{
"$type": "site.standard.document",
"content": "---\ntitle: \"Cybernetic futures explained (maybe)\"\ndescription: \"Futuring works best when it's more like a radar than a lighthouse---a sense-analyse-act feedback loop, not a fixed point to steer towards.\"\ntags:\n - cybernetics\n---\n\nimport YouTube from \"@/components/YouTube.astro\";\n\n<p class=\"post-subtitle\">Futuring as a radar, not a lighthouse.</p>\n\n:::info\n\nOne of my current projects at the ANU School of Cybernetics is to develop tools\n& procedures for futuring. This post is an attempt to get my head around how\nthese things fit together (spoiler: they do!).\n\n:::\n\nFutures/futuring[^terminology] is _a thing_---see\n[Smith and Ashby](https://www.howtofuture.com) for a practical guide or\n[Powers](https://www.press.uillinois.edu/books/?id=p084690) for a more critical\nhistory and review. It's the idea and practice of futuring as a verb, and this\nvideo from the [Institute for the Future](https://www.iftf.org) is a good\narticulation of the \"pitch\":\n\n[^terminology]:\n or foresight, or forecasting... there are a few terms which are used\n relatively interchangeably\n\n<YouTube id=\"5_EsLu4qydw\" />\n\nJust so I'm clear up-front: I think that futuring is a genuinely useful tool in\nthe toolbelt of any individual or organisation trying to figure out what success\nlooks like and how to achieve it.\n\nOne diagram which is often used in futuring is the\n[futures cone](https://thevoroscope.com/publications/foresight-primer/), which\nhelps visualise the relationship between the _now_ and the different potential\n_futures_ which might eventuate.\n\n\n\nThis diagram is just a visual aid---the future doesn't really exist as a series\nof concentric discs of soothing colours---but it helps to anchor discussions we\nmight have and predictions we might make about the future. In this sense, any\nprediction or \"vision\" of the future is a single **point** in the futures cone,\nand exactly where it falls in the space of possible, plausible, probable, or\npreferable futures is part of the discussion. Think about it as\n[monte carlo](https://en.wikipedia.org/wiki/Monte_Carlo_method) of the future:\n\n\n\nOne thing to note here is that futuring is not about predicting the future. In\nmany ways it makes one less certain about the future; futuring requires a\nhealthy dose of epistemic humility, but that's ok. It's hard to expect the\nunexpected and predict the unpredictable, so these sampled points (potential\nfutures) are really just a way of thinking about the current state of the world,\nand especially how we might act and position ourselves in the world now to make\nthe most of future opportunities and steer towards the more desirable potential\nfutures.\n\nThe other half of the story is cybernetics.\n[Stafford Beer](https://en.wikipedia.org/wiki/Stafford_Beer) tells a joke about\ndefining it in\n[an address he gave at the University of Valladolid](https://www.emerald.com/insight/content/doi/10.1108/03684920210417283/):\n\n> ...it concerns three men who are about to be executed. The prison governor\n> calls them to his office, and explains that each will be granted a last\n> request. The first one confesses that he has led a sinful life, and would like\n> to see a priest. The governor says he thinks he can arrange that. And the\n> second man? The second man explains that he is a professor of cybernetics. His\n> last request is to deliver a final and definitive answer to the question: what\n> is cybernetics? The governor accedes to this request also. And the third man?\n> Well, he is a doctoral student of the professor---his request is to be\n> executed second.\n\nIt's a great joke, grounded in a deep truth. One way to define cybernetic\nsystems is as systems with (i) a purpose/goal and (ii) a mechanism for steering\ntowards[^towards] said purpose. In some cases there's a defined end state, where\nupon attaining said purpose victory is declared and the job is done. However in\nmany cases what's desirable is\n[homeostasis](https://en.wikipedia.org/wiki/Homeostasis), i.e. a system which\ncan keep itself \"in its happy place\", stable and resistant to peturbations.\n\n[^towards]:\n The system's purpose can also be stated negatively, e.g. avoiding pain. So\n towards/away from are equally valid when talking about purpose.\n\n\n\nThere are lots of potential illustrations of this idea, but one that many of my\ncybernetic forebears liked is the one of using a lighthouse to steer a ship.\n[Here's](https://history-computer.com/the-complete-guide-to-cybernetics/) a nice\nexplanation:\n\n> In ancient Greece, the\n> [_Kubernetes_](https://en.wikipedia.org/wiki/Cybernetics#Etymology) >\n> [navigator/helmsperson] was in charge of controlling the Grecian longships.\n> The ships had to be steered through all kinds of unpredictable forces,\n> including wind, waves, storms, currents, and tides. The Greeks found that they\n> could ignore all of these and control the ship via a small tiller connected to\n> the ship's larger rudder just by pointing the tiller toward a fixed object in\n> the distance, such as a lighthouse, and making adjustments in real-time.\n\nThere's a\n[YouTube \"What is cybernetics?\"](https://www.youtube.com/watch?v=iXmlbd86YGA)\nvideo which includes a diagram like this:\n\n\n\nIt's not the particular position of the tiller at any one time that's important,\nis the way that the navigator watches the lighthouse and moves the tiller in\nresponse (as the ship is affected by currents & winds). If the ship's bow is\npointing to one side of the lighthouse, then adjust the tiller in the opposite\ndirection until it does. If you keep up that simple procedure, you'll get there\nin the end, and with your ship in one piece.\n\nIt's the navigator's continued monitoring of the difference between the\npurpose/goal (as indicated by the lighthouse) and the current state (as\nindicated by the where the bow is pointing) which matters. Once you know how to\nrespond to that difference, by steering in the opposite direction of that\ndifference, then you've got a simple and reliable procedure for successful\nsailing.\n\nHere's the key point: cybernetic systems don't work by planning out a complex,\ngo-to-whoa list of actions to take and then mindlessly following them. Instead\nthey use a simpler process involving _feedback_: do a thing,\nlooking/listening/sense what happened, compare the new state of the world with\nthe goal, and then do another thing... and so on[^all-encompassing].\n\n[^all-encompassing]:\n If that definition sounds all-encompassing, you're not the first person to\n notice that. Cybernetics\n [isn't shy](https://www.pangaro.com/cybernetics-the-center-of-sciences-future.html)\n about claiming all things as within its purview---Wiener was an ersatz\n theologian.\n\n\n\nHere's another example of a feedback-powered system. The development of radar in\nWWII was deeply connected to the birth of cybernetics (as detailed by Thomas Rid\nin [Rise of the Machines](https://wwnorton.com/books/Rise-of-the-Machines/)\nChapter 1). Take it away, [Wikipedia](https://en.wikipedia.org/wiki/Radar):\n\n> A radar system consists of a\n> [transmitter](https://en.wikipedia.org/wiki/Transmitter \"Transmitter\")\n> producing\n> [electromagnetic waves](https://en.wikipedia.org/wiki/Electromagnetic_wave \"Electromagnetic\nwave\")\n> in the [radio](https://en.wikipedia.org/wiki/Radio_spectrum \"Radio\nspectrum\")\n> or [microwaves](https://en.wikipedia.org/wiki/Microwave \"Microwave\") domain, a\n> transmitting\n> [antenna](<https://en.wikipedia.org/wiki/Antenna_(radio)> \"Antenna (radio)\"),\n> a receiving antenna (often the same antenna is used for transmitting and\n> receiving) and a\n> [receiver](https://en.wikipedia.org/wiki/Radio_receiver \"Radio receiver\") and\n> [processor](https://en.wikipedia.org/wiki/Data_processing_system \"Data\nprocessing system\")\n> to determine properties of the object(s). Radio waves (pulsed or continuous)\n> from the transmitter reflect off the object and return to the receiver, giving\n> information about the object's location and speed.\n\nThe radar sends out the pulses, which bounce (reflect) off the environment---and\nthese these reflections are sufficient (with some\n[tricky maths](https://nato-us.org/analysis2000/papers/moran.pdf)) to figure out\nwhat the environment looks like. Often there's some sort of visual\nrepresentation of the results, like the classic \"beeping dots on concentric\ncircles\" radar sweep interface you'll know from the movies.\n\nThe radar example is different from the ship steering one in that lighthouses\ndon't really move/change (although the currents in the water & other\nenvironmental factors do... hence the need for the feedback-powered steering\nprocedure). Using the radar properly requires sending an ongoing series of\npulses, because they each give an indication of what the environment looked like\nwhen the pulses were reflected, but to track moving objects you need to monitor\nthe dots over time to see how the environment changes.\n\nBringing these two threads together: each future scenario you can imagine (regardless of where\nit falls in the futures cone) is a way of projecting a potential \"end state\".\nThat end state doesn't have to be desirable---many potential futures\naren't---but it gives a \"concrete\" thing against which to compare your current\nstate to inform your current actions. Even the more mundane acts of management\nand oversight---strategy, tactics, contingency planning---when done well they\nall involve articulating goals and thinking about ways to bring them about.\nThere's clearly an echo of the \"steering systems\" thing here.\n\nHowever, one pitfall of that sort of mental picture of the future is that\nfuturing looks like a three step process:\n\n1. come up with (sample) a bunch of potential futures from the futures cone (the\n points in the diagram above)\n2. pick one of the potential futures you like the most (from the _preferable_\n area of the futures cone)\n3. steer towards[^lighthouse-steering] it like a lighthouse (whatever that looks\n like)\n\n[^lighthouse-steering]:\n Yeah, I know, you don't steer towards the lighthouse exactly---they just\n [show you where the rocks and reefs are](https://adventure.howstuffworks.com/lighthouse.htm).\n Don't @ me.\n\n\n\nThat's an unhelpful picture of what futuring is because it implies that the\npotential future you're steering towards is solid & stationary, but that's just\nnot how potential futures work. Instead, I think that futuring works best when\nit's more like a radar:\n\n1. come up with (sample) a bunch of potential futures from the futures cone\n (send out radar pulses)\n2. look for their reflections in the present\n3. analyse these reflections to decide how to act\n4. goto step 1 (because things have now changed)\n\n\n\nAs well as giving a different mental model of what futuring is and isn't, I\nthink there are two implications of this switch in perspective.\n\n1. Just like a ship drifts on the currents, our orientation to these potential\n futures is constantly changing as we & others act in the present---we're\n agents; we have agency. So we need to constantly re-examine our orientation\n towards these multiple futures and reorient ourselves as a result. That's a\n \"radar-like\" model of futuring. Cybernetics doesn't provide a \"once you\n measure & model all the things you can predict the future\" silver bullet\n (although it's not like some folks haven't tried\n [and failed](https://eujournalfuturesresearch.springeropen.com/articles/10.1007/s40309-013-0029-y)),\n but rather it's a commitment to using a simple \"sense-analyze-act\" feedback\n loop to keep the system on track.\n\n1. Don't be fooled by the simplicity of the steering procedure in the\n ship/lighthouse example; the radar example shows that making sense of the\n feedback from the environment can require non-trivial analysis before it's\n useful.\n\nTo close, I want to stress that I'm not saying anything remotely new here about\nthe connection between futuring and cybernetics---they're very often seen &\ndiscussed together. William Gibson, award-winning sci-fi novelist and sometime\nfuturist was\n[steeped in cybernetic lore](https://www.nytimes.com/2007/08/19/magazine/19wwln-q4-t.html)\nwhen he\n[coined the term \"cyberspace\"](https://www.themarginalian.org/2014/08/26/how-william-gibson-coined-cyberspace/).\nMy boss\n[Genevieve Bell AO](https://cybernetics.anu.edu.au/people/genevieve-bell/),\ndirector of the [ANU School of Cybernetics](https://cybernetics.anu.edu.au), has\nthought deeply and written persuasively and generally projected big futuring\nenergy for pretty much her whole career.\n",
"createdAt": "2026-05-13T23:14:49.626Z",
"description": "Futuring works best when it's more like a radar than a lighthouse---a sense-analyse-act feedback loop, not a fixed point to steer towards.",
"path": "/blog/2022/01/20/cybernetic-futures-explained",
"publishedAt": "2022-01-20T00:00:00.000Z",
"site": "at://did:plc:tevykrhi4kibtsipzci76d76/site.standard.publication/self",
"tags": [
"cybernetics"
],
"textContent": "Futuring works best when it's more like a radar than a lighthouse---a sense-analyse-act feedback loop, not a fixed point to steer towards.",
"title": "Cybernetic futures explained (maybe)"
}