From Infrastructure to Networking

I C. Ciborra (red.) From Control to Drift, Oxford University Press, 2000.

Internet is often described as an infrastructure for information society, but it is not obvious that the notion of infrastructure, once introduced to account for the foundation of industrial society, will be suitable for an analysis of the role of information technology. In this chapter we will question the idea that information technology is the stable foundation of information society. Instead, we will suggest that information technology should be understood as a flexible medium by which social structures are formed, reformed, and dissolved, in a continuous process of networking. 

1 A Paradox?

The standard introduction to management literature in the early 1990s told about how things were changing. In the past, they would say, it was possible for companies to rely on a stable market with faithful customers. But all that is different now, they went on to say, with global competition on a global, deregulated, and open market, empowered customers demanding tailor made products and services of high quality, shorter life span of products, a rapidly developing information technology, more knowledge intensive work and therefore a more educated work force demanding empowerment (Drucker 1988).

In the past, the most successful organisation was the huge factory, mass producing one and the same product to the same customers year after year, but now we were entering a situation in which the only surviving companies would be the ones that were able to quickly adjust to a an increasingly changing market with changing demands for new products and services. The recommended recipe for survival always involved moving away from the industrial factory organisation towards a more loosely organised, networked enterprise with alliances, external networking, and outsourcing as major ingredients (Peters & Waterman 1983).

At the same time, in the early 1990s, the US government launched an inspiring program for a national information infrastructure which drew the attention of media to information technology: ”Coming Soon to Your TV Screen: The Info Highway – Bringing a revolution in entertainment, news and communication” (TIME Magazine cover, April 12, 1993). Just like vice-president Al Gore’s father once contributed to taking the US out of the depression by a huge infrastructure project including automobile highways, so the son would now help increase economic growth with an infrastructure project for electronic highways. A well-tested industrial society (Keynesian) method for ending economic depression was to lay the foundation for post-industrial society.

The Clinton administration program for information highways, not only created an information technological media boom, it also helped initiate a research interest in infrastructures. So, while there was a general impression, expressed in popular management literature, that industrial society with is stable infrastructures was being replaced by a panta rei society of highly flexible and mobile organisations, politicians and researchers got engaged in building and examining the infrastructure for information society. Taken jointly this venture seems paradoxical. Why build infrastructures for a society that is described as a society without infrastructure?

One reaction to this paradox is to question the entrenched notion of infrastructure in favour of a more flexible and complex notion more suitable for analysing the role of information technology in information society. This line of argument can be found in the work of Susan Leigh Star (Star & Ruhleder 1996) and Ole Hanseth (Hanseth 1996, and chapter 3 above). Here, in our closing chapter, a slightly more radical remedy will be suggested. There is nothing wrong (or not very much wrong) with the received view of infrastructure as applied to a typical mid-20th century industrial society, we will argue. So, let us keep the notion of infrastructure more or less as it is, but instead realise that the new society we are now entering, is different enough from the industrial society we are leaving, to be best characterized as a society with no infrastructure.

Thus, in this chapter, the notion of infrastructure and its role in our understanding of society will be analyzed, comparing and contrasting the society we are leaving, dominated by machine technology, with the society we are entering, built by information technology. It will be argued that the notion of infrastructure, obviously useful as it is in theorising about industrial society, can be very misleading when used in theories about information society. In particular, we will concentrate on, what seems to be, the four central ingredients in the notion of infrastructure, together giving us an illuminating analysis of modern, industrial society: (a) the idea of infrastructure as a foundation underlying society, (b) the idea of infrastructure as a stable structure, (c) the idea of infrastructure as a common resource, and (d) the idea of infrastructure as a common standard.

It will be argued that these four ingredients, when used to understand the role of information technology in information society will lead us seriously astray. We will try to show that information technology, rather than being a productive foundation for information society, is a flexible means of communication, by which social structures are formed, reformed, and dissolved, in a continuous process of networking. Information technology is characterised by its lightness rather than by its weight and inertia. It is relatively inexpensive so that we can afford to compete rather than share, and it is an adapter technology that invites us to experiment with several standards at the same time. Rather than forming a stable infrastructure for information and service production, information technology introduces a more flexible and lightweight, networking, society without infrastructure. We will end by saying something about that society, about its networking, nomadic, enterprises, and about the nomadic philosophy and social science that will be adequate for its analysis.

2 A Foundation

Infrastructure was introduced during the industrial revolution in the 19th century as a military term to designate railroads, industries, and other resources, behind the actual front, necessary to wage war. Today, the term is normally used for the stable structure – roads, harbours, railroads, airports, energy systems and telephone networks – underlying production and distribution of goods and services in a society. It is common to distinguish three major types of systems within the infrastructure: the system of transportation, the energy system, and the system of communication.

If the term was new a hundred years ago, the concept had a long history in our culture. When Western philosophy began, it was with a quest for foundations. Natural philosophers from Thales to Empedokles asked for the arché of the world, for its ground, its foundations. Most of them ended up in one or another version of the four elements – earth, water, air, and fire – as the foundation of everything else.

Cultural historians have argued that this interest in foundations is typical of a young agricultural culture, trying to come up with an ideology legitimising a life style of staying on the farm rather than roaming the world. Farms are built on foundations, and the whole agricultural society is solidly planted in the ground. So, we get a philosophy concentrating on building stable foundations, sometimes worried by movement and change, but never really taking them seriously, except as aberrations or accidents against a background of stable foundations.

Archaeological excavations in the Mediterranean area typically show layers of buildings on top of each other. New civilisations are built on the old; progress is moving upwards from the ground. This whole idea of having something to build on becomes one of the fundamental philosophical ideas of our culture, from the axiom systems of Euclid and classification schemes of Aristotle to the methods of Descartes and Spinoza. The world, and our knowledge of it, is organised in levels, and the philosopher is moving, like an archaeologist at work, up and down these levels: analysing, reducing, deducing, and explaining. Classical philosophy is a philosophy for farmers.

Modern philosophy remained, from Descartes to Kant, a quest for foundations. An industrial society needs stable foundations, even more than an agricultural one. But there are important differences. With its huge and complex, distributed production system, an industrialised society needs a well functioning and reliable distribution system. It becomes natural for such a society to think of the distribution system – ports, railroads, highways, and airports – as a foundation for all of society. And thus the notion of infrastructure is formed.

The interest in foundations and levels really increases with the modernisation of philosophy. This is especially so in the philosophy that more than anyone else showed an explicit interest in industrialisation. We are thinking of course of Marx and Marxism.

In their theory of historical materialism, Marx and Engels distinguished between the basis and the superstructure of a society. In the basis were the material conditions: productive forces and relations of production. In the superstructure were political, legal, and ideological institutions as well as the ideas dominating them. This way of thinking of society, with a material basis of production, and different institutional levels building on that basis, was to become the de facto standard in modern sociological thought.

Marxism is modern in the way it reckons with historical change. The material foundations of societies evolve and sometimes this evolution results in a major social change, a revolution. The so-called industrial revolution is such a revolution in Marx’s sense. When the tools of traditional societies finally evolve into machines, and the productive forces thus make a leap forward, they will break through the agricultural ways of organising work and a revolution takes place changing the very foundations of society. And when the ways of production change, everything else will have to change as well: everyday life, politics, the judicial system, institutions of education, healthcare, art, ideologies, and so on.

Marx’s theory is a descendant of the Romantic Movement that around 1800 began to play around with the idea of foundations. Poets and artists began to think of the infrastructure as something underlying the visible social systems and activities – the mysterious sewage systems, water pipes, and so on, are the hidden blood and nervous system of society. The infrastructure, when it really is an infra-structure, is subterranean and out of sight, and preferably not only below but also beyond our consciousness. When you are reminded of the infrastructure it tends to make you nervous: there is something down there making everyday activities possible, but you don’t really know what it is, where it is, or how it works, and what will happen, come to think of it, if it breaks down. In the early 19th century people began to hesitate to look down into the entrails and subconscious of society for fear of vertigo and unease.

Historical materialism is a useful tool for understanding the changes from agricultural to industrial to information societies. But one has to be aware of the fact that it relies on an understanding of technology that is very much ”historical” in the sense of being based on the technology dominating industrial society, machine technology. Properties of other sorts of technology are not so well accounted for in historical materialism.

There are things we know about technology, which a theory of technology must account for: (I) Technology is artificial, that is, designed, constructed, made by people. Thus, the modern world is an artificial world. (ii) Technology plays important roles in our lives, determining our space of possibilities, giving us our identity. With glasses I can see, without computers I could not dream of a job as a programmer. (iii) Technology comes in many varieties, as tools, machines, media, and complex systems. New varieties will continue to appear.

Theories of technology typically use one or more root metaphors and expand on its theme. These root metaphors will get their inspiration from one or another preconceived understanding of technology. In the industrial age it is tools and machines that dominate our thinking. Tools, when we look back to the craftsmanship that we fear is disappearing as the machines take over. Machines, when we look into the future and applaud the automation of labour.

When we think of technology as a social construction (Bijker, Hughes, Pinch 1987, Dahlbom 1992) we have a romantic notion of technology as tools, made by us, for us to use at our leisure. Technology is an extension of our mind, a prosthetic device, with which we can conquer the world. When we think of technology as a machine, it becomes a form for our life (Winner 1986), an infrastructure for society (Hanseth 1996).

Marx called his theory ”historical” materialism, but of course he used the conceptions of his time to interpret history. That is, he used the notion of technology as a machine in his theory of technology. So, society is organised by forces and relations of production making up a material basis for the rest. But the very notions of force and production are industrial age notions. In agricultural societies, tools are used to cultivate the land. To think of tools as forces of production is to stretch the imagination. Similarly, in information society the focus of attention will be on transactions (on communication, sales, services) rather than on production. And, information technology is no productive force, it is a medium or an intermediary in a communication network.

We are still waiting for the Marx of information society. The first pretender to this epithet, Manuel Castells, is still stuck in traditional Marxism, defining the information economy in terms of information production (1996, p. 67). Castells relies on Marx’s notion of mode of production to define information society, and fails to see that information society is not really a production oriented or production based society at all. If we want to understand the role of technology in information society, we have to break out of industrial age thinking. We can continue to use Marx as an inspiration, but we have to be careful not to use his machine metaphors (like infrastructure) to force information society into an industrial mold. We can continue to be historical materialists, but we have to be careful how we understand the role of the material in a nomadic society, and refrain from thinking of it as a ”basis.” And we have to be careful in our understanding of materialism itself when we go from agricultural and industrial societies creating wealth by moving and transforming matter to information and media societies in which fortunes are made by exploring and exploiting virtual worlds.

3 A Stable Structure

An infrastructure can be thought of as a platform for implementation that, without interfering with the details, offers the foundation and service needed by the various applications for production, distribution, consumption and recreation. But an infrastructure can also be a regulating skeleton, providing framework and guidelines for the activity. The infrastructure provides stability and security rather than liberty of action. A rudimentary infrastructure will be a liberating platform, but it will leave a lot of work for you to do. A more developed infrastructure will restrict your choices while at the same time doing most of the work.

The infrastructure of modern industrial society is made up of systems. Thus, if you say ”roads” you will have to add automobiles, gas stations, auto dealers, garages, rules of traffic, and a lot more that is necessary for roads to be functional. It is important to see that these systems are not just technical systems, but rather sociotechnical. If one takes the notion of infrastructure seriously enough, one must therefore add a number of less material systems, social institutions, such as educational systems, monetary systems, and languages necessary to make the technical systems operational. In addition to the systems of transportation, energy, and communication, one had better include in the infrastructure those basic industries on which the majority of enterprises in a society depend, such as banking, steel mills, chemical industry, and food industry. Once, we begin to slide down this slope, we will see more and more examples of things to include in the infrastructure. Isn’t the fact that there is a market for its products as vital to a company as the necessary means of production?

All this makes infrastructure a somewhat slippery notion. Perhaps we should only use it in a relative sense: X is an infrastructure relative to Y, meaning that Y depends for its operation on X, and X is somehow more stable and basic than Y. And yet, the notion of infrastructure works well in industrial society with its predilection for levels: base ­ super structure, basic research ­ applied research, basic knowledge, basic technologies, and so on. In comparison, information society seems at first to lack an infrastructure of its own (much of the old infrastructure will still remain, of course), or, perhaps rather, to lack a material infrastructure. The rapid changes characterising that society and the immaterial nature of information, as well as the very idea of networks, seem incompatible with ontological levels, stable foundations, and infrastructures.

A complex industrial society is impossible without a well functioning infrastructure of energy, transport, and communication systems. Like the productive forces in Marx’s theory, the infrastructure is a stable, inert material foundation for production, social life and organisation. These are illuminating theories of the way a modern, industrial, society works, but when the distance from natural resource to final product increases, when focus shifts from products to services, when all labour is well educated, and when technology becomes more knowledge intensive, then that society is turned on its head. The nearest we come to a stable foundation of information society – is its ideas, its educational institutions, its research organisations, its political and legal system, the software, the habits and values of its citizens.

This change is similar to the change experienced when moving from an agricultural to an industrial society. In agricultural society it is food production that is the basis of society. Farmers used to warn their sons not to leave the farm to go to the cities, ”because in the cities there is nothing to eat.” To a farmer, the cities depend on the farms for their survival, but with industrialisation of agriculture this dependence relation changes. Modern agricultural production units depend on the factories in the cities – on roads, machines, fertilisers, and electricity – for their survival. Similarly, factory thinkers continue to believe that the infrastructure of industrial society will remain the infrastructure of information society.

4 A Common Resource

The notion of infrastructure is rich in content. Infrastructures are stable foundations for organisations and activities. We have argued above that we should be careful in extending these ideas of infrastructure to information technology. But infrastructures are also shared, common resources, and that idea may, on the other hand, be quite valuable to retain in our understanding of information technology.

Modern organisation theory treats the division and co-ordination of labour as the two fundamental aspects of organisations (Mintzberg 1983). Management information systems, office automation, groupware and workflow management is all about controlling these internal aspects of the organisation. Infrastructure, in comparison, can be viewed as a public commodity external to the organisation. In that sense, it is a shared resource, a stable, long term investment, preferably paid for by taxes rather than by users, so that it will be used freely without constraints by companies and individuals to increase the wealth of nations.

The infrastructure is what you need to build an organisation. In the infrastructure is included technology, personnel with their basic education and competence, buildings, systems of transport, finance, laws, a market, and so on. Organisations are, of course, vitally dependent on such an infrastructure. As they grow in power, they will try to control it, to ensure its well functioning. They will incorporate aspects of the infrastructure within the organisation, or build their own alternative infrastructure, rather than having to depend on public resources partly beyond their control.

When Internet is being introduced to the general public in the Clinton-Gore initiative, it is described as an infrastructure, a shared resource, extending beyond and between companies. But when companies reluctantly begin to use Internet, they soon find ways to turn it into an internal, proprietary infrastructure. When you decide to use Internet technology internally, enclosed within firewalls, in an organisation, it is used as an infrastructure for the organisation. If the company has several profit centres with independent economic responsibilities, and it uses an Intranet as a shared resource, then this infrastructure can be said to be internally public.

Today, when companies are trying to keep up with the revolution brought on by Internet, when they are ”globalising” in response to the global market made possible by information technology, there is much talk about ”global infrastructures.” So far, this has mostly been a misnomer, however, since what companies do to become global is not to begin to use the Internet as infrastructure, but to introduce global information systems. Just like traditional management information systems were used to support control and standardised conduct in the organisations of the 1970s, so today’s ”global infrastructures” are really global information systems for standardisation and co-ordination of labour.

Large companies today hesitate about how to deal with Internet as an infrastructure. Should they invest in electronic commerce with Internet as the market place, or should they not? In many ways this can be a much more important decision than might first seem. By choosing to use information technology as a public infrastructure, in this case Internet, the company will open its boundaries to the surrounding world. Instead of using information technology to build information systems to sharpen company boundaries, a process will begin that eventually can turn the old factory companies of industrial society into more distributed, customer oriented, sales organisations. If the old industrial companies don’t do this, there will be other companies who will do so at their expense.

A production-oriented company will focus on its production and distribution infrastructure, while a sales-oriented company will focus on external communication. So far, the public globalisation process has been moving much quicker than companies normally have expected, and so they have wasted money on internal infrastructure rather than using what has quickly become publicly available. Why build your own roads, when you can use the public roads? Why develop EDI when there is Internet? A high and reliable quality in production and distribution may demand an internal infrastructure that is yours to own and control. But a powerful sales organisation needs to meet the customers where they are, and therefore has to operate on the public infrastructure.

Industrial companies rely on a well functioning infrastructure for supply and delivery. So dependent are they on such an infrastructure that they try hard to control it, preferably turning it into a company owned system. Sales oriented companies in information society rely on a well functioning electronic market for selling their services. It does not make sense to turn this into a proprietary system, since the very idea of a market is for it to be a shared resource in the sense of being open and accessible to everyone (with enough money).

The conclusion, then, is that the common resource aspect of the notion of infrastructure remains important as we move into an information society. Information technology is inexpensive enough for big companies to invest in their own infrastructures. They don’t need to depend on information technology as a public resource paid by tax money. And yet, it makes good sense for them to orient their operations towards the publicly available communication network. Information technology will provide the market place for information society by providing the communication medium for a society, increasingly engaged in commerce of all sorts on a global level.

5 A Common Standard

To think of an infrastructure as a stable foundation, a common resource, to operate upon, is to think of it as a material structure. But one of the more important aspects of an infrastructure is immaterial. To build a national, or international railway system, one has to agree on how wide the rails should be. And the trains need time tables, which presuppose a standard time. An industrial society needs numerous standards for interfaces connecting the modules making up machines. Nuts, bolts, screws, nails, boards – they all need standards and all these standards rely on standard measures. A national market needs a common currency.

All the vehicles that will run on the common infrastructure will have to be adapted to the properties of that structure. So, all the owners, producers, users of vehicles will have to agree on how to design the infrastructure, what standard to choose (cf. Hanseth 1996).

Information technology too has its standards, of course. Operating systems, protocols, cables, interfaces of all sorts, characterise this technology. To us they may seem stable and impossibly inert in a world of otherwise flexible technology. And yet we are speaking about standards that rarely survive for more than two decades. More important than this is of course the nature of information technology which makes the relation between vehicles and infrastructure importantly different.

Industrial society solves its problems of incompatibility between flexible vehicle technology and inert infrastructure by means of gateways. The very same gateways can also be used, of course, as means to combine infrastructures with different standards. When the trains cross the border their wheel width will have to change. But to machine technology, gateways is a problem. They are expensive and cumbersome. They are awkward and inelegant, and something to be avoided as much as possible. Information technology is very different.

Information technology is a gateway, or adapter, technology. Compilers and interpreters are the very essence of information technology. So, the reason for standards is different. Partly, there are hardware reasons, but with increasing softening of the technology, the reasons are more pragmatic than necessary. Why have different operating systems, programming languages, or communication protocols, when they are equal in capacity, and when gateways have to be installed and, of course, draw a bit on time and space resources?

Only gradually are we beginning to understand this new won freedom from standardisation. With information technology we get gateways and don’t have to standardise any longer. Why introduce a global currency when all it takes is software gateways in all systems of exchange? When there is no need for standards, then standards will no longer slow down the development of new technology. And, the tempo and flexibility of society will increase.

6 A Nomadic Society

Information technology develops and changes quickly compared to machine technology. Investments in industrial society infrastructures – ports, railroads, highways, airports, traditional telecommunications – are large and typically paid for by taxpayers. Investments in information society infrastructures – fibre nets, routers, servers, and software – are comparatively small scale, and commercially interesting with many competing alternatives. Even if they continue to be a common resource, it is not because of their monopoly or that they are so expensive, but because the very idea of a public communications network as a global market is commercially appealing. Indeed, even from a security point of view, it makes more sense that we do our transactions on the same communications network than that we drive on the same roads.

In major cities around the world you can buy Gant shirts, jackets, pants, sweaters. Perhaps you have seen one of their catalogues, made in a truly American style, with colourful pictures in New England settings on nice, thick paper. When the Gant shirts were becoming popular in Sweden in the early 1980s, Swedes used to look for Gant stores when visiting the US. They never found any. No wonder, since Gant was not a US shirt maker. Gant was what Hedberg (1991) calls an ”imaginary organisation.” In the 1980s they were a team made up of a few clothes designers, a couple of business people, a copywriter, and some assistants. Altogether they were about a dozen people, operating out of Stockholm.

When they decided to go into the fashion business on their own, they chose not to invest in textile factories, but decided to stick to doing what they knew: designing clothes. So, once they have designed a new collection, they have some textile factory sew up a few exemplars. They bring them to the US and hire photographers and whatever it takes to make the catalogue, and then they return to Sweden. In the beginning, Gant shirts were sold in various high quality clothes stores, but now there are specific Gant stores only selling Gant. These stores are not owned by the company, of course. They operate on independent contracts. Now, there are even Gant stores in the US.

Companies like Gant are like the hordes of pre-agricultural, nomadic, societies, as described by Emile Durkheim. According to Durkheim, people then lived in small groups of about twenty-five members (the size of a modern school class), roaming around in the fields and forests, gathering roots and nuts or hunting for small animals. Such a group is a horde. They live off the land, but they never stay very long in one place, nor do they sow to harvest.

If by infrastructure, we mean the platform or framework, the foundations, making nomadic life possible, we would perhaps find the infrastructure of the hordes in their kinship system, their marital habits and customs. But that would be silly. We would transport a notion that is really useful in industrial societies (and to a lesser extent in agricultural societies) and use it on a society to which it does not apply. The very idea of nomadic societies is not to rely on infrastructures. If there is anything like an infrastructure in nomadic societies it is nature itself.

Companies like Gant rely on industrial society like the hordes rely on nature as a common resource. The society formed by such ”imaginary” or networking organisations does not have an infrastructure in anything like the sense characteristic of industrial society. What they rely on for their operations are, first and foremost, good communications and the possibility to change communicative partners without delay. Such companies don’t want information highways in any way like the highways or railroads of industrial society.

What we see around us today is the dawn of a nomadic society that, at least to begin with, draws on industrial and agricultural society for its operations, in a way reminding of how the nomadic hordes lived off nature. In a world dominated by such networking, nomadic organisations, there will still be room for industrial type organisations providing a sort of infrastructure for the networking organisations. There will still be a need for steel production, paper mills, railroads, ports, airports, airlines, and the whole system of industrial production and distribution will be necessary for the networking organisations to operate.

But the communication networks that are so vital for nomadic, networking life will not be part of that infrastructure. The optic fibres, satellites, routers and hubs necessary for communication will be much too flexible, changing, and makeshift to be anything like an infrastructure for information society. They will be like the Internet is today.

Information society hordes are small, mobile companies with good design ideas, like Gant, Benetton, Nike, or companies that simply are good at selling, like TV Shop, good at doing business on the new electronic markets and can place orders with the old industrial companies, use their resources, like the nomads used nature, and move on when opportunities are better elsewhere.

The first hordes are already here. And in some big companies, innovation projects are sometimes set up as hordes. When electronic commerce at last becomes a major factor on the business scene, we can expect to witness an interesting game with the old industrial companies and the new hordes as players. One day there may be an automotive company that is a horde! A small group of automobile designers developing a car the production of which they place on contract with different traditional companies (like some automotive companies already do). The horde has the ideas; others can do the low paid, high quality, routine work for them. Information society will be an Idea society.

7 A Nomadic Philosophy

Most of 20th century philosophy remains stuck in infrastructure thinking. This is certainly true of the phenomenological movement, as defined by Edmund Husserl and Martin Heidegger. And it is true of Cambridge analytic philosophy with Bertrand Russell and of the Vienna circle, under the influence of Ludwig Wittgenstein in the Tractatus. But there are powerful exceptions, as well, pointing towards a break with traditional and modern quests for foundations. The most influential alternatives to infrastructure philosophy are to be found in American pragmatism, as developed by Peirce, Dewey and James around the turn of the century 1900, in the Vienna circle’s development after Tractatus, with Rudolf Carnap as figurehead, in Wittgenstein’s later philosophy, and in French post-structuralism.

In his early essays, From A Logical Point of View (1953), and in Word & Object (1960), W. V. O. Quine developed an influential version of a philosophy without foundations, drawing mainly on Carnap and pragmatism. Quine came to play the central role in modern analytic philosophy in the US after the 2nd world war, very much setting the agenda for philosophy there. Quine used a quote from Otto Neurath, member of the Vienna circle, as motto for his philosophy:

Wie Schiffer sind wir, die ihr Schiff auf offener See umbauen müssen, ohne es jemals in einem Dock zerlegen und aus besten Bestandteilen neu errichten zu können.

For Quine there is no such thing as a first philosophy, a foundation, or method on which to base one’s quest for truth. Philosophy is not prior to science, but rather it is the other way around that philosophy emerges in the midst of the scientific project, drawing on scientific results to explain what knowledge is, what is knowable and what is not. Philosophy is not the foundation of the sciences, but instead epistemology is simply a ”chapter of psychology.”

Ancient and modern philosophy, as defined by Plato and Descartes, respectively, can be described as examples of ”vertical thinking,” trying to organise knowledge in levels or layers. The philosophical method is typically one of scepticism, of analysing knowledge claims by questioning assumptions, weeding out what is uncertain, doubtful or contingent in order to reach an unquestionable core. Quine’s philosophical position and method is different and truly provisional. Philosophy grows dialectically with science, supporting itself once here, once there, advancing and retreating, building and rebuilding its vessel of knowledge, increasing its understanding by horizontal expansion rather than by vertical restriction.

While ancient and modern philosophy drives poles down into the mud to reach the hard rock underneath, Quine invites us to think horizontally and use the flimsy network of a continuously revised science to stay afloat on the mud, saying, in effect, that there is no rock beneath, that it is mud all the way down.

This is a philosophy for a society without infrastructure. It is a philosophy without foundations, a ”tough minded” philosophy, to speak with William James. It is a philosophy for a society in which life has become horizontal rather than vertical, in which people engage in networking rather than edification, establish multiple connections rather than hunt for Arkimedean points, and encourage diversification rather than core activity. In short, it is a philosophy for a nomadic society.

Quine is the most powerful nomadic philosopher on the American scene, but there is a whole group of sympathetic philosophers in his vicinity: Wilfrid Sellars, Nelson Goodman, Hilary Putnam, Richard Rorty, Daniel Dennett. In their different ways they all make their contributions to a philosophy for information society – beginning their work when that society is barely discernible at the horizon.

There are extensive similarities between Quine’s philosophy and that of the late Wittgenstein, even if there also are major differences. Instead of science, Wittgenstein relies on everyday experience as his background for philosophical therapy. But the message is very much the same: mistrust in foundations, and in the traditional (and modern) ambitions of philosophy to find an infrastructure on which to place everything else. In practice, Wittgenstein is more of a nomadic philosopher than Quine, roaming from subject matter to subject matter in his attempts to make his point, while Quine has a more restricted view, tending to repeat the same argument, the same illustrations, again and again.

There are relevant similarities, as well, between Quine’s philosophy and that of French post-structuralism, especially as expounded by Jacques Derrida in the late 1960s and early 70s. Derrida’s criticism of Husserl’s foundationalism and his break with the closed systems of Lévi-Strauss structuralism, in favour of an open ended, continuously changing networking, have close parallels in Quine’s philosophy (Carlshamre 1986). We are not saying that these three philosophers, Quine, Wittgenstein, Derrida, coming out of very different traditions, and with very different agendas, have so much in common that it warrants close comparisons. What we are saying is only that against a background of traditional and modern infrastructure philosophy, they all share a more horizontal way of thinking, giving, each in his own particular way, a powerful exposition of an alternative, nomadic philosophy.

In the last decades, there have been backlashes against the very radical philosophies of Wittgenstein, Quine, and Derrida, backlashes with a more traditional (or modern) approach to philosophy. But it is a safe bet that, when in the next century, the origins of information society philosophy are traced, these are the three major contributors that will be singled out.

The social sciences, too, will have to change, of course, as we leave modern industrialised society behind, entering post-industrial, networking, nomadic, information society. The process of modernisation, the industrial revolution taking us from a traditional, rural, agricultural society into a modern, urban, industrial world, can no longer continue to set the agenda for the social sciences. Marxism, structuralism, Weberian rational organisation theory, Parson’s universals, systems thinking, Gesellschaft­Gemeinschaft, can no longer provide the conceptual framework for social theorising. They all obviously suffer from industrial age, infrastructural thinking.

We have to realise that the social sciences were established, in the midst of the industrial revolution, to explain that revolution, as theories of modernisation and industrialisation. As we leave industrial society behind, there will be need for a more fundamental rethinking of the conceptual and theoretical frameworks of the social sciences. In my own discipline, informatics, we have to get rid of systems thinking, and we have to develop new ways of thinking about technology. And, we should be very careful when importing such politically loaded, industrial age, concepts as that of infrastructure to analyse the social role of technology.

With this warning in mind, here are some examples of metaphors for technology, which go beyond the tool-machine discussion typical of industrial age social science: Technology as medium posits technology as a place for us to meet and interact in. It is an attractive metaphor, but we have to realise that we are yet a long way away from that situation. Today, the way we use information technology rather involves a lot of interaction with the technology itself. Here are some metaphors that try to capture this: Technology as stranger (Ciborra 1996); Technology as intermediary (Latour 1992); Technology as organism (Dahlbom & Mathiassen 1993).

These theories break away from the two layered theory of society, at least when it comes to understanding technology, and Latour is quite explicit in this respect. These theories all stress the interactive aspect, technology as something people interact with.

It has often been observed that our Western culture has a penchant for nouns. Farmers and factory owners are people of substance, and processes and change are always secondary to substance. Infrastructure is a very powerful expression of substance thinking. A networking, nomadic society may perhaps be better described with verbs than with nouns. It is a networking society, not a network society. It is activities and actions rather than organisations and agents that make up that society. (Even when we want to use verbs, we find only nouns.) So, we should follow Weick (1976) and study organising rather than organisations, and Czarniawska (1997) and speak of ”action-nets” rather than ”actor-networks.”

Barbara Czarniawska has been a long time proponent of an ethnographic approach to the study of organising (Czarniawska 1997, 1998). Ethnography is, of course, the approach to use when meeting with foreign cultures and social practices. And, ethnography therefore is the most suitable approach to our own organisations and social practices, when these undergo revolutionary change. We who grew up in industrial society were of course experts on that society, being shaped by it and being responsible for shaping it. But growing up in industrial society we are obviously uncomprehending when confronted with the society coming out of the information technology revolution. Our conceptual frameworks only hinder. We have become foreigners in our own society, and we have to approach our own society as foreigners.

Eventually there will come a new Marx, Durkheim, Weber, Pareto, and Tönnies, and they will provide the conceptual frameworks needed to understand the information technology revolution and the new society growing out of it. They will of course build on the nomadic philosophy already outlined and on all the ethnographic research going on today as the revolution is unfolding. The general character of their analysis will be to show how information technology takes us out of a vertical society with levels and layers, infrastructures and super structures, into a horizontal society of networking nomads, a society without foundations.

8 Conclusion

The modern organisation is a creation of the industrial revolution. Systems thinking is the adequate foundation for theorising about that organisation. If information technology is really changing the focus of business from production to sales, by automating industrial production on the one hand and creating a global market on the other, then those organisations will have to change too. The opaque factories at the outskirts of society, so typical of industrial society, will change into mobile, distributed sales companies with no clear boundaries to their environment. Rather than using information technology to control their internal organisation, they will prefer to use information technology as a public communication medium to participate in the world.

An infrastructure is a stable foundation, a common resource, a standard for business activities. But if you think of IT in this way, you will soon find yourself in trouble. Here we have argued that IT is characterised by its lightness rather than by its weight and inertia. IT is no productive foundation; it is a flexible means of communication. It is relatively inexpensive so that we can afford to compete rather than share. It is an adapter technology that invites us to experiment with several standards at the same time.

The infrastructure of information society is not Internet, or information technology more generally. If there is an infrastructure of information society at all, it is the old industrial society with its industrial production factories and distribution systems. But this is a misleading way of thinking. The lightness and flexibility of information society, made possible by its light, flexible and inexpensive technology, will spread to the old industrial structures too. They will not serve as an inert foundation for information society. Instead, they will be transformed by the new social organisations in order to meet new demands for flexibility and change.

Information technology is no infrastructure issue to decide upon and then lay aside in order to attend to business. Information technology is what we make business with. Electronic commerce is not a question of finding a standard for payments on Internet. It is instead, as the name indicates, the nature of commerce, the market, of electronic society. And it is on this market we must compete, by experimenting with new technology, new ways of organising, moving quickly, like nomads.

References

Bijker, W.E., Hughes, T.P. & Pinch, T. (1987) The Social Construction of Technological Systems. Cambridge, MA: The MIT Press.

Carlshamre, S. (1986) Language and Time. Acta Universitatis Gothoburgensis.

Castells, M. (1996) The Rise of the Network Society. Oxford: Blackwell.

Ciborra, C. (1996) Introduction: What Does Groupware Mean for the Organization Hosting it? In Ciborra, C. (ed.) Groupware & Teamwork. Invisible Aid or Technical Hindrance? John Wiley & Sons.

Czarniawska, B. (1997) Narrating the Organization: Dramas of Institutional Identity. Chicago: Chicago University Press.

Czarniawska, B. (1998) A Narrative Approach to Organization Studies. SAGE Publications.

Dahlbom, B. (1992) The Idea that Reality is Socially Constructed. In C. Floyd, H. Züllighoven, R. Budde & R. Keil-Slawik (eds) Software Development and Reality Construction. Berlin: Springer-Verlag.

Dahlbom, B. & Mathiassen, L. (1993) Computers in Context. Oxford: Blackwell.

Drucker, P. F. (1988) The Coming of the New Organization. Harvard Business Review, 88: 45-53.

Hanseth, O. (1996) Information Technology as Infrastructure. Gothenburg Studies in Informatics, Report 10, Department of Informatics, Göteborg University.

Hedberg, B. (1991) The Role of Information Systems in Imaginary Organizations. In R. K. Stamper, P. Kerola, R. Lee & K. Lyytinen (eds) Collaborative Work, Social Communications and Information Systems. Elsevier Science Publishers (North Holland).

Latour, B. (1992) Technology is Society Made Durable. In J. Law (ed.) A Sociology of Monsters: Essays on Power, Technology and Domination. London: Routledge.

Mintzberg, H. (1983) Structure in Fives: Designing Effective Organizations. Englewood Cliffs, N. J.: Prentice-Hall.

Palvia, S., Palvia, P. & Zigli, R. M. (1992) The Global Issues of Information Technology Management. Harrisburg, PA: Idea Group Publishing.

Peters, T. J. & Waterman Jr., R. H. (1983) In Search of Excellence. Lessons from America’s Best-Run Companies. Harper and Row.

Quine, W. V. O. (1951) From A Logical Point of View. Cambridge: Harvard University Press.

Quine, W. V. O. (1960) Word & Object. Cambridge, MA: The MIT Press.

Star, S. L. & Ruhleder, K. (1996) Steps Toward an Ecology of Infrastructure: Design and Access for Large Information Spaces. Information Systems Research, vol. 7, no 1, pp. 111-134.

Weick, K. E. (1979) The Social Psychology of Organizing. Reading, MA: Addison-Wesley.

Weill, P., Broadbent, M. & St. Clair, D. (1996) IT Value and the Role of IT Infrastructure. In J. Luftman (ed.) Strategic Alignment. Oxford University Press.

Winner, L. (1986) The Whale and the Reactor. Chicago University Press.