When is a document a document?

I read a story that delightfully illustrates how 'contingent', or perhaps more accurately, how 'convenient' the meaning of technology is. In Japan parties have started campaigning for the upcoming national election. There is a law that limits the overall number and distribution of election material such as posters and flyers during the campaigning period. A recent move by officials to apply this law to Internet campaigning has involved a rather unique interpretation of the definition of a "document". According to the interpretation, a "document" can include a single web page or email and each "download" of a "document" is seen to be equivalent to a single physical exchange of a file, such as handing someone a brochure. The article describes how those who defy the law and who go ahead and update their home pages or disseminate emails risk being stripped of the votes cast in Sunday's pol and jailed for two years or fined up to 500,000 yen ($6000). (SMH 9/9/05)

Campaigners immediately responded to this by putting a quarantine on their web sites. The article goes on to explain that, "The threat has caused the home pages of the main parties, usually dynamic, to be frozen in time, appearing the same as they were on August 30, the first official day of campaigning...While party platforms can be downloaded, there are no updates, interviews or transcripts, and no way for voters with questions to get a reply." (SMH 9/9/05)

There are a number of questions about the technical understandings that must come into play to equate a web page with a document and a download with handing someone a brochure. I will return to this point about the properties of technology later. What initially impressed me about this story is how it illustrates that the definition of a technology, in this case a "document", is directly connected to an interpretative operation. This operation defines not just what we understand the technology to be, but also what it does. Woolgar offers us the anti-essentialist perspective;

"The objective effects of technology are anything but self-evident. In each case the effect of technology necessitates some form of human interpretation; even if debate about such effects may be brought to (temporary) closure, and a consensus constructed around the alleged effects, this consensus is socially constituted, not the result of an autonomously and exogenously imposed truth." (p138)

The interpretative operation can also be seen to follow the stages of translation in actor-network theory as described by Woolgar in "The Machine at Work".

1. The nature of technology is problematised (what is the definition of a document?)
2. The meaning of a technology is translated to mean something else (a web page and an email are documents.)
3. The meaning is stabilised (in law and by the policies of authorities) and fourthly,
4. The new meaning is mobilised (through the threat of action by authorities and the response of campaigners to this threat).

So this case demonstrates that any understanding of technology, what it is and what it does, must be viewed firstly, as an interpretative operation that endows it with meaning and secondly, that the operation is a translation, or a strategic set of moves that involves a transfer of meaning (and power).

Although this account is pretty comprehensive, I am left feeling uneasy. It is not just the meaning of the object - the document - that is problematised and then re-interpreted but also the act of exchange itself - the download. Here we are to understand that downloading can be equated with handing someone a brochure. My initial reaction to this was that it was clearly the work of a bunch of lawyers who no nothing about IT. In both cases we are required to suspend certain technical properties that make the equivalence extremely difficult to maintain and to enforce. An anti-essentialist perspective would argue that the inherent technical properties are not inherent to the technology but have come to be known through a social consensus involving technicians and programmers etc.

But to my mind, downloading is really nothing like handing someone a brochure. Perhaps my IT background is clouding my reading but surely there has to be some acknowledgement of attributes that belong to the object at the time of the interpretative operation. Surely this bears greatly on the conditions of the interpretation and its success.

When I open my browser and visit a home page, I am downloading, or more accurately, my browser is loading the data that constitutes the web page through a series of requests and replies to the server hosting that page. The download is more correctly understood as a stream of queries and replies. It is not a single transaction of a single contained piece of information. Secondly, given this definition, each time I point my browser to that web page, another download occurs. And yet, this is not equivalent to me returning to the campaigner standing at a street corner and being handed another brochure because, unlike this scenario, re-loading a web page is not cumulative. I am left with no more data 'in my hand' so to speak after re-loading the page.

In the story about Japan, the interpretation seems to have accounted for this cumulative property by equating it with changes to the web page. And yet there still seems to be a mis-match of equivalence. What I can do with a bunch of brochures in my hand is completely different to what I can do with a re-loaded web page on my browser even with updated information. In this case, the mis-alignment of equivalence has resulted in a kind of slippage of meaning, rather than a complete substitution. While the ANT approach might read this as an effect of the incomplete enrolment of actors in the network, I am left unconvinced.