3D-printing: form and substance

3D printing is widely predicted to be the future of manufacturing, by groups as varied as the excitable fringes of Silicon Valley and the accelerationist left. They believe that everyone will have a 3D printer in their home, just like everyone has (had?) a ‘2D’ printer, and these printers are going to produce the quotidian objects of our lives. Just as a printer allows you to produce a piece of paper with whatever words or pictures you desire, your desktop 3D printer will produce any arbitrary physical object.

This is, as everyone knows, still a dream of the future rather than our present situation: 3D printing is still primarily useful for rapidly building physical models of objects to be produced in conventional methods (whether buildings or aerofoils).

The fundamental distinction between 3D and 2D printing is illustrated by what was one of the most eye-catching early stories about 3D printing: the 3D printed gun. The news that someone had managed to produce a gun provoked panicked predictions of killers downloading plan from the dark web then embarking on a murder spree with their freshly printed assault rifle. The reality was very different. The gun could only be fired once before collapsing, and one of the key components (the firing pin) was manufactured separately (this story reminds me of ‘internet ham’). It may have looked like a gun but it wasn’t a useful weapon. Both 2D and 3D printers can reproduce the form of objects, but neither can reproduce their substance. A photocopy is not made of the same material as the original but because they resembles each other very closely, they are functionally equivalent. For most objects we own, shape isn’t enough, because materials matter: nobody would use a chocolate teapot.

This poses a big problem for the concept of a universal 3D printer: because the differences between the materials also demand different manufacturing processes, for example the laser heating 3D printing techniques that work for metals will wreck plastics. In many cases additive manufacturing (the layer-by-layer processes mostly used by 3D printers) is unlikely to ever be able to deliver the properties needed for an object. The single crystal slabs of silicon used for computer chips or solar cells won’t be grown using a 3D printer because the orientation and size of the crystals critically matters. There is such a variety of manufacturing methods for all the different kinds of things we made that it is difficult to imagine a single machine being able to replace them. This shouldn’t be seen as a failure - we should instead embrace the diversity, and the range of different properties that are afforded.

One common criticism of technological evangelists is that they fail to think about society, because they start from an individualistic point of view. It’s also important to think about the physical, rather than digital abstractions - a catastrophic example being Theranos, marketing tests for diseases claiming to use a volume of blood which wouldn’t contain a single molecule of the marker. It may be plausible for the 24th century Star Trek universe to have manufacturers of everything, but if we are to speculate about society within our lifetimes, we should take more care to think about materials.

Figure: Object (or Luncheon in Fur), by Meret Oppenheim, 1936. Flavia Brand/Flickr