide' certain kinds of 'politically charged' documents deep within the site, with only one entry and exit point into and out of those documents.
What would be the effect of having these different types of information operating independently of each other? This is frequently what happens in traditional, unilinear, hierarchical presentations of data when people emphasise one 'history' over another. For example, many 'Black History Month' presentations (a large component of educational curricula in the United States) do not discuss the many ways that whites and blacks interacted and influenced each other in the past. Similarly, traditional 'fancy house' plantation tours do not usually discuss the lives of enslaved people, and sometimes even professional historians take an 'either-or', rather than 'both-and' approach to the writing of history.
If we did not provide internal links from material about the enslaved people' lives to material about the owners' lives we would, first, fail to illustrate the contextual nature of knowledge production - but something even more important would happen as well. We would perpetuate the stereotypical view that there was little meaningful interaction between the two groups of Jordan Plantation residents. Traditional binary modes of interpreting human behaviour would be reinforced. We would be masking the multiple, overlapping, and complex roles that all of these individuals experienced.
This would not only be a rhetorical act, we maintain that it would also be a political and social one. The Net does offer archaeologists many opportunities to deconstruct the ideologies - the 'taken-for-granteds' - that exist in writing about archaeology and history. However, unless we construct web sites, and the links within them, critically and reflexively, those opportunities will be lost. This paper gives some ideas of ways in which the Jordan project collaborators have attempted to capitalise on these opportunities.
Go to the Levi Jordan Web Site
© Internet Archaeology
URL: http://intarch.ac.uk/journal/issue6/mcdavid/contextuality.html
Last updated: Wed Apr 28 1999