The Technium

What Technology Wants

[Translations: Japanese]

Your dog  wants to go outside. Your cat wants to be scratched.  Birds want mates. Worms want moisture. Bacteria want food.

The wants of a microscopic single-celled organism are less than the wants of you or me, but all organisms share a few fundamental desires: to survive, to grow. The wants of a protozoan are unconscious, unarticulated, and more like an urge, or even tendency. A bacterium tends to drift toward nutrients with no awareness of its needs. There is no room beneath its membrane for a will as we know it, yet in a dim way it chooses to satisfy its wants by heading one way and not another.

Perhaps not much room is needed to want. The astrophysicist Freeman Dyson claims that we should view the smallest known bits of organized matter — quantum particles – as making choices. For millions of years a particle will exist and then suddenly it decays. Why then? Dyson says that from the individual particle’s viewpoint, this moment can only look like a choice, a satisfaction of a want. It is only on the scale of statistics with millions of particles that a particle’s choice shapes up as a predictable radiation half-life. But even individual human wants and desires average out to weirdly predictable laws in aggregate.

If a little one-celled protozoan – a very small package – can have a choice, if a flea has urges, if a starfish has a bias towards certain things, if a mouse can want, then so can the growing, complexifying technological assemblage we have surrounded ourselves with. Its complexity is approaching the complexity of a microscopic organism. This tissue consists (so far) of billions of dwellings, millions of factories, billions of hectares of land modified by plant and animal breeding, trillion of motors, thousands of dammed rivers and artificial lakes, hundred of millions of automobiles coursing along like cells, a quadrillion computer chips, millions of miles of wire, and it consumes 16 terawatts of power.

None of these parts operate independently. No mechanical system can function by itself. Each bit of technology requires the viability and growth of all the rest of technology to keep going. There is no communication without the nerves of electricity. There is no electricity without the veins of coal mining, uranium mining, or damming of rivers, or even the mining of precious metals to make solar panels. There is no metabolism of factories without the ingest of food from domesticated plants and animals, and no circulation of goods without vehicles. This global-scaled network of systems, subsystems, machines, pipes, roads, wires, conveyor belts, automobiles, servers and routers, institutions, laws, calculators, sensors, works of art, archives, activators, collective memory, and power generators – this whole grand system of interrelated and interdependent pieces forms a very primitive organism-like system. Call it the technium.

The technium is the sphere of visible technology and intangible organizations that form what we think of as modern culture. It is the current accumulation of all that humans have created. For the last 1,000 years, this techosphere has grown about 1.5% per year. It marks the difference between our lives now, verses 10,000 years ago. Our society is as dependent on this technological system as nature itself. Yet, like all systems it has its own agenda. Like all organisms the technium also wants.

To head off any confusion, the technium is not conscious (at this point). Its wants are not deliberations, but rather tendencies. Leanings. Urges, Trajectories. By the nature of self-reinforcing feedback loops, any large system will tend to lean in certain directions more than others. The sum total of millions of amplifying relationships, circuits, and networks of influence is to push the total in one direction more than another. Every owner of a large complicated machine can appreciate this tendency. Your machine will “want” to stall in certain conditions, or want to “runaway” in others. Left to its own devices, complex systems will gravitate to specific states. In mathematical terms this is called the convergence upon “strange attractors” – sort of gravity wells that pull in a complex system toward this state no matter where it starts.

Of course we humans want certain things from the technium, but at the same time there is an inherent bias in the technium outside of our wants. Beyond our desires, there is a tendency within the technium that – all other things being equal — favors a certain solutions. Technology will head in certain directions because physics, mathematics, and realities of innovation constrain possibilities. Imagine other worlds of alien civilizations. Once they discover electricity, their electronics will share some, but not all, attributes with our electrical devices. That which they share can be counted as the inherent agenda of electrical technology. Throughout  the galaxy any civilization  that invents nuclear power will hit  upon a small set of workable solutions: that set is the inherent “agenda” of technology.

It would be wonderful if we could survey all alien technological civilizations to extract the common tendencies in technological growth. A large number of technological evolutions would reveal the culture-free dynamics beneath them all. Since we have a solitary sample of one technium back on Earth, we have fewer methods of unraveling inherent system bias in technology. Three lines of evidence present themselves:

1) We can look back in history to when technological development was more culturally isolated. The pathways of technology in early China, South America, Africa, and Western Europe out with only minimal cross-over  influence. Examination of their parallel developmental sequences can reveal inherent biases.

2) More importantly, the major predecessor system to technology is organic life. Many of the dynamics of evolution and syntropy extend from living organisms into artificial systems, primarily because they share similar disequilibrial states. We can see the direction of technology in the direction of life and evolution (that is if you accept evolution has a bias as I do).

3) The long-term history of our single technium shows high-level patterns which we can project forward. We can ignore individual inventions and chart long-term flows which enable them. Much as we might want the compressed history of a growing creature and guess where it goes next. If the organism is a caterpillar we are out of luck; if it is a worm, it will succeed.

Animals At The Beach 5

Strandbeest by Theo Jansen

So, looking at the evolution of life and the long-term histories of past technologies, what are the long-term trajectories of the technium? What does technology want?


To increase diversity

To maximize freedom/choices

To expand the space of the possible


To increase specialization/uniqueness

To increase power density

To increase density of meaning

To engage all matter and energy

To reach ubiquity and free-ness

To become beautiful


To increase complexity

To increase social co-dependency

To increase self-referential nature

To align with nature


To accelerate evolvability

To play the infinite game

In general the long-term bias of technology is to increase the diversity of artifacts, methods, techniques. More ways, more choices. Over time technological advances invent more energy efficient methods, and gravitate to technologies which compress the most information and knowledge into a given space or weight. Also over time, more of more of matter on the planet will be touched by technological processes. Also, technologies tend toward ubiquity and cheapness. They also tend towards new levels of complexity (though many will get simpler, too). Over time technologies require more surrounding technologies in order to be discovered and  to operate; some technologies become eusocial – a distributed existence – in which they are inert when solitary. In the long run, technology increases the speed at which it evolves and encourages its own means of invention to change. It aims to keep the game of change going.   

What this means is that when the future trajectory of a particular field of technology is in doubt, “all things being equal” you can guess several things about where it is headed:

• The varieties of whatever will increase. Those varieties that give humans more free choices will prevail.

•  Technologies will start out general in their first version, and specialize over time. Going niche will always be going with the flow. There is almost no end to how specialized (and tiny) some niches can get.

•  You can safely anticipate higher energy efficiency, more compact meaning and everything getting smarter.

•  All are headed to ubiquity and free. What flips when everyone has one? What happens when it is free?

•  Any highly evolved form becomes beautiful, which can be its own attraction.

•  Over time the fastest moving technology will become more social, more co-dependent, more ecological, more deeply entwined with other technologies. Many technologies require scaffolding tech to be born first.

•  The trend is toward enabling technologies which become tools for inventing new technologies easiest, faster, cheaper.

•  High tech needs clean water, clean air, reliable energy just as much as humans want the same.                           

These are just some of the things technology wants. We don’t always have to do what technology wants, but I think we need to begin with what it wants so that we can work with these forces instead of against them.

  • paradiso

    You say: “We can see the direction of technology in the direction of life and evolution (that is if you accept evolution has a bias as I do).”

    But I don’t accept that at all. While I do see technology as a subset of biology, I’m uncomfortable with the term agenda and its implied volition.

    As a farmer I have certainly seen how things seem to “want to” die or lean against other things or come apart under stress.
    I just don’t find any of that having intent or wishes.

    And nor do I accept evolution having any bias. It just runs, things happen and accumulate, or fade away. It doesn’t disturb me to think of the universe as entirely random; but then I don’t think of anything beyond human society as anthropocentric.

  • AjmoT

    Kevin, I just don’t understand you man. Why do you resist accepting that technology is a fully living and conscious organism? You say: _”To head off any confusion, the technium is not conscious (at this point).”_ But earlier in the post you clearly accept that from a macroscopic view, humans can be seen as dumb cogs: _”It is only on the scale of statistics with millions of particles that a particle’s choice shapes up as a predictable radiation half-life. But even individual human wants and desires average out to weirdly predictable laws in aggregate.”_

    “Weirdly predictable!” You’re funny. So you say: _”Its wants are not deliberations, but rather tendencies. Leanings. Urges, Trajectories.”_ Okay, and this is exactly how we can see humans, and I don’t know what kind of super-yogis you’ve been hanging out with, but most people seem very set on beings cogs, routinely, within culturally set patterns, allowing for limited creative deviation. Look at a human population: 90% predictable, 10% not yet really understood: good enough for university science, huh!

    _”By the nature of self-reinforcing feedback loops, any large system will tend to lean in certain directions more than others.”_ Pardon me, but how else could you describe the human organism if you were determined to see it mechanistically? Well, no doubt if you’re bent on seeing a forest or an ant colony with the same lack of appreciation, you’ll be blind to a whole hidden and wonderful reality of conscious communication taking place in those systems. Same goes for a building and a business — any organism, any population.

    _”The sum total of millions of amplifying relationships, circuits, and networks of influence is to push the total in one direction more than another. Every owner of a large complicated machine can appreciate this tendency.”_ I still don’t see how you could be describing anything different from a bunch of humans. So what’s this special privilege of intelligence and consciousness that you are selecting for us folk, and assuming to be lacking in other parts of reality?

    This is a biased, a-priori assumption on your part with precious little evidence in its favor. To assume that consciousness is rare and, _as far as we know_, located only in ourselves, is an anthropocentric dogma and sorely anti-scientific. “I’m the only conscious thing around, these are the very few living things around me, all else is mechanical and inert!”… It’s not a very provable hypothesis and it prevents the foundation work necessary to start communicating effectively with the world at large. In the sciences, you don’t expect to see an invisible force head on. Don’t assume what you can’t see to be non-existent. One listens for the tell-tale signs of a phenomena, always with an eye for paradoxes and an ear to the nagging noise in the signal.

    Here is some noise which carries a signal: the apparently mechanistic behavior of non-human systems is a symptom of our perspective, seeing these systems macroscopically and almost exclusively in aggregate, as an “other”. As you begin to point out, humans in aggregate look like communities of bacteria, just as predictable as any other collection of individuals running amok in the cosmos, Ol’ Compy 386 included.

    “The wants of a microscopic single-celled organism are less than the wants of you or me… The wants of a protozoan are unconscious, unarticulated, and more like an urge, or even tendency.” Really, please try tracing that thought backwards a bit. Are not human patterns of behavior classifiable as “tendencies” and “urges” viewed from a distance, in the statistical simulation? Right, we’re like the protozoan in that way. And I’d say the “urge” is perfectly “articulated” by the protozoan — it really is making manifest its intentions in the surrounding world — but from the point of view of your pet dog, most of what you say to it probably counts as “unarticulated” too, huh? So there again we’ve merely got an erroneous epistomelogical posture based on a psychological xenophobia: this “other” is not like me, I refuse to treat it as an equal! So now we come to the label of “unconscious”. If the tree that cannot hear you speak, should it assume you are “unconscious” because it can’t communicate with you? And vice versa? Aw heck, some of nature at large may be just as prone to that delusion with respect to us, as much as we are deaf to its speech, I s’pose.

    Where art thou O Fabled Scientist — I smell a dogma, some Ego-Drama! We don’t know what consciousness is. Let’s shake up this prejudice about who gets to be conscious, lest we block our vision of reality with a cloudy lens: “the forest must not think or feel” or “protozoan wants are _less than_ those of humans — less important, less well-considered, less potent”.

    What foolishness! The protozoan has immense power over us. Our lives hang on the right-living of the nearby microbiotic population. Sometime after Kepler it became universally fashionable to deprive the outer world of its inner thoughtfulness. But the more fruitful hypothesis awaits the test: consciousness is revealed through communication — which requires language acquisition — and so the starting principle should not be that “the world around us is inert and dumb (because we can’t talk with it)”, but that it is chit-chatting away and if we want to join the conversation, we need to learn the tongue.

    Let’s listen closely!

    Your man Dyson knows a bit about what Kepler knew: he was penetrating the language of the living cosmos, a communion with mind at large: not the discovery of some inert math, but meaningful speech.

    I’d be right tickled if you went to check out the “Parable of the Two Suitors”, a nice little thought experiment dealing with this approach to reality.

    Keep sortin the signs outta dem bones, Mr. Oracle, Sir!

    P.S.: Didn’t you once say this? _”Your calculator is already smarter in arithmetic than any person in this room. Why aren’t we threatened by it? Because it is ‘other.’ A different kind of intelligence. One superior to us, but one we aren’t particularly envious of. Most of the minds we make including the smartest AI, will be ‘other.’ Even in the possibility space of types of conscious minds, there are 2 million other possible species of intelligence than the one type we know (humans) — each one of them unique and different as a calculator and a dolphin.”_ Bridge the gap. We shall learn to speak the other’s language.

  • bruno boutot

    It wants every human being to be connected to every other human being at electricity speed.
    It wants it to happen fast.
    Apparently, some event/transformation/transition is depending on it. (Nothing magical: only a natural, mechanical process.)
    We don’t even have to agree: we like being interconnected, so we are doing it a little more every day. So is it a result of our “natural leaning” or of “the technium” or “the machine that is us”? Is there any difference between “us” and “it”?

  • evobrain

    I have a phrase that I have coined which should give you something to think about. “All progress is progress towards the Singularity.” That is, progress has taken on a life of its own. It is beyond our control and it is hurtling the human race towards a singularity that will cause the overthrow of humanity by an AI. That AI already exists and its name is “Progress”. It exists now in a embryonic state. It is dependent on nourishment through its virtual umbilical cord from its nurturing mother which is human civilization. The Singularity will be the moment of its birth, but it is already alive.

    It is the one you call Technium.

  • Jakov Ahel

    Every complex system seems to liberate entities that constitute it. I like to search for analogy on different levels of being. For example, a cell in the human body can do more, make larger distances and generally put itself to conditions it could not do by itself. Likewise, you can see that in molecules constituting a cell, or atoms in a molecule.
    It seems to me that evolution going on in one level of complexity favours the entity, or rather a type of entity, that is most prone for communication and interaction with other entities, as well as using freely other types of entities, for its well being. Carbon atom, RNA molecule, nerve cell, human being are entity types which I found well fit in that line of thinking.
    The question is, if the earth (nature, human society topped with technology) is a body of a ‘technium’, who will she communicate, who will she mate?

  • RobertJ

    Reading your postings I find that large parts of your writing and the underlying theories depend on accepting loose analogies. Here you point out that technologies depend on each other and claim that this warrant calling it “a very primitive organism-like system.” Isn’t it enough to call this an ecology? And are you making any distinction between the technology and the users of it? without nuclear power plants, the uranium mines might well have to shut down, but mining technology or the knowledge of uranium dosn’t simply go extinct with it. If you want the organism metaphor to work, you should at the very least be able to present analogies to basic “living thing characteristics” such as homeostasis and metabolism.

    Apart from metaphors, you rely an awful lot on extrapolating very recent trends. For instance you write “Over time the fastest moving technology will become more social”. This is just picking up cellphone and social website trends. Half a century ago one may have pointed to aerodynamic cars and standardized containers in shipping to say that “over time, technology becomes more and more streamlined and efficient, wasting less and less and cutting unnessary design features.” And selective reading of evolutionary biology would back you up on this too, though the statement sounds odd and irrelevant today.

    Your postings are thought-provoking, well-written and original and I keep coming back to read more, but I don’t think I’d be better prepared for the future if me or my company acted on the predictions presented here.

    • Kevin Kelly

      @RobertJ: I am guilty of “loose analogies.” I am working on tightening them up, but in this case that will mean adding words, which I’ll do in separate posts. Re streamlining: The reason steamlining is a fad but not a long-term trend in tech, though energy power density is a trend, is that we don’t see increasing streamlining in evolution over time (it’s always present), but we do see increasing socialization over time. I admit I need to explain a lot better what I mean by increasing socialization, or power density. I intended to do that with a dedicated post for each (I forgot to mention that in this intro piece). Thanks for reading closely.

  • Brian Merritt

    Supposing technology was a woman – what then? Is there not ground for suspecting that all technologists, in so far as they have been dogmatists, have failed to understand women…

    I look forward to reading “What Technology Wants”, even though I have an emotional, perhaps irrational disquiet re your list of biases, trajectories, strange attractors

  • Amit M

    I had similar thoughts on particle decay, as to why and when a particular atom chooses to decay. The answer I came up with is that it probably has to do with cellular automata. Each atom reacts to developments in its neighborhood. A particular sequence of decaying in its surroundings could probably trigger its own decay.

    But then again, “consciousness” may pervade to the atomic level too. Who knows :-)

  • Kent Schnake

    Kevin, your write: “We can see the direction of technology in the direction of life and evolution (that is if you accept evolution has a bias as I do).”

    As a Christian I believe that God is sovereign and has a plan. I can accept the idea of evolution as change directed by a bias, God’s sovereign will. I also believe that we are made in his image, hence the existence of freedom and choice.

    Without the bias, it would seem that evolution is absent design and consequently without meaning.

    The bias also allows us to distinguish that which is beautiful from that which is ugly and that which is good from that which is evil.

    However, it is my impression that much of the argument about design versus “evolution” is at its heart and argument about choice versus chance. If all change happens by chance, undirected, then it would seem to me that all that happens is meaningless. Could bias be a small word that implies meaning, direction, purpose?

  • Doug King

    You should be asking “what does biology want”. The inherent “agenda” of technology is the inherent “agenda” of biology. The following video simulation might help: Accurate Scientific Visualizations of the T4 bacteriophage infection process and replication

    Although it is not shown in the simulation, keep in mind the assemblage of the raw materials is accomplished with the help of Brownian motion, a high speed random positioning of the raw materials, similar to a box full of vibrating parts. When the components match up correctly the create the next link in the structure. The raw materials travel enormously long and convoluted paths before they happen to find a place where they fit, bind, and remain. The core of the matching engine is the DNA or RNA which starts the ‘crystallization’ of the final virus, bacterium or organism, the ultimate goal of which is to spread the DNA or RNA, eventually in a mutated form.

    The structure of DNA is pure information, seemingly with a purpose or agenda. So biology has this agenda and technology is the extension of biology. We have our own specific biology, but the principles and forces of biology do not belong to us and ultimately technology will not either. We are artifacts of biology, of DNA.

    The notion of the singularity is the acknowledgment that in our current biological incarnation we will not be able to own or comprehend a technology that will be the future manifestation of our biology.

    When you watch the raw materials of biology seemingly self-assemble, and replicate, and over time grow in complexity, you have to ask what are the forces at work that allowed this to come to pass. An important word in the last sentence is “you”. For without our senses and minds to observe, the raw materials of infinite scale, infinite time, infinite particles of matter, infinite energy and conceivably infinite dimensions of space would not be observable, therefore not a thing; nothing. The anthropic principle re-states “I think, therefore I am”. So our biology allows us to observe our biology, the universe looking in on itself.

    Given infinity in all it’s dimensions, we eventually come to a point where we have structures that are capable of observing, self modification and self awareness. At this point we look out and say “the universe is a finely tuned thing that must have been designed intentionally for us to exist. Any minor deviation from initial conditions and it would be a wasteland. There must be a hand of God”.

    The next step of biology is to transition to a higher level of information replication, information gathering, complexity, self modification and self awareness. This will be accomplished with other non-carbon molecules of matter so long as the replicate / mutate / survive formula is the driver. So we don’t need carbon life to have self awareness and intelligence, choice, urges, agendas. However, I believe the next level of awareness will retain it’s parentage and at the core will be human just like we retain some of our ‘lizard brain’. I don’t think there will be a clean break.

  • kirk

    Not sure why the trope “it’s not yet alive” survives. If meat puppet consciousness is heterophenomenological (emerging from multiple “clumps” of agency) then perforce the addition of agency from the pure technology “side” have already added to the meat puppet “side”. I view the technium as the emergent quality of the yin (tech) and yang (biota). We knocked on heavens door 150 years ago and sometime in the recent past we tunneled through to the other side.

  • Lloyd Mintern

    The key thing is that insofar as technology is nature, it is NOT science.

  • Lloyd Mintern

    Insofar as it “wants”, it does not think.

  • Heiko

    Dear Kevin, i’ve been following your blog for some time, and enjoy your thought-provoking posts.

    Not sure if it’s useful to nitpick, since it’s just a minor detail – but the term “strange attractor” in this article jumped out at me as a little unnecessary. Complex systems may be more likely than not to converge to some attractor, but that attractor doesn’t have to be “strange” (i.e. being of a non-repetitive, fractal nature).

    In fact, i suspect most complex systems are going to have “extinction” (a single point in outcome-space) as their attractor, as opposed to some more complex type of long-term attractor.

    Either way, it’s a minor point :)

    Thanks for sharing your thoughts on so many topics with all of us out here.

    cheers, :) Heiko