Above I claimed that technologies don’t go extinct. By that I mean that no species, no specific invention, has been eliminated from the world at a global level. Somewhere, someone is still making that technology brand new today. Anything I could think of.

I repeated that claim in What Technology Wants, but science journalist Robert Krulwich was not buying it. Krulwich challenged me in public, on the radio, that he could easily find extinct technologies. He crafted a very fun NPR segment (worth listening to), where he sets out the challenge. Krulwich admits

I tried carbon paper (still being made), steam powered car engine parts (still being made), Paleolithic hammers (still being made), 6 pages of agricultural tools from an 1895 Montgomery Ward & Co. Catalogue (every one of them still being made), and to my utter astonishment, I couldn’t find a provable example of an technology that has disappeared completely.

Krulwich then crowdsourced the problem and asks NPR listeners to suggest technologies that have gone extinct globally. Lots of listeners responded and the NPR kept a log of the suggested extinct technologies. They ranged from obviously wrong candidates such as anvils and telegraphs, to more promising nominations, like shoe X-ray machines, whalebone corset stays, or the solar bath (as depicted below by NPR’s Benjamin Arthur).

From all the candidates, Krulwich selected three that they submitted to me as very likely to be extinct: The Roman Corvus – a pivoting bridge with a spike dropped on enemy ships. Ferrite Memory Cores – found in early jukeboxes and primitive computers. Radium Suppositories – radioactive sticks you insert into your anus.

I supplied counter examples of ways in which I felt these species are still alive, still being used. I found radioactive therapy devices for the rectum, and gang planks for boats, and ferrite cores being made for non-computer use.

It may be a game of semantics whether these are the same species, but you can judge for yourself in Krulwich’s great summary of our wager. In the final episode he gracefully conceded: “consider me beaten. To a degree I didn’t appreciate until Kevin forced me to look — technology does indeed persist. Tools, machines, they change, they adapt, they morph, but they continue to be made. I hadn’t noticed this tenaciousness before.” This story of technology’s persistence is told in full by a book that was published after I finished my book (and so it was not included in my bibliography) – The Shock of the Old. Author David Edgerton makes the important point that old technologies don’t die; they continue to influence us, and often to shape our lives more than the new does. Just consider concrete, slaughtering tools, sewing machines and so on. There is more old technology around they we admit, and it is more powerful than we recognize.

If you have not seen Kelly’s talks check out the videos below.

This year one of the more thought provoking thought experiments to appear in recent memory has its tenth anniversary.  Nick Bostrom’s paper in thePhilosophical Quarterly “Are You Living in a Simulation?”” might have sounded like the types of conversations we all had after leaving the theater having seen The Matrix, but Bostrom’s attempt was serious. (There is a great recent video of Bostrom discussing his argument at the IEET). What he did in his paper was create a formal argument around the seemingly fanciful question of whether or not we were living in a simulated world. Here is how he stated it…

 

This paper argues that at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a “posthuman” stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation.

To state his case in everyday language: any technological civilization whose progress did not stop at some point would gain the capacity to realistically simulate whole worlds, including the individual minds that inhabit them, that is, they could run realistic ancestor simulations. Technologically advanced civilizations either die out before gaining the capacity to produce realistic ancestor simulations,  there is  something stopping such technologically mature civilizations from running such simulations in large numbers or, we ourselves are most likely living in such a simulation because there are a great many more such simulated worlds than real ones.

There is a lot in that argument to digest and a number of underlying assumptions that might be explored or challenged, but I want to look at just one of them, #2. That is, I will make the case that there may be very good reasons why technological civilizations both prohibit and are largely uninterested in creating realistic ancestor simulations. Reasons that are both ethical and scientific. Bostrom himself discusses the possibility that ethical constraints might prevent  technologically mature civilizations from creating realistic ancestor simulations. He writes:

One can speculate that advanced civilizations all develop along a trajectory that leads to the recognition of an ethical prohibition against running ancestor-simulations because of the suffering that is inflicted on the inhabitants of the simulation. However, from our present point of view, it is not clear that creating a human race is immoral. On the contrary, we tend to view the existence of our race as constituting a great ethical value. Moreover, convergence on an ethical view of the immorality of running ancestor-simulations is not enough: it must be combined with convergence on a civilization-wide social structure that enables activities considered immoral to be effectively banned.

I think the issue of “suffering that is inflicted on the inhabitants of a simulation” may be more serious than Bostrom appears to believe. Any civilization that has reached the stage where it can create realistic worlds that contain fully conscious human beings will have almost definitely escaped the two conditions that haunt the human condition in its current form- namely pain and death. The creation of realistic ancestor simulations will have brought back into existence these two horrors and thus might likely be considered not merely unethical but perhaps even evil. Were our world actually such a simulation it would confront us with questions that once went by the name of theodicy, namely, the attempt to reconcile the assumed goodness of the creator (for our case the simulator)  with the existence of evil: natural, moral, and metaphysical that exists in the world.

Questions as to why there is, or the morality of there being, such a wide disjunction between the conditions of any imaginable creator/simulator and those of the created/simulated is a road we’ve been down before as the philosopher Susan Neiman so brilliantly showed us in her Evil in Modern Thought: An Alternative History of Philosophy.  There, Neiman points out that the question of theodicy is a nearly invisible current that swept through the modern age. As a serious topic of thought in this period it began as an underlying assumption behind the scientific revolution with thinkers such as Leibniz arguing in his Essays on the Goodness of God, the Freedom of Man and the Origin of Evil that “ours is the best of all possible worlds”. Leibniz having invented calculus and been the first to envision what we would understand as a modern computer was no dope, so one might wonder how such a genius could ever believe in anything so seemingly contradictory to actual human experience?

What one needs to remember in order to understand this is that the early giants of the scientific revolution, Newton, Descartes, Leibniz, Bacon weren’t out to replace God but to understand what was thought of as his natural order. Given how stunning and quick what seemed at the time a complete understanding of the natural world had been formulated using the new methods it perhaps made sense to think that a similar human understanding of the moral order was close at hand. Given the intricate and harmonious picture of how nature was “designed” it was easy to think that this natural order did not somehow run in violation of the moral order. That underneath every seemingly senseless and death-bringing natural event- an earthquake or plague- there was some deeper and more necessary process for the good of humanity going on.

The quest for theodicy continued when Rousseau gave us the idea that for harmony to be restored to the human world we needed to return to the balance found in nature, something we had lost when we developed civilization.  Nature, and therefore the intelligence that was thought to have designed it were good. Human beings got themselves into trouble when they failed to heed this nature- built their cities on fault lines, or even, for Rousseau, lived in cities at all.

Anyone who has ever taken a serious look at history (or even paid real attention to the news) would agree with Hegel that: “History… is, indeed, little more than the register of the ‘crimes, follies, and misfortunes’ of mankind”. There isn’t any room for an ethical creator there, but Hegel himself tried to find some larger meaning in the long term trajectory of history. With him, and Marx who followed on his heels we have not so much a creator, but a natural and historical process that leads to a fulfillment an intelligence or perfect society at its end. There may be a lot of blood and gore, a huge amount of seemingly unnecessary human suffering between the beginning of history and its end, but the price is worth paying.

Lest anyone think that these views are irrelevant in our current context, one can see a great deal of Rousseau in the positions of contemporary environmentalists and bio-conservatives who take the position that it is us, that is human beings and the artificial technological society we have created that is the problem. Likewise, the views of both singularitarians and some transhumanists who think we are on the verge of reaching some breakout stage where we complete or transcend the human condition, have deep echoes of both Hegel and Marx.

But perhaps the best analog for the kinds of rules a technologically advanced civilization might create around the issue of realistic ancestor simulations might lie not in these religiously based philosophical ideas but  in our regulations regarding more practical areas such as animal experimentation. Scientific researchers have long recognized that the pursuit of truth needs humane constraints and that such constraints apply not just to human research subjects, but to animal subjects as well. In the United States, standards regarding research that use live animals is subject to self-regulation and oversight based on those established by the Institutional Animal Care and Use Committee (IACUC). Those criteria are as follows:

The basic criteria for IACUC approval are that the research (a) has the potential to allow us to learn new information, (b) will teach skills or concepts that cannot be obtained using an alternative, (c) will generate knowledge that is scientifically and socially important, and (d) is designed such that animals are treated humanely.

The underlying idea behind these regulations is that researchers should never unnecessarily burden animals in research. Therefore, it is the job of researchers to design and carry out research in a way that does not subject animals to unnecessary burdens. Doing research that does not promise to generate important knowledge, subjecting animals to unnecessary pain, doing experiments on animals when the objectives can be reached without doing so are all ways of unnecessarily burdening animals.

Would realistic ancestor simulations meet these criteria? I think realistic ancestor simulations would probably fulfill criteria (a), but I have serious doubts that it meets any of the others. In terms of simulations offering us “skills or concepts that cannot be obtained using an alternative” (b), in what sense would a realistic simulations teach skills or concepts that could not be achieved using something short of such simulations where those within them actually suffer and die? Are there not many types of simulations that would fall short of the full range of subjective human experiences of suffering that would nevertheless grant us a good deal of knowledge regarding such types of worlds and societies?  There is probably an even greater ethical hurdle for realistic ancestor simulations to cross in (c), for it is difficult to see exactly what lessons or essential knowledge running such simulations would bring. Societies that are advanced enough to run such simulations are unlikely to gain vital information about how to run their own societies.

The knowledge they are seeking is bound to be historical in nature, that is, what was it like to live in such and such a period or what might have happened if some historical contingency were reversed? I find it extremely difficult to believe that we do not have the majority of information today to create realistic models of what it was like to live in a particular historical period, a recreation that does not have to entail real suffering on the part of innocent participants to be of worth.

Let’s take a specific rather than a general historical example dear to my heart because I am a Pennsylvanian- the Battle of Gettysburg. Imagine that you live hundreds or even thousands of years into the future when we are capable of creating realistic ancestor simulations. Imagine that you are absolutely fascinated by the American Civil War and would like to “live” in that period to see it for yourself. Certainly you might want to bring your own capacity for suffering and pain or to replicate this capacity in a “human” you, but why do the simulated beings in this world with you have to actually feel the lash of a whip, or the pain of a saw hacking off an injured limb? Again, if completely accurate simulations of limited historical events are ethically suspect, why would this suspicion not hold for the simulation of entire worlds?

One might raise the objection that any realistic ancestor simulation would need to possess sentient beings with free will such as ourselves that possess not merely the the ability to suffer, but to inflict evil upon others. This, of course, is exactly the argument Christian theodicy makes. It is also an argument that was undermined by sceptics of early modern theodicy- such as that put forth by Leibniz.

Arguments that the sentient beings we are now (whether simulated or real) require free will that puts us at risk of self-harm were dealt with brilliantly by the 17th century philosopher, Pierre Bayle, who compared whatever creator might exist behind a world where sentient beings are in constant danger due to their exercise of free will to a negligent parent. Every parent knows that giving their children freedom is necessary for moral growth, but what parent would give this freedom such reign when it was not only likely but known that it would lead to the severe harm or even death of their child?

Applying this directly to the simulation argument any creator of such simulations knows that we will kill millions of our fellow human beings in war and torture, deliberately starve and enslave countless others. At least these are problems we have brought on ourselves,but the world also contains numerous so-called natural evils such as earthquakes and pandemics which have devastated us numerous times. Above all, it contains death itself which will kill all of us in the end.

It was quite obvious to another sceptic, David Hume, that the reality we lived in had no concern for us. If it was “engineered” what did it say that the engineer did not provide obvious “tweaks” to the system that would make human life infinitely better? Why, are we driven by pain such as hunger and not correspondingly greater pleasure alone?

Arguments that human and animal suffering is somehow not “real” because it is simulated seem to me to be particularly tone deaf. In a lighter mood I might kick a stone like Samuel Johnson and squeak “I refute it thus!” In a darker mood I might ask those who hold the idea whether they would willingly exchange places with a burn victim. I think not!

If it is the case that we live in a realistic ancestor simulation then the simulator cares nothing for our suffering on a granular level. This leads us to the question of what type of knowledge could truly be gained from running such realistic ancestor simulations. It might be the case that the more granular a simulation is the less knowledge can actually be gained from it. If one needs to create entire worlds with individuals and everything in them in order to truly understand what is going on, then the results of the process you are studying is either contingent or you are really not in possession of full knowledge regarding how such worlds work. It might be interesting to run natural selection over again from the beginning of life on earth to see what alternatives evolution might have come up with, but would we really be gaining any knowledge about evolution itself rather than a view of how it might have looked had such and such initial conditions been changed? And why do we need to replicate an entire world including the pain suffered by the simulated creatures within before we can grasp what alternatives to the path evolution or even history followed might have looked like. Full understanding of the process by which evolution works, which we do not have but a civilization able to create realistic ancestor simulations doubtless would, should allow us to envision alternatives without having to run the process from scratch a countless number of times.

Yet, it is in the last criteria (d) that experiments are  “designed such that animals are treated humanely” that the ethical nature of any realistic ancestor simulation really hits a wall. If we are indeed living in an ancestor simulation it seems pretty clear that the simulator(s) should be declared inhumane. How otherwise would the simulator create a world of pandemics and genocides, torture, war, and murder, and above all, universal death?

One might claim that any simulator is so far above us that it takes little concern of our suffering. Yet, we have granted this kind of concern to animals and thus might consider ourselves morally superior to an ethically blind simulator. Without any concern or interest in our collective well-being, why simulate us in the first place?

Indeed, any simulator who created a world such as our own would be the ultimate anti-transhumanist. Having escaped pain and death “he” would bring them back into the world on account of what would likely be socially useless curiosity. Here then we might have an answer to the second half of Bostrom’s quote above:

Moreover, convergence on an ethical view of the immorality of running ancestor-simulations is not enough: it must be combined with convergence on a civilization-wide social structure that enables activities considered immoral to be effectively banned.

A society that was technologically capable of creating realistic ancestor simulations and actually runs them would appear to have on of two features (1) it finds such simulations ethically permissible, (2) it is unable to prevent such simulations from being created. Perhaps any society that remains open to creating such a degree of suffering found in realistic ancestor simulations  for anything but the reason of existential survival would be likely to destroy itself for other reasons relating to such lack of ethical boundaries.

However, it is (2) that I find the most illuminating. For perhaps the condition that decides if a civilization will continue to exist is its ability to adequately regulate the use of technology within it. Any society that is unable to prevent rogue members from creating realistic ancestor simulations despite deep ethical prohibitions is incapable of preventing the use of destructive technologies or in managing its own technological development in a way that promotes survival. A situation we can perhaps see glimpses of in our own situation related to nuclear and biological weapons, or the dangers of the Anthropocene.

This link between ethics, successful control of technology and long term survival is perhaps is the real lesson we should glean from Bostrom’s provocative simulation argument.

You are looking at, more or less, a portrait of the internet over an average 24 hours in 2012—higher usage in yellows and reds; lower in greens and blues—created by an anonymous researcher for the “Internet Census 2012″ project. How, exactly, do you gather this much data? Well: not legally, that’s for sure.

 

In order to track the geographical location and usage patterns of the internet, our researcher created a “botnet”—a network of nearly half a million hacked computers, chosen from a selection of Linux machines with no or default passwords, pinging everything they could and reporting back. The researcher says one of the chief concerns of the project was to “be nice”—”[W]e did not change passwords and did not make any permanent changes… We also uploaded a readme file containing a short explanation of the project as well as a contact email address”—but the botnet, dubbed “Carna,” was ultimately highly illegal.

It’s also not quite comprehensive. The computers being pinged to gather the data were limited to to IPv4 address space—leaving out the newer IPv6 protocol. But it still paints a extensive picture of the geographical internet. Here’s another map showing the location of reachable computers across the world:

The full report is here.

[Internet Census 2012 via Motherboard]