Our best machines are made of sunshine

The title is a quote from Donna Haraway’s A Cyborg Manifesto (1985). Here’s the relevant passage:

The third distinction is a subset of the second: the boundary between physical and non-physical is very imprecise for us. Pop physics books on the consequences of quantum theory and the indeterminacy principle are a kind of popular scientific equivalent to Harlequin romances* as a marker of radical change in American white heterosexuality: they get it wrong, but they are on the right subject. Modern machines are quintessentially microelectronic devices: they are everywhere and they are invisible. Modern machinery is an irreverent upstart god, mocking the Father’s ubiquity and spirituality. The silicon chip is a surface for writing; it is etched in molecular scales disturbed only by atomic noise, the ultimate interference for nuclear scores. Writing, power, and technology are old partners in Western stories of the origin of civilization, but miniaturization has changed our experience of mechanism. Miniaturization has turned out to be about power; small is not so much beautiful as pre-eminently dangerous, as in cruise missiles. Contrast the TV sets of the 1950s or the news cameras of the 1970s with the TV wrist bands or hand-sized video cameras now advertised. Our best machines are made of sunshine; they are all light and clean because they are nothing but signals, electromagnetic waves, a section of a spectrum, and these machines are eminently portable, mobile — a matter of immense human pain in Detroit and Singapore. People are nowhere near so fluid, being both material and opaque. Cyborgs are ether, quintessence.

I interpret Haraway’s quote quite literally: our best machines are made of pure energy, of the same stuff as sunshine. Think of fiber optics, or of all the signals that fill the air broadcasting information at some frequency of the electromagnetic spectrum. Haraway’s argument is ultimately that the image of the cyborg overcomes many of the difficulties associated with the modernist conception of the human being, and this short quote captures one of the more compelling ways she tries to overturn modernist thought.

But there is something sneaky going on in Haraway’s quote, so I have a question:

Are electronics, strictly speaking, machines?

The common notion of a machine is of some device that has moving parts. The steam engine is a paradigmatic example of a machine. Simple tools with no moving parts, like a hammer, aren’t usually consider to be machines. Transistors don’t move either; they just store an electronic charge. So electronics seem to fall pretty far outside these paradigmatic cases. We learn in grade school that the inclined plane is called a ‘simple machine’, because it can be used to do work, which is defined in terms of movement over a distance. In this sense, transistors don’t do any work, and so wouldn’t even qualify as a simple machine.

For the record, Wikipedia puts ‘electronic machines’ into a separate category from other machines (including biological machines), so at the very least there is an important distinction to draw. But by the common notion of ‘machine’ discussed above, biological machines share a closer family resemblance to the steam engine (they move around and do work) than do electronics. So my question might be recast: Are electronics a species of machine, or do they belong to their own family?

Wikipedia also identifies ‘electrical work’ as a special case of mechanical work, but since the source referenced here (Wiki’s page on electrical fields) doesn’t employ the notion of ‘work’ at all, this seems to be an ad hoc way of fitting electronics into the conventional idea of ‘machines’, and so I don’t find this very persuasive way of settling the above question. However, I suspect there will be many objections along these lines.

There is another notion of ‘machine’ that we inherit from 17th century philosophers like Descartes, where ‘machine’ simply means ‘obeys physical laws’. On this view, a falling rock is ‘mechanical’, even though it wouldn’t qualify as a machine in any of the above senses. Clearly, electronics would classify as machines in this sense. And here’s where Haraway’s quote gets interesting; the idea of a machine takes on a very negative connotation in the enlightenment, as correllary base materialism and determinism that is to be distinguished from the vital or spiritual force supposed to exist in living things, and especially in the minds of men.

But life derives its energy from the sun, so sunshine itself paradoxically inherits the mystical properties attributed to life. Haraway’s quote plays on this paradox by pointing out that our best machines today operate on the same principles as sunshine, and thereby turns the modernist conception on its head and endows our machines with the spirituality that we are increasingly incapable of finding in ourselves. We are clunky and material; our machines are quintessence.

But if electronics don’t count as machines, then maybe Haraway really hasn’t overturned the modernist view of man and machine, but has pointed in a third way: that our best technology is neither living force nor brute material but a new kind of creature altogether, one that can operate simultaneously on the principles of inorganic nature and at the rhythm of organic life. I’ll leave this third alternative vague for now, because I’m not sure its worth developing if my original question is a bad one.

Some instructive comments from the D&D thread:

If you read a little past the quote you gave, she goes a bit more into what she’s trying to say. It’s completely about this blurring of the organic/inorganic. You can’t tell that these are actually machines anymore, therefore it’s difficult to see where the machine ends and the inorganic begins.

“The ubiquity and invisibility of cyborgs is precisely why these sunshine-belt machines are so deadly. They are as hard to see politically as materially. They are about consciousness – or its simulation.”

The “deadliness” of these “sunshine-belt” machines is found in that we, as humans, have lost our fear of integrating with them, and as such, are becoming cyborgs. With the politics associated with the cyborg – a breakdown of gender, sex, race, etc – this change is dangerous in that it goes against the modernist viewpoints and challenges the status quo. The danger is not for the cyborg, but for those who have yet to grasp onto its ideals.

Its about blurring the organic/inorganic, the human/animal, and the physical/nonphysical. Part of my question is whether these things should be as blurred as Haraway suggests. Maybe we ought to reserve the word ‘machine’ for particular kinds of things, and use different words for describing other kinds of things. Maybe a better way of rephrasing Haraway is to say: Our best machines aren’t machines at all.

I’ve made this kind of argument before with respect to robots. It is easy to agree with Dennett when he says we are made of lots of tiny robots, and he means that in exactly the sense of being made of lots of tiny machines. But I want to object: no, there actually are robots in the world, and they don’t look or act anything like the little mechanisms that make up your body. Robosapien and Kismet and Optimus Prime are all part of the same family, and the little mechanisms in your body just don’t belong to that family. Blurring the line between the organic and the inorganic might be helpful for resolving some crucial metaphysical disputes, but when that blurring begins to confuse our understanding of the kind of creatures that uncontroversially inhabit our world, like machines and robots, then perhaps its best to step back and take stock.


  1. If these things are energy and more importantly light energy then they have both wave and particle properties as such it makes sense for any “thing” like this to not be easily defined in a particular category.

  2. “Wikipedia also identifies ‘electrical work’ as a special case of mechanical work, but since the source referenced here (Wiki’s page on electrical fields) doesn’t employ the notion of ‘work’ at all, this seems to be an ad hoc way of fitting electronics into the conventional idea of ‘machines’, and so I don’t find this very persuasive way of settling the above question.”

    I don’t get it. Our machines that are “made of sunshine” are all there to do work. My computer and ipod etc are there to give me information and help me communicate. We’ve gone from spoken word to hand written letters to morse code to telephones to email. That’s not work?

    I’m not sure how our electronics are made out of sunshine either. They’re still made out of materials (I almost said “they have material bodies”). I guess you can say that they run off of sunshine or whatever, but they sure as fuck aren’t made of sunshine.

    I guess I don’t really understand this post at all.

  3. This all seems very confused to me, but maybe I’m just confused. Here are the electronics that figure most prominently in my life: my clock radio, my other radio, my iPod, my stereo, my laptop, my computer at work, my television, my PS2, my cell phone, my other cell phone. All of these things have moving parts, so all of them fit the common notion of a machine. Admittedly, I can think of electronics that seem not to have moving parts (e.g., a wireless router, a cable modem), and maybe I’ll come back to those in a minute, but let me just say for now that I suspect that the problematic class of devices is smaller than you suspect.

    Maybe this is partly explained by one of your key examples, the transistor. It’s true that a transistor doesn’t move, and that the transistor replaced something which did — a switch or a valve usually. But isn’t it a little bit odd to call the transistor a machine, rather than part of a machine? And doesn’t doing so allow you to bracket off the moving parts of electronic devices as if they didn’t matter?

    This actually reveals what may be an interesting question, and one which seems to overlap with some of the arguments you’ve made about Google, the internet, and wikipedia in the past. Are there non arbitrary borders we can draw to say where a machine ends? I could, if threatened, put together an argument to the effect that my laptop and wireless router are part of the same machine. Then, because my laptop has moving parts the machine would have moving parts and the common definition would be satisfied.

    Admittedly, that argument seems sophistical, but that’s how this whole project of drawing distinctions between what isn’t and isn’t a machine feels to me. I need to know why the distinction matters to understand why it’s worth making.

    Still, if I were to engage the project the second definition seems more promising. But then I have to ask, could we get a little more clear on ‘movement’? Doesn’t it make sense to say that current moves through a transistor, or through any kind of circuit? Maybe you answer that the transistor isn’t moving. But why isn’t the current part of the machine? Or maybe your answer is that when we get down to the nitty gritty physics of the situation, current doesn’t move. Well, I don’t know. I do know that, depending on the strength of the current going into a transistor there either will or will not be a measurable current on the other side. Which is to say that there is a measurable change of state it current that passes through a transistor. This seems not to be all that different from a measurable change in the altitude of one of galileo’s balls.

  4. Also, I was taught in elementary school that a lever is a machine. But if a lever is a machine, then so is a hammer.

Submit a comment