An Engineer's Perspective on Deleuze and Guattari's Thousand Plateaus

You might want to buckle in: this could be a long and bumpy ride. I'm about to tackle Deleuze and Guattari's A Thousand Plateaus. This was an outcome of the Centre for Disability Studies' reading group in which we tackled Margrit Shildrick's Why Should Bodies End at the Skin? and mostly debated terminology around Lines of Flight, Assemblages and Becoming. The comments I had upon announcing that I'd do this weren't encouraging: From "Above and beyond the call of duty" to "You need to drop acid to read it... that, or it'll make you feel like you've dropped acid".

Well, I've done it. Read the book that is, not dropped acid. And now I shall do my best to explain (or should that be eng-splain?) it from an Engineer's Perspective - or horribly misinterpret it all. Like I said: this could be a bumpy ride, and it'll be going round the houses. Sorry about that.

Just for the record, I'm starting this on January 9th 2017. Let's see when I finish...

Caveat Lector
Warning: once again I am about to discuss matters outside my sphere of expertise, so this stands more for discussion and correction than for education. Also, when I say an engineer's perspective, I mean it literally: that engineer is me. For other engineers, your mileage may vary. It's also worth noting that I am of a different political stripe than Deleuze and Guattari. If I'm honest, I find people squishy and unpredictable and that makes me uncomfortable. I like hierarchies and boundaries and structures and norms and machines. I suspect that I'm very much part of the State Deleuze and Guattari complain about, so that filters my take on things. I get the feeling that were we to meet, we would disapprove of each other.

Then again, I'm a great believer in trying to see other points of view, and I also like prodding things to see what happens.  You have been warned.

Obscurantism and Complexification
As the warnings I received attest, the language and presentation in A Thousand Plateaus are... tricky. Of course, I am reading in translation, so perhaps that is part of the problem, though I don't envy the translator (Brian Masumi) the complexity of ideas and wordplay here.

Indeed, Felix Guattari was among those criticised by Richard Dawkins among others for being deliberately obscure. This takes in matters such as the Sokal affair, and probably touches on some of the fundamentals of the philosophy of science and positivism vs constructivism in which I am not well-versed (I'm just sufficiently well-versed to know that the debate isn't easily characterised in a few words). I suspect we'll be coming back to this issue and matters of knowledge,  reality and uncertainty later in the post. The key point is that it is hard to differentiate between something badly explained and something badly conceived in the first place. Between something difficult to grasp and plausible-sounding rubbish. A Thousand Plateaus is written in an ambiguous style: imprecise, perhaps? Impressionistic? Not my prefered style of academic writing. What works well in, say,  If on a Winter's Night a Traveller or The Sound and the Fury or The Unconsoled is less welcome in a work that seeks to inform.

Guattari seems opposed to reductionism and simplification, preferring "complexification" (I'm relying here on what Wikipedia says that Guattari says - I have not read his Chaosmosis). They are interested in the Nomad rather than the State: in flux and change, you might say. I have some sympathy with this view: there is a danger in oversimplification in science and engineering in particular, extrapolating from models that rely on simplifications and assumptions and making unwarranted sweeping generalisations. Yet there is the reverse danger - saying "everything is just a bunch of stuff, so we might as well just muddle through any old how". There is also the risk that complexification becomes about withholding or obscuring knowledge, rather than communicating it; or making your ideas so intangible and fluid that they cannot be critiqued. It's the sort of criticism one might well level at the bombastic style of A Thousand Plateaus. Happily, they do actually criticize that approach, as we shall see under the bit on the Body without Organs.

Of course, all this presupposes that the purpose of the work is to inform. It may not be: Deleuze was apparently committed to "thinking differently", and read in that light, A Thousand Plateaus becomes a provocation to the reader to think anew about the familiar, more about the impression than the detail. In that spirit, let me give Deleuze and Guattari the benefit of the doubt and give you my impressions.

Specifically: what impressions from this book did I find useful? To explain that, I'll need to take a detour into Soft Systems Methodology (SSM).

Flux, Appreciative Systems and SSM

Peter Checkland's SSM was hugely influential on me (and our research group as a whole) during the course of my PhD. It grappled with exactly the interface between the soft, fuzzy, amorphous nature of the wicked problems presented in design, and the rigorous, structured, systematic tools we try to apply to them.

The roots of SSM lie in systems theory and cybernetics. Systems Theory started out as one of those Grand Unifying Theories, or a Theory of Everything. In this view, everything is a system: a set of interacting parts. Those parts are themselves systems, made up of smaller parts which are also systems... and so on.  Hence for any given system, we can speak of its subsystems (the parts making up the given system) and its supersystem (the system it is part of).

You will note that this creates an onion-like layering: every system is part of a supersystem, which is part of a wider supersystem which is part of a still wider supersystem and so on until you get to the whole of reality. The Universe, or multiverses or whatever. And of course, every system is made up of interacting subsystems, which  are each made up of their own smaller subsystems which are made up of even smaller subsystems, and so on until you get to whatever the smallest particle in the world turns out to be. It was meant to be the atom, but that has long since been subdivided, and as the recent discovery of the Higgs-Boson demonstrates, they're still going.

One of the key features of a system are its emergent properties: properties not exhibited by its individual subsystems, but generated by their interaction. For example: if I have an electric motor, it does nothing. If I have a battery, it does nothing. Put the two together, though, and suddenly the motor draws current from the battery and begins to spin: I have the emergent property of rotary motion. Attach this to a chassis and a set of wheels, and I have linear motion. In each case adding more systems creates new emergent properties. Indeed, the putting systems together to get desired emergent properties is the basis of engineering design.

Of course, I don't need to worry about every subsystem: just the interfaces between the ones I'm combining. I don't need to know about electrons or quarks or Higgs-Bosons to attach a battery to a DC motor and a set of wheels to get them moving. Not at the scale a mechanical engineer needs to work with, anyway. Nor do I need to know about the entire supersystem: the moon will have a gravitational pull on this makeshift electric buggy, but it's pretty small, and I can safely ignore it. I really just need to worry about the immediate environment. If, on the other hand, I am making a rocket to send a satellite to Jupiter, the gravitational pull of the moon is going to be very important. Which highlights a key issue: for all that systems may be comprised of energy and matter that can be measured and modelled, exactly where a given system is taken to begin and end (the "system boundary") is amorphous and depends upon what you want to do with it.

   In the case of, let's say, a car, this is pretty straight forward and extremely handy. As these products have become more complex, they are no longer designed and made by a single artisan or craftsman: instead, there will be a range of teams of specialists responsible for different parts. Someone designing the chassis, someone the interior, someone the braking system, someone the engine, someone the steering, someone the hardware for the onboard computer, someone the software, and so on and so forth. Taking a systems perspective means you can give each subsystem its own requirements and the team can go away and build it knowing that as long as they meet those requirements, their part will work with the others.  You've got this much space, and you need connectors here and here - the rest is up to you. This doesn't eliminate all problems. As Bucciarelli (1994) notes, different engineers have very different perspectives on what the system is and does - a bit like the blind men and the elephant - but as long as someone is overseeing the process, this doesn't matter too much.

    This approach led to huge gains in the aerospace and automobile industries in the mid-20th century, naturally leading to the suggestion that we should apply it to other systems: businesses, judicial systems, governments. There is, however, a problem here: these social systems don't enjoy clear and objective boundaries. They are amorphous and change overtime. Different actors may disagree on their purpose.

   Social networks are a topical example. Facebook and Twitter exist objectively, as software and hardware. A set of datacentres and algorithms. But their emergent properties include the behaviours of their users: fake news and trolling being two recent controversial examples. The problem is that trying to deal with these problems relies on predicting the behaviour of users. It's not the same as solving a bug (hard enough in itself) or excessive load (remember the Fail Whale?). And herein lies the problem: I can model the behaviour of users, but I can never know it's right. That means that unlike the physical behaviour of the system, they are very difficult to predict. Thus, Vickers coined the idea of Appreciative Systems. We exist in a complex flux of events and ideas, and in order to determine which actions will lead to desirable outcomes, we must first appreciate (i.e. perceive and interpret) this flux. This necessarily involves simplifications, attending to some things and ignoring others. My illustration of this (as drawn for my thesis back in 2005!) is shown below: you can find a really good description of Appreciative Systems here.

Vickers' Concept of an Appreciative System, as Visualised by Checkland and Casar, and redrawn by me for my Thesis.

The Tacoma Narrows bridge is again a good example: the developers probably ignored the gravitational pull of the moon and Mars and Jupiter. That was reasonable. They also ignored resonance, which proved to be a mistake: a failure of appreciation. And therein lies the rub: you can't model every relationship between systems, so where do you draw the line? The same applies with human systems: I need to predict user behaviour. Phishing, trolling, identity theft, social engineering - they all require designers of systems to appreciate complicated and sometimes unpredictable behaviour.

Which isn't to say that systems ideas are useless in human or organisational  contexts. Rather, Peter Checkland used them as a tool for inquiry: by modelling a system from the perspective of different stakeholders, you can look for areas of agreement and disagreement and begin to discuss actions for meaningful change. This is Soft Systems Methodology: Systemic Inquiry rather than the assumption of a simple systemic world. 

SSM has developed over several decades, and Checkland has, with others, written several books on the subject. Systems Thinking, Systems Action lays out the original thinking behind it, including a handy potted history of systems theory and science. Soft Systems Methodology in Action is a series of case studies, illustrating the application of SSM (the significance to A Thousand Plateaus will appear soon), and the 30 Year Retrospective included in later editions of these books (dating to 1999 - so a little old itself now!) is an excellent introduction to the topic.

A further concept worth addressing is self-assembly: that random combinations of parts can spontaneously create ordered systems with emergent properties. This tends to apply at a molecular or crystal level, and doesn't really crop up that much in the work of Checkland, except perhaps as a natural system. Worth noting that this can also apply to social systems: as groups come together and formulate norms and behaviours, something that will crop up as we discussed Deleuze and Guattari's concept of rhizomes.

Anyway: this is a very sketchy outline of systems concepts, particularly Appreciative Systems, but they are central to my reading of A Thousand Plateaus.

A Thousand Plateaus

A Thousand Plateaus was published in 1980, written by philosopher Guillaume Deleuze and psychotherapist Felix Guattari - one of several collaborations between them, and the second volume of Capitalism and Schizophrenia. I have not read the first volume, Anti-Oedipus, just to make my life harder.

The book comprises a series of "plateaus", each identified with a date and a concept related to the topic. They were (according to the introduction) written piecemeal, and are intended to be read that way: dipped into in any order. I didn't do that: I started at the Introduction and read to the end.

This structure is not as mind-bending as it sounds. Essentially, the book opens by introducing the concept of the rhizome, then proceeds to apply it to a variety of different concepts, before concluding by summarising the ideas that have developed. In this sense, the plateaus are like the case studies in Soft Systems Methodology in Action. Not as formal, but essentially the book could be understood as: "here is a way of looking at the world... Let's see what happens when we look at different concepts in this way..."

I won't review the chapters in detail, but instead look at the concepts that evolve through them, particularly in the introduction and summary.

Rhizome
The book opens with the concept of the "rhizome" - in biology, the underground stem of a plant that spreads out into a complex network of roots and other plants - which they contrast with the hierarchical "tree" of conventional systems science. Whereas a tree (in biology) is vertical, a rhizome is horizontal. So, Deleuze and Guattari contrast their approach with the more conventional hierarchical "tree" which classifies and subdivides from root to leaves.

Deleuze and Guattari set out six principles for rhizomes:

1) Connection and 2) Heterogeneity: "Any point of a rhizome can be connected to anything other, and must be" (p7).

Connection is tautological: it wouldn't be a rhizome without connections. In this sense, a rhizome is like a system: a system needs to have constituent parts, or it isn't a system. Deleuze and Guattari refer to this as an assemblage, and they refer to assemblages as producing affects: analogous to emergent properties. However, Heterogeneity is more interesting: any two (or more) things can be connected in a rhizome, and have the capacity to affect and be affected. Is this the same as a system? Can any two things form a system? Or only those with emergent properties? In other words, if the parts don't interact, do they still form a system? Then again, if the elements of a rhizome don't interact, it isn't a very interesting rhizome. Perhaps two parts that don't interact just make a pointless system. Is an affect different from an emergent property? Is an assemblage without affects still an assemblage?

By this definition, I can form a rhizome from the star Betelgeuse and the poem Mid-term Break by Seamus Heaney. Do they form a system? It's not obvious that they interact at all, the poem being a designed abstract system, and Betelgeuse a natural system, to use Checkland's classification. They certainly don't have a hierarchical relationship. Then again, they don't make a very interesting rhizome, either. But what if I, say, read Mid-term Break, then look at Betelgeuse. Do I think about it differently? Does it lead me to think different thoughts? A poem about death, and a dying star - perhaps it causes me to reflect on mortality, and the human scale compared with the Universe in a way that I wouldn't have if I'd encountered them more separately. Is that an affect after all? Is it an emergent property? After all, the poem has changed my appreciation of the flux of facts and events. Of course, that makes me (the one being affected) part of the rhizome - the poem and the star may not interact physically (or even conceptually), but each changes my interaction with the other.

3) Principle of multiplicity: " it is only when the multiple is effectively treated as a substantive, "multiplicity," that it ceases to have any relation to the One as subject or object, natural or spiritual reality, image and world." (p8)

Well, the grammar is unhelpful: you can see what Dawkins was complaining about.  Here, I found the explanation at this link helpful. If this interpretation is correct, then the point is that a rhizome is a multiplicity of things, not a multiple of units. I suppose this follows from Heterogeneity: it is a network of potentially unlike things. Again, the Betelgeuse/Mid-term Break example makes this point: not a set of stars, or of poems, not a set of anything, but a multiplicity of unlike things. Also worth noting is that, by virtue of the Connection principle, a rhizome is not a sequence or a hub, or a hierarchy. It doesn't have (or doesn't need to have) a definite start or end, or centre. This leads neatly onto the next principle:

4) Principle of asignifying rupture: "a rhizome may be broken, but it will start up again on one of its old lines, or on new lines" (p9)

This is where it starts to get interesting. This principle imbues rhizomes with a life of their own. You can rupture them, but they'll just form a new rhizome. This is where self-assembly comes in: you may separate out parts of a rhizome, but if you aren't careful they'll form new systems with new properties, and not necessarily the properties you want.

This brings in four important concepts: becoming, lines of flight, territorialisation and overcoding.

According to Deleuze and Guattari, everything is in a state of Becoming. We could say that everything is in a state of flux or changing - I think that isn't quite what Deleuze and Guattari mean (we'll revisit this under lines of flight), but it does raise an important point. In a systems perspective, parts are brought together to make a new whole and emergent properties... well, emerge. Of course, this implies that parts weren't always part of the system, and won't always be. If we take a car, for example: components and assemblies are brought together to form it, then tyres and timing belts and gaskets and brake pads are removed and replaced as needed. Taken to the extreme, you get the Ship of Theseus or Trigger's Broom: every piece is substituted, but the car, as a system, remains. Not dissimilar to the human body, where cells are renewed and replaced, I suppose.
  
   Of course, the parts removed from the car don't vanish. They get reconditioned or melted down or thrown away, and become parts of other systems. Likewise the body, again. I consume food and incorporate it into myself, I burn some of myself for energy and breathe out carbon dioxide which is later incorporated into a plant or the atmosphere... I sweat, shed skin and hair - and these feed into other systems. In this sense, everything physical is in a state of flux, removed from one rhizome/system/assemblage and incorporated into another.

Linguistics is probably another good example (I say as a non-linguist): language evolves constantly and uncontrollably. Bits of other languages and jargon get incorporated, words get abandoned, etc. In essence, Becoming is the process of breaking and reconnecting parts of rhizomes, and it is going on all the time. Becoming - change, flux - will go on whether you like it or not.

The Line of Flight (flight as in flee, escape rather than birds or planes) captures this - the line of flight defines the direction of change, of becoming. Or of rethinking, perhaps, since the line of flight implies something deliberate. A desired direction for the becoming instead of just leaving events to take their course and parts to self-assemble as they happen to fall.

Following a line of flight means breaking down and reassembling rhizomes:  deterritorialising parts from one or more rhizomes and territorialising them into another, steering their Becoming. This notion of territory carries the notion of taking control of something. It is also referred to as overcoding (as in overwriting): obliterating the original and replacing it with something new. It is, I suppose, a way of imposing one's will on the rhizome or system, rather than letting it take its own course.

5) Cartography and 6) Decalcomania: "a rhizome is not amenable to any structural or generative model" (p12).

These last two are expressed as a rhizome being "a map and not a tracing" (p12).  By their reckoning, the purpose of a tracing "is to describe a de facto state, to maintain balance in intersubjective relations or to explore an unconscious that is already there" whereas a map "is entirely oriented towards an experimentation in contact with the real." (p12). It's the difference between giving someone directions (which must be followed as prescribed, and are useless if you wander off course) and a map (which allows you to find your own path and explore areas away from it). It's not quite that a rhizome is a map of what could be rather than what is: more that the rhizome permits multiple interpretations and reinterpretations. This is where the link with systems theory becomes interesting: the vagaries of system boundaries, the well-defined "decal" of a designed physical system vs the porous and shifting boundary of a social system.

Of course, Deleuze and Guattari are all about "nomad" thought, pushing back against the boundaries and structures imposed by norms and convention, which is what the rhizome concept is all about.

Body Without Organs

So far so good. Rhizomes are a way of thinking about the world, an alternative to hierarchy and imposed structure. Now we come to a more difficult concept: the plane of consistency or Body without Organs. Initially, I took this literally: as in  a body of work, or a body of water - something that coheres conceptually, rather than physically. Actually maybe a body of water does cohere physically, but I'm thinking here that a lake or river remains the same body of water, even though the water molecules making it up  change. Yet, I'm not sure that's quite what they mean.

    The Body without Organs is a phrase apparently borrowed from playwright Antonin Artaud, and refers to the "virtual dimension" of the body, away from its physical manifestation. Not a platonic ideal, exactly, but a potential - all the infinite ways that the limited, defined, physical body might be interpreted or reinterpreted. It is "opposed to the plain of organisation" (p507): on the Body without Organs, the plane of consistency, any line of flight can be drawn; any part of the rhizome can be connected with any other part.

   If I have interpreted this correctly, then the Body without Organs is the rhizome, the map, and the process of deterritorialisation (decoding) and reterritorialisation (overcoding) produces tracings from it. Some of these "tracings" will be imposed by norms and conventions around us, but the Body without Organs permits experimenting with different lines of flight. Not all of these experiments will generate a useful new way of thinking about the body in question. But some will.

    Language might be a good example. I can put different sounds together in all sorts of combinations, most of which will be rubbish and no use for communication. I could arbitrarily make up my own language and attach meaning to sounds, but no one else would know what it meant. I would need to overcode the listener's meanings for the sound with my own. I suppose this is how children learn to speak: by attaching and reattaching meanings to combinations of syllables. Misusing and mispronouncing, reusing sounds heard from the family and those around them, tuning them until they prove useful. Or an adult with jargon: we see a new word or phrase and try to attach meanings to it. I'm doing that right now with the term "Body without Organs", deterritorialising the meanings of the terms in my mind, and putting them together in new ways until I find one that makes sense.

   Take the word Spam. How did a trade name for chopped pork and ham come to mean electronic junk mail? Someone had to overcode the syllables to form the name; Monty Python's Flying Circus had to deterritorialise it and appropriate it for their sketch, thereby overcoding the associations with it in a subset of the population who would later reterritorialise the phrase as part of the jargon of electronic mail. And so forth.

    A more physical example, perhaps lies around systems boundaries in engineering. The boundary of an engine may be physically well-defined, but it is still, at some level, arbitrary. In design, the system boundaries define who has responsibility for which sub-systems, and what is taken as given: but they can be redefined.

    Take Apple's new headphones. In PCs and mobile phones headphones are normally an external system, taken as given, interfaced with a 3.5mm jack. You can put whatever headphones you like in. By removing 3.5mm jacks from their devices, Apple have effectively territorialised or overcoded headphones, absorbing it into the rhizome of the mobile phone.

This is perhaps a slightly naff example of the Body without Organs, but it serves the point (I hope). Experimenting with the concept of what a phone is, what a computer is, what a music player and a camera are, have led to modern smartphones: these are lines of flight along the Body without Organs, experimenting with different potentials.

Deleuze and Guattari view the Body without Organs as a basis for experimentation and rethinking. However, they warn against two extremes. Firstly the Body without Organs can become empty, like a drug addict (p163), devoid of structure; no longer a map for experimentation, but a chaotic free for all. Like trying to talk your own language and expecting people to understand or randomly clagging things together and demanding the result be treated as original and insightful when there was no insight behind it. At the opposite extreme, the Body without Organs can be cancerous, trying to replicate its structure over everything: insisting that no new words can be admitted to a language orc that words cannot change their meaning; or insisting that art must be reproductions of something real. 

In essence, a cancerous Body without Organs would be saying "Look, I've rethought the issue, and here is the right answer, and now no one else should rethink it".

Let's Recap

And there we have it: those were the main ideas I could locate in A Thousand Plateaus. Let's just recap the main points:

1. Everything is in flux, a state of Becoming.

2. Rhizomes represent an alternative way of thinking about systems, being assemblages of different parts that combine to create affects: everything has the capacity to affect and be affected.

3. Because a rhizome is "a map, not a tracing", it can be explored and rethought, allowing its Becoming to be steered along certain Lines of Flight.

4. These Lines of Flight represent different ways of combining the elements in a rhizome - tracing different routes through its map - by reterritorialising them into new assemblages.

5. For a rhizome, there exists a Plane of Consistency (the Body without Organs): all the possible Lines of Flight that could be taken. This plane is purely conceptual: it cannot be grasped in its entirety, only investigated or explored. Exploring this plane is the process of forming a Body without Organs and this can be done chaotically (clashing ideas together without control), cancerously (trying to create a new rigid view to overcode others) or healthily (experimenting with different combinations and following those that are most promising). The map analogy is helpful here - the difference between running around wildly trying to find your way; drawing a single path and insisting everyone use it; and exploring different routes to see which have value.

Gravity provides an interesting example of this. At one extreme, we have the fabricated interpretation put forward by Alan Sokal in the Sokal Affair: an example of an empty Body without Organs, a haphazard breaking down and recombining of terms and concepts into something useless - rejecting gravity won't save you if you dive out of the window (to paraphrase Sokal very, very loosely). On the other hand, we have to recognise that Newton followed this process in developing his model of gravity, a model rethought and reassembled by Einstein. In this sense, we can view Einstein's thought experiments as a healthy exploration of a Body without Organs.

Also, note that sometimes it is useful to make a tracing. Drawing a system boundary is a process of agreeing to use the same tracing from the rhizome. This is helpful in co-ordinating different actors, just as a given marathon follows a fixed route, rather than giving competitors a map and saying "Choose any 26 miles" (contrast with Orienteering, perhaps...). Likewise, if Airbus have asked you to design a new engine, and you come back with a whole plane, you won't be popular.

The key point is recognising that the system boundary agreed on is a tracing, and not the absolute fixed and inevitable truth. In fact, this is the very process of appreciation outlined by Vickers, and expanded by Checkland (among others) in Soft Systems Methodology - hence the lengthy digression around those subejcts at the start of this post. Of course this raises another question: do the concepts addressed in A Thousand Plateaus have anything to add this? Actually, I think they do. Rhizomes, the Body without Organs, Becomings, and Lines of Flight are more-or-less reflected in many of the tools of Soft Systems Methodology (the mapping is not precise, but I think the same sort of ideas apply - root definitions and conceptual models of systems of purposeful activity may not be as evocative terms, but the basic idea of exploring and mapping out a complicated and uncertain world that is always in flux is there).

The things I find most interesting are the concepts of overcoding, territorialisation and deterritorialisation. This covers something not really addressed in SSM: the locus of control, and process by which we impose our will upon others. Hence I can territorialise language, or a computer system, or a social network, incorporating it into my designs for action. This has a lot of relevance to design and the dual nature of technical artefacts: you produce a design, but it is then territorialised by the users, who will incorporate into their own designs, possibly in ways you don't want or couldn't predict (using a bottle as a weapon, perhaps). It's interesting to reflect on the AHRC Tracking Network in this way, the idea of being tracked as being territorialised by another system. Rehabilitation likewise, perhaps: being territorialised by a medical system, or overcoded by illness or injury. Anyway: these are the ideas that most provoked new thoughts in me.

Armed with this understanding, we are now in a position to return to Margrit Shildrick's paper on Why Should the Body End at the Skin?

Why Should Bodies End at the Skin? Revisited

Seen through the lens of the above concepts, this becomes (what I assume is) a classic Deleuzian question. Not in a random, chaotic, or arbitrary way: "Hey! My body includes the pavement! I want everyone walking on it arrested for assault!" - the empty Body without Organs. Nor in a controlling way - "You're all wrong! Bodies don't end at the skin and anyone who thinks they do is an idiot!" - the cancerous Body without Organs. But taken as a healthy exploration of the Body without Organs, it becomes an inquiry: " Is there any value to viewing the body as extending beyond the skin?".

It's also worth noting that the terminology is unfortunate here. The Body without Organs has nothing to do with whether the human body does or doesn't have organs. It's a metaphor, so it might well be  better to use the term Plane of Consistency. Hence we can talk about the Plane of Consistency of the Body (rather than the Body without Organs of the Body!), representing all the feasible ways in which the body might be mapped. The enquiry takes the form not of looking for a single definition of The Body to supplant all others, but to see if alternative definitions are useful. In this case, the outcome is different "lenses" for different situations: sometimes useful to consider person plus assistive technology; sometimes not. This brings us neatly to the issue I was grappling with here.

And since that was more or less where we came in, it seems as a good a place as any to stop.

References

Bucciarelli LL (1994) Designing Engineers, MIT Press: Cambridge, MA 

Checkland, P. and A. Casar (1986). ‘Vickers’ concept of an appreciative system: a systemic account. Journal of Applied Systems Analysis 13, 109–115.