The latter half of the 20th century saw the built environment merged with media space, and architecture taking on new roles related to branding, image and consumerism. Augmented reality may recontextualise the functions of consumerism and architecture, and change in the way in which we operate within it.

A film produced for my final year Masters in Architecture, part of a larger project about the social and architectural consequences of new media and augmented reality.

First, here’s the original version of the video: a very ordinary first-person view of someone in the kitchen of a small, cramped flat (presumably in Britain, judging by the power-sockets), engaged in the mundane task of making a cup of tea. (The reasons for the strange hand-gestures will become apparent in the second video.)

domestic robocop: original footage from Keiichi Matsuda on Vimeo.

And here’s the same footage with an overdose of ‘augmented reality’ applied…

Augmented (hyper)Reality: Domestic Robocop from Keiichi Matsuda on Vimeo.

I love the detail and precision here: the clock and kettle-timer ticking away, the banal music, the equally-banal recipe and voice-over instructions. And the subtly ironic sense of humour, too: one of the messages (at 01:00) asks plaintively “Anyone up for a RL [real-life] meeting this weekend?”; the departure from the kitchen is signalled (at 01:30) by a red-flashing warning on the ‘Liquid waste’ bar on the personal-status indicator.

I can see that augmented-reality does have its value, but this is definitely not a future I’d like to live in!

As I read the article, my heart sank still further – though perhaps not for the reasons you might expect. Yes, the ‘bomb-detector’ has proved to be unreliable: there are huge problems on that score, without doubt. But to me the ‘betrayal’ turns out to be much more complex than it seems on the surface – because despite the ‘military-hardware’ packaging of the device in question, and its impressive-looking dials and cables and the rest, the underlying technology of the ‘bomb detector’ is a plain old ordinary everyday dowsing-rod.

Dowsing has been a serious interest of mine for several decades: over the years I’ve written what are now some of the best-known books on dowsing, in fact. Hence – unlike many of the critics – I do have some solid understanding of what’s going on in this case. And because of that longstanding background in the field, I’ll freely admit that I have few fundamental doubts about the use of dowsing in this context, not least because there’s plenty of long-documented, long-proven military practice in dowsing for land-mines and the like (contact the British Society of Dowsers for case-studies in Aden, for example, or the American Society of Dowsers for US use in Vietnam).  Like most people, I would much prefer a predictable and reliable machine to do the job, if there’s one available and it actually does work – which many don’t. But when lives are on the line and you don’t have anything else, a dowsing-rod in experienced hands can work wonders: so at least that part of this sad, messy story is no fraud. Yet that point about ‘experienced hands’ is extremely important: in unskilled hands a dowsing-rod can easily be worse than useless – as those on the receiving-end of those undetected explosives would have discovered to their cost…

(This is getting very long: better put a ‘Read more… link in here.)

Despite the protestations of pseudoscientist ‘skeptics’ like James Randi, the blunt fact is that dowsing works. Interestingly, how it works is almost irrelevant, though these days we do have a much better understanding of the psychology and physiology of what’s going on in dowsing – especially how the brain enacts pattern-processing against unconscious cues, much as in some aspects of proof-reading, for example, or ‘reading’ a stock-market ticker or a full-on fire in an apartment-block. But we also know that the instrument itself has very little impact on the quality of dowsing:  the relevant physics are so trivial that good dowsing-work can be done with a couple of bits of bent fence-wire or even a used tea-bag. The physiological constraints are also trivial – so trivial that just about anyone can do it one way or another if they put their mind to it. And the amount of knowledge needed to get started is trivial too – so trivial that most people can pick up the basics in a couple of minutes. But where so many people go so badly wrong with dowsing is that beyond that simple base, everything else depends on personal skill, on observation and self-observation, experience and interpretation – and none of that is trivial at all.

Every true skill depends on the development of judgement and awareness: my real professional interest is around identifying common-factors that apply in every skill, and using that knowledge to improve skills-education in any domain. In that sense, dowsing has been a very good test-case for that research – research which has been applied in a whole swathe of much more ‘conventional’ skills, from archaeology to enterprise-architecture, from software-development to quality-system design, and just about everything else in between. (Except sports, for some reason. I’ve never understood sports. I don’t know why, but there ’tis. )

Any skills-based technology will depend on the specific combination of the equipment and the operator: it needs to be understood as a interacting, interdependent system, not solely in terms of any of its individual components. The closer we get to a ‘pure’ skill, the more important the capabilities of the operator become – and dowsing is actually one of the closest examples we have to a ‘pure’ skill, because it consists of almost nothing but ‘judgement and awareness’. In practice, the choice and characteristics of a dowsing-instrument are often almost irrelevant, because the real ‘instrument’ you’re using in dowsing is you.

So yes, I do have some real concerns about the company selling a dowsing-based ‘bomb-detector’ for £15,000 each (or £45,000 each in Iraq, apparently), because in principle at least the job could have been done just as well with tuppence-worth of power-cable. (Belief and credibility do play an important part in dowsing-success, so there are some arguments for putting a serious price-tag on what is always going to be a very simple piece of kit – but for ethical reasons if nothing else, that price-tag should be measured in tens or low hundreds at most, not tens of thousands!)

Yet what worries me far more is the risk that this has been presented as a ‘deus ex machina’, inducing people to rely on the ‘machine’ itself (which, as the ‘skeptics’ correctly state, has just about zero capability on its own) rather than the personal skills of the operator (who can or may have the required capability, but only if the right skills-development has been applied). If the marketing-literature purports that the ‘bomb-detector’ itself does the task, it would be technological incompetence at best, because dowsing simply does not work that way – and the company should certainly have known this before they sold anything at all. If they knew, and went ahead anyway – especially without a rigorous focus on really solid skills-development – then it would indeed be fraud of a very serious kind. I would hold back any judgement on that until we’d had a chance to scrutinise the training-regime. I’d have to admit, though, that so far it doesn’t look good – which is unfortunate, to say the least.

Yet it’s important to tackle the right target here:  and I’m sure that in this case dowsing itself isn’t it. But you may think otherwise, of course – your comments, perhaps?

Update: 24 Jan

Seems there’s been quite a follow-up on this on the BBC. As usual they’ve tried to make sense of it in conventional ‘deus ex machina’ terms, with the obvious and correct conclusion that it doesn’t and can’t make sense in those terms: see the BBC article ‘Export ban on ‘useless’ detector‘. The ‘sensor card’ turns out to be a very ordinary RFID tag or some such, without any connection to any real electronics: there’s no possible way in which it can work in any conventional physical or chemical sense. But dowsers would recognise this straight away as what’s known as a ‘sample’ or ‘witness’: in effect, it’s best understood as a psychological trick to focus the operator’s mind on only the specified substance, in much the same way as the ‘cocktail party effect’, selecting out a single conversation at a very noisy party – though here the basic signal-to-noise ratio is vanishingly small, and needs to be enhanced in any way that we can. As with all dowsing, it’s based far more on psychology than physics, so attempting to assess in strictly physical terms not only makes no sense, but is literally unscientific. Worse, as with all skills – in fact, exactly as with operation of a conventional sonar/radar mine-detector – the ‘bomb-detector’ process needs to be understood as a complete system, the intersection of equipment and operator: but in this kind of analysis they’ve ignored that fundamental constraint, and instead tested the least-active part of the overall system, which again is flat-out unscientific – applying controls for parameters which are not even in play, and applying no controls for the parameters that do affect the system. Would be good if some of those self-styled ‘scientists’ had any real grasp of what true scientific investigation actually requires, but there ’tis…

Hence bleakly amusing to see the BBC’s evident surprise when the Iraqi Interior Ministry says that it does trust the devices: see ‘Iraqi Interior ministry still backing ‘bomb detector’‘. Part of that trust may come from the fact that they’ve so far spent a staggering $85m on the devices, of course. But at least that article does also make clear that a lot of emphasis has been placed on proper training: so actually there’s a fair chance that the device will work – in dowsing terms at least. And unlike the BBC, they do seem to be aware of the centrality of the operator in the system; but even so,the reliability will always be somewhat in question, especially for tired, scared, bored operators out on the street, day after day, dealing with an endless stream of vehicles and insults.

Judging from the descriptions, I really don’t think it’s fraud: at least, not in the sense of deliberately selling something that they know can’t work, though it may be fraud in other ways, of course. Only a proper investigation will be able to tell on that – and I don’t include the clumsy, myopic BBC hatchet-job as a ‘proper investigation’…

In many ways this whole sad mess is best understood as a clash of worldviews: the Western view, which focusses solely on the ‘objective’ world and tries to remove the person from the equation wherever possible; versus a more systemic worldview – pretty much any ‘non-Western’ worldview, in fact – which focusses far more on the person as an intrinsic and interdependent component of the ‘system’ in scope. The other deeper clash is between science-as-religion, which is obsessed with finding the Ultimate Answer to how everything ‘really works’, but will only accept answers within the confines a very rigid set of materialist assumptions; versus the much more practical technology view, which is less interested in ‘how things work’, and far more in ‘how things can be worked’ – which is not the same question.

The saddest part of this will be obvious to anyone who understands the psychology of skills-education, though will certainlynot be obvious to the BBC or the other ‘skeptic’ critics. This is that a very large part of this overall system depends on belief, and especially on the operator’s belief that the ‘sensing’ is separate from themselves, even though the operator is actually responsible for virtually all of the sensing and sense-interpretation – a complex double-bind described by the psychologist Kenneth Batcheldor as ‘ownership resistance’. The result is that if we ‘prove’ that the device ‘does not work’, in a suitably convincing manner, the belief in its efficacy will be destroyed, and hence the overall system will cease to work. Which would no doubt be taken as vindication of the aggressive ‘investigation’ – but in fact the investigation itself is the primary cause of failure. (Given his evident dislike of this class of technologies, Dave Snowden will no doubt not like me saying this, but this is in fact an exact illustration of his own dictum that “every diagnostic is also an intervention” – in this case, an inherently destructive intervention.) Which, in practice, could well leave Iraq without any form of mass-scale streetside bomb-detection – which would not be a good outcome… yet an outcome that arises directly from the so-called ‘scientific’ investigation.

I’ve no doubt that the BBC investigators were unaware of these complexities or their impacts: but there are real and very serious ethical issues here, which they appear to have failed to understand at all. Religious fanatics like James Randi are beyond the pale, and beyond reason, of course; but given that the only real ‘fraud’ here is the fundamentally non-scientific nature of their investigation, and that the destruction of a dowsing-based capability may well cost far more lives than the inevitable human-based limits on the system’s reliability, I would certainly caution the BBC and others to be a bit more cautious next time of the consequences of their no doubt well-meant but potentially lethal actions.

As David puts it, “WOW!!!” – very impressive indeed, and definitely reminiscent of the system shown in the sci-fi film Minority Report. What worries me, though, is that there’s a significant risk of a serious term-hijack here. As a “wearable gestural interface” to contextual information available via the net, ‘SixthSense’ is certainly an innovative form of augmented reality; but that’s all it is – it’s just clever technology, not ‘sixth sense’ in the traditional meaning of the term.

Probably the closest definition of ‘sixth sense’ would be “access to information which is not available directly via touch, taste, sight, sound or scent”. So any form of indirect sensing – such as plain old telephone or television, just as much as internet data-sources – is technically a kind of ‘sixth sense’. Another often-cited component is synaesthesia, any kind of cross-merging of the senses – so that aspect of the definition would apply to SixthSense too, because it cross-maps the indirect net-derived information with that arising from the immediate physical world. But not only is there a real danger of IT-centrism – where the technology becomes the sole centre of attention, ignoring the purpose for that merging of information – but we also risk assuming that we should constrain the meaning of ‘sixth sense’ to the available information solely to that which already exists in accessible form on the net. If we do the latter, without full awareness of doing so – in other words, if we fall for the implied term-hijack – we could entrap ourselves within three potentially lethal problems:

• we may shut out other information-sources, including possibly our own senses – “lost in cyberspace” etc
• we may limit ourselves only to what is already known – risking loss of insight or innovation
• we may be unable to test or verify the reliability or trustworthiness of the ‘augmented’ information-sources

From a human perspective, it’s essential not to limit our sources of information, because each can both provide unique information of its own, and also provide cross-checks against the sources, This is a key theme in enforcing transparency via the ‘social web’, for example. But it also brings us to the more traditional meaning of ‘sixth sense’, via the often strange concepts – or experiences, rather – such as psychometry, remote-viewing, telepathy, dowsing and the like. Generically these are often classified as ‘inituive skills’ – where the word ‘inituion’ literally translates as ‘teaching from within’. I’m well aware that self-styled Skeptics and other followers of the fundamentalist religion of ‘scientism‘ may have difficulty with any such notions, but as it happens, I’ve studied dowsing or ‘water-witching’ for several decades now: my first book, a kind of ‘teach-yourself guide’ nowadays known as The Diviner’s Handbook, was first published way back in 1976, and has been continuously in print ever since. This perhaps seem a bit of a surprise if you’ve only only known me as an enterprise-architect, but as far as dowsing is concerned, I’m generally regarded as one of the world experts in the field – particularly in its intersection of theory and practice as methodology. I do know what I’m talking about here: most self-styled Skeptics don’t. (At which point I’m reminded of Isaac Newton’s retort to astronomer Edmond Halley when the latter mocked his extensive writings on astrology: “I have studied the subject, sir, and you have not!” )

The point there is that in all of these intuitive-skills there’s a clear gradation from straightforward physical synaesthesia (one that for some people does quite literally resemble IT-based augmented-reality) all the way through to what we might describe as ‘good question…’; most people seem to make a big fuss about the ‘good question’ end of the scale, but in practice it’s the more ordinary world that is more important in most dowsing work, and, crucially, it is a learnable skill, dependent on much the same disciplines as for any other skill-based technology. (More info on that in my book Disciplines of Dowsing, co-authored with archaeologist/archaeographer Liz Poraj-Wilczynska.) Because this is technology, not science, it’s not ‘fraud’ for such intuitive information to come from any mixture of sources: it’s just information. (Though it’s often important to be able to identify which source the information arises from, so as to be able to verify the information-value – a theme we’ll return with the third point above.)

If we only limit ourselves to known sources of information, we’ll be unable to discover anything new. Often, for example, we’ll come across instances of a tactic I describe as “In order to remember something you never knew, first set out to forget it”. The mathematician Henri Poincaré provides one famous anecdote of this kind:

The circumstances of the journey made me forget my mathematical work; arrived at Coutances we boarded an omnibus … At the moment when I put my foot on the step the idea came to me, without anything in my former thoughts seeming to have paved the way for it, that the transformation that I had used to define the Fuchsian functions were identical with those of non-euclidean geometry. I did not verify this, I did not have the time for it, since scarcely had I sat down in the bus than I resumed the conversation already begun, but I was entirely certain at once.

This is a key theme in one of my favourite books, William Beveridge’s The Art of Scientific Investigation, which explores the use of chance, the use of intuition, the hazards and limitations of reason, and suchlike concerns in the process of scientific research. (Another example quoted in Beveridge’s book is Kekulé’s well-known story about how he discovered the ring-like structure of benzene: at the end, he urges his fellow-scientists, “Gentlemen, we must learn to dream!”) So the science of science itself is still something of a mystery: a century or more later, we still don’t know much about how these processes work, but we do have a much clearer understanding of how they can be worked – in other words, the technology and methodology, rather than the science. To quote Louis Pasteur, “In the field of scientific endeavour, chance favours the prepared mind”; yet if we arbitrarily constrain our sources of information, we’re limiting our chances. An open mind matters here – and ‘open’ in every sense, too.

Finally, by what means can we test and trust the information from these ‘augmented’ sources? Much of the self-styled ‘New Age’ teachings, for example, might perhaps be described not so much as ‘channelling’ as an open drain: no cross-checks of any kind, and far too often just ‘received truth’ for the gullible and self-deluded. But is much of what’s on the internet really any better? Google Maps’ interpretation of British post-codes is notoriously variable in its accuracy: I’ve sometimes found it to be half a mile or more off-target, especially in the smaller towns and villages. As a desmonstration of the technology’s potential, the SixthSense presentation was brilliant – but I really do have serious doubts as to how well it would work in practice. As with cloud-computing, Enterprise 2.0 and the rest of the current hype-wagons, there are some really serious questions about security and data-quality and the like that will need to be addressed before it could be trusted for use in any non-trivial real-world application. And as with other IT-hype term-hijacks, that’s exactly what usually doesn’t happen, because the hype is itself used to block out any visibility of those broader issues.

So yes, SixthSense is an excellent demonstration of net-based augmented-reality’s potential: but it’s important that we don’t let the hype and excitement block out the broader, richer, traditional meaning of ‘sixth sense’.