The Duration of the Present and the Risk of Not Telling Large Stories

A Comment on The Plenary EASST-Debate on Technological Determinism

At one of the plenaries at the EASST-conference in Lausanne this year Sally Wyatt asked us – the STS-community – to rethink “technological determinism”, not because we need to redo “our” understanding of its falsity, but because it is still alive and kicking “out there”. As a social fact, as an epistemic reality, an instance of what anthropologists recognise as an emical category, it still makes a difference. On the train back home from Lausanne, I picked up on Wyatt’s invitation, and started to think, ontologically, about technological determinism and the nature of time, or times. Not of “spaces”, as the conference was dedicated to, and that much of STS-research has been so devoted to for a long time now. I thought about times and their durations, on that Swiss train that I will return to in a moment, and the Whiteheadian understanding of time as a succession of durations rather than a succession of moments. Technological determinism is the belief in the technological inevitability of historical development, and thus, somehow, the inevitability of the future, given a particular technology. To understand the duration of the present then, is crucial if we want to say something about technological determinism, because, somehow, at the end of the present lies the future.

I also started to think about the politics of “technological determinism”, and in particular the politics in the widespread STS-scepticism towards this large, fuzzy beast. Perhaps we are not always political responsible in our rejection of it.

The event at EASST 2006 The event that set off this little cosmopolitical speculation was the plenary session around Stephen Graham’s paper Software-Sorted Technologies: On Space, Technology and Inequality (Graham 2005). The paper was commented by Sally Wyatt and later discussed by the plenary and particularly commented upon by John Law. Here is a summary of the central arguments of the event:

Stephen Graham, invited to be “external” to STS, first presented his paper on “software sorted technologies”. (Or really, he did not present his paper, he had it “ghost read” by one f the conference-organisers, because he himself was home with a flue.) In short, Graham’s paper explored, quoting its abstract: “the central role of computerized code in shaping the social and geographical politics of inequality in advanced societies …” (Graham 2005: 1) It did so by presenting three different cases. The clear political message of the paper was that the production and reproduction of various social inequalities, by means of software and digital technology, is worrying. The paper is a warning of a present and possible future risk. The discussion afterwards revolved around last of Graham’s three cases, future, computerized face recognition, using public surveillance cameras. These cameras are most often monitored manually. That is, some person has to sit down and actually look at the filmed material, live or recorded. There is however an emerging technology that can recognise people’s faces automatically, by means of computer programs. Wanted persons can thus be found automatically. This technology is developed to find people whose faces have already been scanned and stored in databases, but it is also developed to find “suspect” persons in general, using artificial intelligence to look for generally “deviant” behaviours. Moreover, the cameras “tends also to be used”, to quote Graham, to scan people’s faces without their consent, presumably to build databases of faces, later to be used to find people who were not known to be doing anything wrong when their faces were scanned. There is a “very real risk” that camera systems that are now isolated will be integrated into “massive and geographically-stretched facial recognition CCTV systems” (Graham, 2005: **)

After Graham’s presentation, the “internal STSer, Sally Wyatt for this occasion, presented what she called the “usual STS-criticism” of such stories, namely that these stories tend to take the advent of some technologically produced future for granted, and that they are de-contextualised. That is, they are typical stories of technological determinism. However, rather than just dismissing Graham as “wrong”, Wyatt made the general point that we need to understand the way in which such notions of determinacy works, without judging their ontological truthfulness.

In the plenary discussion afterwards John Law further discussed the problems of the possible determinism of Graham’s paper. I’ll only repeat his last argument here, concerning the critical stance of Graham. Law was worried that by telling such large scaled “critical stories” one runs the risk of colluding with the advocates of these systems, because one basically agrees about what these systems can “really do”, and in how massively they can do it. The only difference between the advocate and the critic is that the former wants the system whereas the latter fears it. But they agree on what the system really is, will be or can be. This collusion between the critic and the advocate, then, closes the a space for a possible criticism, a criticism that might constructively try to see other realities, more desirable realities. Thus, the mega-machine that Graham fears is unattainable to the common deadly who might want to make a contribution – small but possible – to a better world.

I do very often share this argument with John Law. To take a personal example: I am worried about the way in which large computer corporations (notably Microsoft) work to monopolise the communication standards, the infrastructure, of the Internet. But I have not started to study this by investigating Microsoft. Rather I have started a study of some parts of the diverse group of computer activists, often known as “hackers”, who work relentlessly and in large varieties of ways – commercially, technically and politically – to undo that monopoly. So I am STSer enough to believe in the usefulness of the small stories. By telling stories of hackers and Free Software, I hope to help redo received stories about who is big and small. Moreover I hope to do that in ways that do not reify sides of good and evil, as it is now the case that the “hacking” of Free Software includes million-dollar investments from companies like Sun Microsystems, Novell and IBM, in cooperation with large public bodies, such as the Brazilian government. But still also in cooperation with non-profit organizations such as Free Software Foundation.

But then, I am not sure if it is always a good idea to tell the small stories rather than the large ones. I’m not sure it is always the responsible thing to do, and I am not sure if it was a bad idea of Graham to tell his story the way he did. Here, at least, is a partial defence of his paper. First I’ll do a little Whiteheadian inspired discussion of what technological determinism may be, and not be. Then I’ll look at some alternative political implications of writing in the style of Graham rather than in the style of Law.

The ontology of time To Whitehead, time, or rather space-time, is always made up of “atoms” that have a real substantial existence. Points in space and moments in time are just limits, mere relations, in themselves with nothing to make relations between. Points and moments then, exists as limits between something and something else, some things substantial, some things that have extensions in space and time. Time is not made up of a moment that travels on a time-line. It is made up of “chunks” of space-time, that is of actual events (Whitehead 1926: 158-159). These “things” are atomic to Whitehead, they are “building blocks” in a monadological universe. But unlike the undividable classical atom, the atom of Whitehead is always dividable. Time and space can always be divided into smaller parts, but these parts will always be substantial things, “gatherings” as Bruno Latour would say, always drawn together and always substantial.

Now, the reader may not believe in this Whiteheadian conception of space-time, but this is not the place to run a long argument in to show the soundness of this metaphysics. So I will just have to ask you to entertain the metaphysical premise of seeing space-time as made up of substantial atoms, atoms that are always drawn together in events, “prehended” in Whiteheadian vocabulary, and that are therefore always dividable.

If time-space atoms may vary in size, they may also be large. The question, in relation to technological determinism, then, is this: How long does the present last? How large is the present? This is a relevant question because if we want to understand what the allegedly inevitability of the future might be, we need to sort future from present. Her is an extreme case to think with, in thinking about technological determinism:

I left EASST 2006 to go to Geneva airport by train, and started doing this speculation. The train was scheduled to leave Lausanne at 16:17 and to arrive Geneva Airport 43 minutes later. I had reasons to believe it would be fairly on time in Lausanne and actually use about 43 minutes (being in Switzerland …). The train was on time and it used 43 minutes.

Was I being a “technological determinist” in believing in the inevitable and technologically determined unfolding of a future, a future “determined” by the Swiss railway system? No, I think I was not. This is not what we mean by that term. So, then, how do we analytically separate the possible or impossible inevitability of the future from a case such as the stability of the Swiss railway system? I suggest we do it by separating the (alleged) inevitability of the future from the inevitability of the present. There is such a thing as the inevitability of the present. The present is inevitably here. Thus, when travelling with a train in Switzerland, the whole trip is part of a present. The event or duration, the “chunk of time”, that is the present in relation to such a trip includes the whole trip. This present is not a given. It is an achievement, an achievement of the Swiss railway system (In Britain it is not the same kind of achievement, and the present is a different thing when travelling with trains in the UK). Doing the trip from Lausanne to Geneva was an unfolding of a present. And the present has to unfold, as it is not a moment, but a duration.

The making into unity, the work and achievement of a “gathering” which by STS-researchers is studied in detail – such as the one Latour describes , for example in the unity (and disunification) of the Challenger space shuttle (Latour 2004), or the “coordination work” that Mol describes in the unification of a disease (Mol 2002) – are not only spacial unifications. They are also a temporal unifications, “prehensions”, graspings into unities of durations, in Whitehead’s terms. They are unification of presences. Presences can always be divided in smaller presences, as atoms are dividable. Thus presences can also be unified into larger presences. And in relation to the arbitrary scale of the life of a human being, these presences might well be long.

If, then, we want to criticise technological determinism, we should not criticise descriptions when they describe an unfolding of a present, even when that present consists of a long chunk of time, like, say, 20 years (“cars and roads will still be the dominant machinery of transportation in 20 years time”).

If, however, we want to argue that a particular unfolding of events is an unfolding of a present time, we have to argue the case empirically. It cannot be assumed. And the arguing may be difficult and uncertain. There are epistemic uncertainties: often we may not now – because we have not figured it out – if some event belongs to the present or the future. And there are ontological uncertainties, notably in accidents when the future interrupts the present: An accident occurs to the train, even the Swiss one. To the dedicated determinist these two uncertainties are really the same: “Perfect knowledge” would not only annihilate epistemic uncertainty, but also ontological uncertainty, as all possible accidents could be foreseen and avoided. Thus, to the dedicated determinist the future could in principle implode into an eternal present. We are always within this present. There is really no future to the determinist, only a never ending unfolding of the spacial and temporal unity that we call the present.

The present and future of CCTV In relation to Graham fear of the allegedly coming of new systems for electronic face recognition we may thus ask: Does Graham describes the unfolding of a present, or does he speculate or make unjustified claims about an uncertain and open future?

There are good empirical reasons to argue that he is at least partly describing a present, at least if we stick to the UK. Several UK-firms are now selling fully operating and stable face-recognition systems.[1] Aurora Computer Services Ltd. claims that their 3D scanning of faces are extremely reliable: “We can’t say it’s 100% but we’ve done tests and have a zero failure rate.” These systems are installed many places in the world, with Heathrow as one of the likely next customers. The new UK-biometric passport (and the EU passport to come) holds electronic information about the face of the holder, and the image of the holder is stored electronically.[2]

Some of the possible technology that Graham describes may belong to an uncertain future, like the machine-intelligent recognition of “suspect” behaviours (as opposed to the recognition of already electronically known and wanted people). Some of the technology most certainly belongs to an unfolding present, even if it may not yet be in place. It is unjustified to dismiss this unfolding as “technological determinism”. It is also unjustified to “understand” this determinism emically (i.e. symmetrically with respect to its truth or falsity) as Wyatt suggest, because it might not be technological determinism at all.

Moreover, in one important way we do not have to settle if some of the technologies that Graham describes belongs to an uncertain future or an unfolding present. He is presenting a scenario, a possibility, a risk. (This is more clear to me after having read his paper than after having heard his paper presented.) And so we might ask if presenting this scenario is a Good and Critical thing to do, or if it, as Law worried about, is to collude with those he want to criticise. And if the telling of other stories, smaller, more diverse stories, is the more responsible option.

I do not want to choose between these two options. I want to have both, at least sometimes. And I’ll argue the case by reference to an example that I know better than surveillance by CCTV, namely the free software movement. Free or Open Source Software, in short, is software that is owned differently than proprietary software. All users of Free Software have the right and possibility (given their own skills) to change this software, and they have the juridical right to own their own changes – but only their own changes, not the whole program. Free Software has grown into a big thing. 70 % of all servers on the Internet run the Free Software server program known as Apache. As I mentioned above, hackers of all sorts, that is those who make Free Software – the nerdy “geeks”, both at universities, at home and in large corporations – strongly dislike the communication monopoly of the large software companies, notably Microsoft these days, but increasingly also Apple through their ipod.

Here is a brief example of such ownership; the text editor Word. We all send Word-documents to each other. It is a wonderful thing that “we all” (that is: most of academia) can open and read these documents. Word, then, is a communication standard. But the way a word-document is technically structured is a business secret of Microsoft, controlled by Microsoft. To most end users Word is naturalised. To hackers it is not. That is partly because hackers run Linux on their PCs and have been confronted with the hassle of opening word-documents, partly because the very communication standards of the Net is something that they care most passionately about.

Unix-hackers have for a long time disliked the MS Word-format, for its technical inaccessibility. The last 5 years or so, they have got company by an increasing groups of academic lawyers and politicians who have come to understand the democratic problems of letting single corporations control public communications standards. A political change is under way many places. Sun Microsystems has now made an open standard for word-like documents, called the Open Document Format (ODT). This format is now ISO-certified as a public standard, and activist, academics and politicians are working to make it into a practised reality.

The MS Word-document format is only one of many commercially owned communication formats that are politically and technically challenged by hacker-activists. Now, an important backdrop for much of the hacker-activism is a great aversion to Microsoft. Microsoft is “evil”, or at least dangerous, and there are, and have been for a long time, a lot of “large stories” out there who describes this danger. These stories are not very different in their structure form the story Graham told about CCTV. They tell about one “monstrous” beast. They don’t contextualise. They do not tell about how Microsoft users appropriate Windows in a great variety of ways. One could, following John Law, tell of the fluidity of Windows, just as de Laet and Mol have done very nicely about the Zimbabwe Bush Pump (de Laet and Mol 2000). In ways similar to this water pump, Windows have become a global technology not by being an “immutable mobile”, but by being a mutable mobile. Windows is “fluid”: It has been hacked, pirated and tweaked all around the world from Brazil to China. The users of Windows are not, generally, passive victims of an “evil empire”.

There are probably times and places were a possible story about the fluidity of Windows would be a Good Thing. But there are probably also places where it would not be a Good Thing. The critical stories of “Microsoft the Monster” have fuelled a productive hacker activism. What would a story about the fluidity of Windows produce? Legitimacy to Microsoft? Who, then, runs the risk of colluding with the Bad Guys? And, returning to Stephen Graham’s paper, in the case of computerised face recognition, who is colluding with the Bad Guys when Graham’s worry is dismissed as “technological determinism”?

I am not asking these rhetorical questions simply to take the side of good old fashioned criticism, against STS. I just think that sometimes we need the one, other times the other. Large, “de-contextualised” stories may describe large events, large drawings together into unities of large present times. Describing them is describing a “here and now”, not determining a future, even when they are stretched out in time.

Notes

1 For instance Aurora (http://www.facerec.com/company.html) and Dectel (http://www.dectel.co.uk/prodssfacialrec.asp).

2 See http://news.bbc.co.uk/1/hi/magazine/4035285.stm See http://www.urban75.com/Action/cctv.html See http://news.bbc.co.uk/1/hi/uk/4776562.stm

References

Graham, Stephen (2005), ‘Software-sorted geographies’, in Progress in Human Geography, 29,5: 1-19

de Laet, Marianne and Annemarie Mol, (2000), “The Zimbabwe Bush Pump: Mechanics of a Fluid Technology”, Social Studies of Science, 30/2 (April 2000) 225–63.

Latour, Bruno (2004), “Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern”, in Critical Inquiry, Volume 30 no. 2. Web: http://www.muhlenberg.edu/mgt/provost/frg/LatourCriticalInquiry.pdf

Mol, Annemarie (2002) The Body Multiple: ontology in medical practice, London and Durham: Duke University press

Whitehead, Alfred North (1953) [1925] Science and the Modern Word, Cambridge: Cambridge University Press Wyatt, Sally, in press “Technological determinism is dead; long live technological determinism”, New Handbook of Science and Technology Studies, edited by Ed Hackett, Olga Amsterdamska, Mike Lynch & Judy Wajcman (MIT Press)