The Early Days of a Better Nation

Friday, April 17, 2009

Existence Proof of Von Neumann Machine, Placing Imaginary Bets, and other Cultural Learnings

This year's Eastercon, LX 2009 was a blast. (For anyone who doesn't know: Eastercon is the largest annual British SF convention. A science fiction convention isn't - contrary to popular misconception - a gathering of people dressed as Klingons (not that there's anything wrong with that). It's a gathering of people dressed as Victorians. (Actually, that's an exaggeration too. A couple of nice photosets by Mark Bukumunhe and Ian Sales give a fair idea of what an Eastercon crowd looks like.))

The Night Sessions (just out in paperback, by the way) won the BSFA Award for best novel of 2008, and the awards are, as usual, works of art. I totally didn't expect this, given the strength of the list. I'd already read out the short-list and presented the Award for best non-fiction, and was delighted that it went to Farah Mendlesohn for her brilliant and original Rhetorics of Fantasy. Congratulations to all the winners.

Thanks to my cunning plan inveterate procrastination, I didn't register for the con until I arrived, so I wasn't on any other programme events. As a result I saw far more of the programme, and far more of the people there, than usual. Another advantage was that my badge had my name on both sides.

The Science Fiction Foundation's annual George Hay Lecture was presented Adrian Bowyer, on the RepRap machine, an attempt to build a useful self-replicating desktop factory. The lecture was perfectly pitched to a scientifictional audience (Bowyer had told Cory Doctorow: 'You write novels about what I'm doing!') and I was sold on the whole thing the moment I handled one of its products, a chunky and robust plastic coat-hook. RepRap ain't a Von Neumann machine just yet - it only makes all the plastic parts needed for another copy of itself - but the mad scientists are already working on a version that makes its own electrical circuits, and after that, who knows?

The very first BSFA lecture (the intended series will complement the Hay lectures by being on the humanities, rather than science) was given by historian Shana Worthen, on 'Visualising Time in the Middle Ages', and was about calendars and clocks. She began by wishing us a happy new year - one calendar, the Gallic, had the year starting, most inconveniently, on the moveable feast of Easter Sunday. Much that we take for granted had to be invented: the twenty-four hour day dates back to Babylon, but the length of an hour as measured by a sundial varied with the length of the day - and the first mechanical clocks tried to reproduce this. It's kind of heartening to know that the classic mistake of designing the solution before specifying the problem didn't start with IT departments.

Tim Powers was author Guest of Honour, and he was great. In his GoH interview, conducted by his bibliographer John Berlyne, Powers told lots of entertaining tales about his life and gave lots of advice about writing, including a method for finding or inventing weird events in history by looking for clues in biographies of historical characters: watch out for when the biographer says something like 'and then, inexplicably ...'; combine the subject's timetable with someone else's timetable or a known series of events (like Bordeaux good and bad years), spot any coincidences and ask 'Can this be coincidence?'; and always say 'Yes, but why did he really do that?' In short, write your own conspiracy theories! A consequence of doing this, Powers said, was that he could never take conspiracy theories seriously. It's like, hey man, making this stuff up is my day job. Go bother someone else.

In his talk on 'How to Plot a Novel', Powers gave a masterclass on the subject, which I can't begin to reproduce here (but may attempt some other time). One memorable tip was to start with 'placing imaginary bets' - to write down every idea that occurs to you, without editing (a process he said was inspired by a ridiculous system for winning at games of chance by first losing ten thousand dollars on imaginary bets, thus using up all your bad luck) and then critically examining the ideas: questioning them, reversing them, often enough rejecting them. As he pointed out, 'But I thought of it!' is not a good reason for keeping an idea.

Lots of other panels, mostly good. Kari ended one on 'Re-creating History' by revealing a closely-guarded secret of medieval historians: 'People knew how to hem.' Clute said the genres of the fantastic were born of ruins and futurity. At the panel on 'Alternate Socialist Britains' I kept my mouth shut, perhaps wisely. At 'SF as Protest Literature' artist GoH David Lloyd said that the Watchmen film was 'pretty good' and that he didn't mind surveillance cameras. On 'Bad Biology' Paul McAuley remarked that silicon-based life 'could be squishy'.

The panel convened by Farah Mendlesohn on 'Pacifism and Non-Violence in SF' benefited from being on a subject on which there is a manageably small amount of source material. The discussion led me to make one of my very few comments from the floor. A more articulate and argued version of that comment would be this:

We already know how to have peace over large areas of the Earth, and that is by having large states covering those areas. (The combat death rate for men of military age in typical stateless societies far exceeds that in inter-state wars, including world wars.) SF has in its default assumptions a way to get to peace without pacifism, and that is the World State. Even Starship Troopers gives this answer, just as much as Star Trek or anything by H. G. Wells, Isaac Asimov or Arthur C. Clarke. Heinlein's Federation is a World State, and (consequently) there is peace within the human species. It just has wars with aliens.

But there are no aliens. So we could have peace.

Labels: , ,


This duality of fixed hours and variable hours has come up in my current project for Steve Jackson Games: GURPS Low-Tech, a catalog of equipment available in various eras before 1720. In the section on time measuring devices, I had to discuss the two types of hours, and specify whether each device typically kept fixed or variable hours. Variable hours were the natural effect of measuring time by the sun and of sundials, but mechanical devices as early as the water clock or clepsydra were easier to make to keep fixed hours—and apparently some ancient Egyptian clepsydrae were built with several different scales for different months, to approximate the changing length of sundial hours.

I also ran across the amusing anecdote about an Athenian courtesan who was nicknamed "the clepsydra" because she kept one by her bed to time her clients' visits. It's curious to see a situation where "punching the time clock" (so to speak) is a device used by labor to control the demands of the employer. . . .

Anyway, it's an old problem in time measurement. The spread of mechanical clocks seemingly didn't so much inspire a new approach as change the relative ease of applying the two approaches.

Congratulations on winning the BSFA!

On states and violence, though, I've got to disagree -- I think it's confusing cause and effect.

States are a luxury good (well, a luxury bad from my point of view -- but a luxury commodity in any case); they fund themselves out of the social surplus. So a society needs to achieve a certain level of prosperity before it can have much in the way of a state; and it can't achieve that level of prosperity if it's racked by constant tribal warfare. So it's no surprise that the societies that are racked by tribal warfare tend to be the stateless ones -- but it's the violence that explains the statelessness, not vice versa. As Thomas Paine noted, states piggyback on autonomously arising social order and then claim to have created it.

I think this is because states are essentially parasitic and don't contribute to social order at all -- rather the contrary, when they arise they *hinder* the further advance of cooperation and economic development more than they help it. (Certainly when states are imposed, or attempted to be imposed, on violent tribal societies it tends to *exacerbate* the violence, since there's now a big gun in the room -- the state apparatus -- that each tribe needs to seize lest some other tribe seize it first.) But even if one thinks states are a good thing, they're still an expensive thing, and so require a *pre-existing* attainment of a fair degree of peaceful commerce and productivity before they can get going.

Moreover, when large states consolidate their power and displace a previous more decentralised and more peaceful state situation, the result is often genocide (as the history of the 20th century demonstrates). That's another reason for thinking that states are the effect rather than the cause of peace.

If some degree of peace and prosperity is needed to make states possible, then we're going to get misleading data when we compare economically undeveloped, culturally tribal, relatively stateless societies with economically advanced state-ridden societies; the latter will often be more peaceful, and so we'll be tempted to think that the state is what's making the difference, but that inference just doesn't follow.

Thus a more interesting comparison is to compare relatively stateless and relatively state-ridden society that are *otherwise at comparable levels of economic development and cultural mores*.

When we do that, I think we get a very different picture. Ben Powell's research, for example, shows that stateless Somalia, while undoubtedly a crappy place to live, has been both more peaceful and more prosperous than either its earlier stateless self or its economically and culturally comparable neighbours. I would also point to the research of Bruce Benson and David Friedman on how relatively stateless medieval Iceland and the relatively stateless American frontier were far less violent than comparable state-ridden societies of the time.

Oops -- in my last paragraph, "earlier stateless self" should be "earlier state-ridden self."

P.S. - A couple of links:

On Somalia's improvement under anarchy: medieval Iceland, the American frontier, and other cases of relatively orderly anarchy:

I want to correct a common bit of linguistic stretching concerning von Neumann and those machines named after him. The fact is this. Johnny (Janos, Johan) v, Neumann was thinking about computers and automata in general. At that time (c. 1950) he thought that artificial life was possible by using automata. Someone challenged him by claiming that no conceivable machine could crank out a reproduction of itself (a replica). Von Neumann thought about that and went to work. His posthumous MS shows how you can make such a device. It describes an automaton with less than 30 possible states to be in (I'm simplifying a bit). You give those states an initial condition, let it alone for a while while programmed state-transitions occur, and out comes the replica. A bit later mathematicians simplified his design by producing gizmos with fewer states that replicate. OK. Now note the word REPLICATE. That's ALL they CAN do. Indeed, SELF-REPLICATE. But today's SF has changed this meaning into one that denotes gadgets that can change just about anything into something else. So we have machines that eat up Jupiter---or something like that, for their own or their controllers' purposes. I PROTEST. This is NOT what von Neumann machines do. They do ONLY what v.N. proved they can do. Other machines might be able to change ME into Jupiter for all I know, but then we've left Von Neumann Janos very far behind. OK. That's an historical point about the correct use of words.

An alternative reversal of your logic to Roderick T. Long's would be that a World State might be created or might hold together because there are wars with aliens - in much the way that conservatives try to hold societies together by declaring "Wars on *".

Actually, George Berger, Von Neumann also showed that his self-reproducing automata were Turing-complete. Therefore, while they might not be able to turn anything into anything else or eat Jupiter, they could not only reproduce themselves, but they could also perform any computational algorithm.

a World State might be created or might hold together because there are wars with aliens

Actually that's sort of the ending of Watchmen ....

Before Watchmen it was a Theodore Sturgeon story, "Unite and Conquer," if I remember the title right.

In Minnesota the sci-fi convention is the largest BDSM event of the year, under the cover of being a Gor event.

Renegade eye, something like that may have been true at one time, but after the SF fans got tired of being sneered at by all the other 'alternative' types who turned up, Minicon went very deliberately back to being an SF convention. Or so I was told when I went there a few years ago.

Re the state stuff: the first part of my argument, about state versus non-state levels of violence, was going by my memory of a point made in, I think, one of Steve Pinker's books, a point which surprised and disturbed me. The point was that in tribal hunter-gatherer societies (which is what I meant by 'typical stateless societies' because historically and prehistorically they are typical), even ones that anthropologists and other observers had often regarded as pretty idyllic, the actual levels of combat deaths - while small in absolute terms - were proportionately very high. This only became apparent when the numbers were crunched.

The second part, about the World State, is independent of this point (on which I'd be glad to be shown wrong). My point there was that the World State just is the default SF solution to war within the human species, even in works that seem light-years away from that nostrum, such as Starship Troopers.

Thanks Edward, I didn't know that. Also, "george" is good enough. I only use my full name so that friends and colleagues will know that it's me at work or play.

So... where is the evidence that stateless societies are more violent? Where are MacLeod's examples? Is it early Iceland, which was far less violent than its contemporaries, the Wild West, where violence went UP after the establishment of civilization, or Somalia, which has been improving economically ever since they became Stateless?

I have read David Friedman's account of early Iceland, but I have also read William Ian Miller's much longer and more detailed account. From the latter, it does not appear that Iceland was stateless.

The Icelandic thing did not have a monopoly of force; Iceland did not have a big enough economy to support a division of labor that would have let anyone specialize in force full time, and retaliatory force was largely self-help and mutual aid. Nor did it have a monopoly of arbitration; any two people could select any third person to mediate a dispute, or one disputant could grant the other "self-judgment." But Iceland also had compulsory process, in which someone could be ordered to show up and face legal charges, or face outlawry, which stripped him of all legal rights. And the thing had a monopoly of compulsory process. It also had a monopoly of the ability to outlaw people, and a monopoly of the ability to define the law of the land. It even had one paid government official, the Lawspeaker, whose job was to recite the law of the land over the course of three annual sessions of the thing. All those functions were exclusive.

It would be fair to describe Iceland as an ultraminimal state. But it did have the exclusive ability to unleash legitimate retaliatory force against an offender, and to compel an accused person to attend the annual meeting and answer legal charges. And compulsory legal process appears to be the irreducible nub of state authority.

Uh, that's not an answer to my question. You're digressing.

François: No, of course it's not an answer to your question; it wasn't meant to be. It's a rejection of one of your proposed examples of a stateless society, on the ground that it's not truly stateless.

I reject your standard as being too stringent.

Francois Tremblay, we have little direct evidence about stateless societies because of an observer effect: records were generally made of, say, African or Irish tribal society by colonialists who were bringing in a state and who disrupted records and traditions left by those. Furthermore, when they tapped into local accounts of the times immediately preceding, they weren't getting accounts of untouched societies but of societies that were heavily disrupted by the precursors of civilisation, e.g. slavers disrupting everything they left behind or traders selling guns which let some local groups become dominant - de facto or de jure states (as in Hawaii, Madagascar or among the Maori, as when some of those wiped out or forcibly assimilated the Chatham Islanders).

Still, we can get some surrogates for what happened by looking at, say, Scottish and Irish history, or Arab culture in which polygamous practice only made sense when many men died before marrying, in Razzias and the like. Similar things apply to, say, Zulu and Danakil traditions of who was qualified to marry by success in war.

Certainly, these are also not pure cases, rather the emergent unsettled patterns arising from unresolved tensions between clan, feudal and/or other systems. But there was a word like cateran and traditions associated with that, even before Gilderoy MacGregor had his hash settled by, basically, a posse from a town. And Irish oral traditions like "near every man's house there grows a thorn bush", as well as the obvious botanical-geographical meaning, carried down esoteric meanings - that you were always near the uncanny symbolised by the bush, and that the makings of a shillelagh were always to hand (a tradition my grandfather made use of during the troubled times after the Liberation of France; when my mother returned from the wars he greeted her at the door with a shillelagh). And the Australian aborigines had a kind of boomerang whose only use was on men, which can be confirmed for times past by archaeology...

To sum up, we do know that endemic violence did occur in stateless societies. We also know that it was in some sense an exception, or it would not have been survivable; but the exception needed to occur often enough that its presence worked as a threat and warning to prevent further cases. Also, it was kept exceptional mainly by confining it to an exception group - young men - for whom it was more of a norm, so allowing the main system to thrive largely unaffected. (And, of course, there is a thing about young men that they do indeed accept personal risks; to thwart them can itself be a poor trade off by their standards, just as Geordie fishermen were annoyed when their wives agitated for greater safety standards.)

This is the main difference. Endemic violence, while still full of personal tragedy, never reached the scale or systemic threat that war or harmful "peace" does, with no pacifications like the Wasting of the North that halved England's population after the Norman Conquest, no peaceful elimination of Jews from Nazi pacified territories and - dare I say it - no ever-shrinking place for Palestinians today (though even the stateless Caribs of the West Indies or the stateless Jews under the Judges could achieve this, it could never happen on an industrial scale as it were). The "peace" of the state can always include what Tacitus reports the Pict Calgacus saying of Pax Romana, "ubi solitudinem faciunt, pacem appellant", where they make a desert they call it peace. (We should also note, peace is not necessarily a good thing in either absolute or relative terms - it depends what it involves and facilitates.)

On the whole, casualties and all, a workable stateless society is better than pacification by a state if for no other reason than the smaller down side risk of huge occasional disasters - epidemic rather than endemic violence, often left out of the calculations on the spurious grounds that it is not how a state is supposed to work, even though, observably, from time to time it does.

No, the main problems history shows with stateless societies is their internal and external weakness. Their structures can be overridden or even co-opted by internal or external would-be creators of states. To make them workable in the face of what we might call infection would need an analogue of an immune system, institutions within the whole to face and undercut threats without themselves becoming what is feared (think the kings of ancient Israel), cures worse than or tantamount to the disease.

I agree that Iceland wasn't perfectly stateless (though it was much closer to statelessness than to the contemporary conception of a state). But I think when you talk about the thing having a monopoly you mean the Althing. Because things precisely didn't have a monopoly; you could change your thing membership (and thus your godhi -- who was both the provider of security services and your representative at the Althing) without changing your geographical residence. The Althing (an assembly of all the things) was a monopoly (though seats on the Althing were private property and could be bought and sold), but it had no powers of enforcement; all enforcement took place at the competitive level of the godhi.

(I also disagree that there were no specialists in security provision; it seems to me that that's precisely what a godhi was.)

The real monopoly element in Icelandic law, and the one that eventually doomed it, came with the introduction of Christianity, when churchsteads (and the associated fees) were made territorial monopolies, thus allowing their possessors (who were also godhar) to reap monopoly rents and so accumulate power.

On this see Birgir Solvason's dissertation Ordered Anarchy, State, and Rent-Seeking: The Icelandic Commonwealth, 930-1262.

Francois, you quite rightly ask for evidence. Thanks to a commenter on Roderick T. Long's blog, and a few clicks, I've tracked down what seems to be the strongest current source: War before Civilization: The Myth of the Peaceful Savage by Lawrence H. Keeley. Keeley's book came out after Steven Pinker's How the Mind Works (which was, I think, where I originally came across this point, though I no longer have a copy of that book to hand) so I guess Pinker used the same or similar original sources in the anthopology literature.

From reviews of Keeley's book, it seems I understated the case by confining it to men of military age: the toll of primitive warfare on women and children is also proportionately worse than in modern warfare.

None of this means, of course, that peaceful stateless societies haven't existed, or couldn't exist in the future.

To go back to the SF context, I'm reminded of the immortal words of that great philosopher, Jack Handy: 'I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it.'

I don't believe that one second. In our modern warfare, civilians represent 90% of casualties. There's no way you can prove to me that it was worse in primitive times.

It's not the proportion of civilians in the casualties, but the proportion of the civilian casualties to the population, that tends to be higher in primitive warfare, according to reviews of Keeley's book. Which makes sense: few (if any) inter-state conflicts have led to the population of the losing side's being wiped out to the last man, woman, and child, whereas plenty of inter-'tribal' conflicts have.

I don't deny that primitive societies tend (with perhaps some exceptions) to be fairly violent. But the equation of primitive societies with stateless ones is problematic, since the examples of relatively orderly stateless societies I cited are not primitive. Hence my point -- that it's the primitiveness, not the statelessness, that cause the violence.

As for vulnerability -- nearly stateless Ireland fended off the English for over a thousand years.

Aside: I should have written "near every man's door there grows a thorn bush".

Ken, I found the higher endemic level of primitive violence utterly unsurprising, but the idea of higher levels for others is surprising, if nothing else because the young men formed a protective barrier and the others were a prize rather than a target. Certainly African tribes carried off women and children into domestic slavery to strengthen their own tribes. Perhaps you are classifying these as victims of violence, as though they were harmed by it violently as opposed to suffering reduced status? That is, they certainly suffered but they were not casualties of violence in the sense of dead and injured. I'm not trying to belittle it but to categorise it.

I seriously doubt that extirpation was more common at the hands of stateless primitives than of states. Many examples of the latter come to mind: the many groups that dwelled in the Crimea at one time or another (Scythians, Greeks, Goths, possibly even the Tartars); Moors/Arabs and Jews in Spain; Carthaginians, Corinthians and many others like those Celts who entered the Balkans, at the hands of Rome (and even some other Celtic tribes); Vandals and other Goths (and almost the Bulgars) at the hands of Byzantium; the Pelasgoi and even some Greeks (e.g. Malians) at the hands of Greeks; many groups at the hands of Assyria, of whom we only remember the lost tribes of Israel... The list goes on. Quite simply, primitives could not often be that thorough without state machinery, and we usually only see such thorough cases driven by tribal pressures when those take place within such machinery; the only exceptions I know are the Caribs and ancient Israel, and possibly the Maori tradition about the Moriori (which is contentious, and arguably their later genocide of the Chatham Islanders was via a nascent Maori state).

Roderick, I must take issue with you about your definitions in "nearly stateless Ireland fended off the English for over a thousand years". That simply isn't true; what happened was, the Irish could resist the deeper penetrations but never managed to fend the Norman then English presence off, merely managing to confine it to certain areas at times - often quite long ones. And this long retreat only lasted some six centuries, not ten.

Roderick, it seems that we are in violent agreement, as folks used to say on Usenet.

Peter: the extirpations Keeley is talking about are of small groups, so it's not like they were technically difficult. I'm not sure how many of your examples of historical extirpations by states are actual extirpations or forced assimilation or dispersal. And I'm sure we can remove ancient Canaan from the list of stateless massacres! There's no strong archaeological evidence that any Israelite conquest and extermination actually happened.

Sorry to bring the tone down but Ken, have you ever worn a Klingon costume?

No, AVPS, I've never worn a Klingon costume, or any other skiffy costume.

One good example of yr non-statists resisting a state for some time is the Mapuche resistance to conquest by the Spaniards and later the Chileans.

I can't think of any others off-hand which weren't very powerfully aided by tropical diseases, though, so the general point might stand.

Chris Williams

"And I'm sure we can remove ancient Canaan from the list of stateless massacres! There's no strong archaeological evidence that any Israelite conquest and extermination actually happened."

I wasn't clear enough. I didn't intend to suggest that the whole thing was one genocide, rather I was thinking of the individual episodes that went on within that larger account. I'm pretty sure that some campaigns wiped out single tribes.

But that's the thing about "the extirpations Keeley is talking about are of small groups, so it's not like they were technically difficult". Precisely because these were usually conflicts between small groups, they usually were technically difficult. The difficulty came from not being able to get every last one, so that none escaped. Sometimes they could manage it, e.g. by getting them "while they were still sore".

On the rest, "I'm not sure how many of your examples of historical extirpations by states are actual extirpations or forced assimilation or dispersal", well, that's why categorisation is important.

I would draw the line where the group vanished as a result, so Carthage and Corinth were wiped out even though - technically - only the men were killed at the time but the rest were dispersed by being carried off into captivity in which many may have died but the rest may have had assimilated descendants. That would really only take the Spanish cases off the list, because there were places for the survivors of those to go and regroup. The Celts that entered the Balkans were subjugated and their survivors were forcibly resettled in Asia Minor, in a defence zone cut off from other Celts, where they still were centuries later although largely assimilated (as in the Epistle to the Galatians). But this fate guaranteed that they would eventually vanish from a combination of assimilation and difficulty replacing their numbers in local conditions, just like the Nazi plan for settling Jews in Madagascar. (For the Galatians, the local difficulties were raids in force from the east, for Madagascar tropical diseases.)

I think my conclusion still stands, that these things were more practical and more common at the hands of states than of stateless groups, though cases of each can be found.

Charles Johnson has an interesting blog post about our discussion here.

Ken says: I've never worn a Klingon costume, or any other skiffy costumeI beg to differ. I have reason to believe that Ken MacLeod constantly goes around dressed as Ken MacLeod, the famous science fiction writer.

The ST state seemed to be larger than a world. Single state may be a good description of the default solution. The single state needs to be one that regards itself as a whole, not one that seeks distinctions among its population, and kills them.

Speaking of Tim Powers contriving conspiracies from the coincidences that are more common than people suppose, this TED talk does the same thing to the phrase "4 in the morning", weaving a sinister conspiracy out of the times and places the phrase appears.

Chris, I gather Keeley is well impressed with the resistance capacity of tribal societies, and argues that they have been underestimated and patronised despite giving would-be conquerors decades of grief. Ten thousand pounds of education/felled by a ten-rupee jezail, sort of thing.

That's actually "Two thousand pounds of education Drops to a ten-rupee jezail" [emphasis added], according to Arithmetic on the Frontier - there's inflation for you. The rest of it is well worth digesting, too.

Thanks for posting about this, I would love to read more about this topic.
halı yıkama makinası -

I need them all. I'm totally in love with the fabulous sporty shoes like Hogan
. The house of Hogan scarpe donna
is renowned for its casual-chic sneakers for him and her. You can look for hogan donna
. Even if you always choose the proper shoes in town, you can pick Hogan scarpe uomo
along with give it a try.

Post a Comment