March 2008

Strange, almost eerie: a pundit is found actually pondering the possible policies of the presidential pretenders.  Paul Prugman, er, Krugman, compares the candidates’ ideas on the mortgage crisis:

 Krugman, an economist, spent much of 2000 hammering at the fiscal irresponsibility of George Bush; people really didn’t want to hear it, but in fact Bush is fiscally irresponsible.  So I’d take it as a serious concern that McCain sees the mortgage crisis– which was largely the result of an unregulated shadow banking system making irresponsible loans– and thinks the answer is deregulation.  If we’re going to spend billions of dollars of taxpayer money to keep the system from collapsing– and McCain is in favor of “preventing systemic risk”, i.e. helping big business though not mortgage holders– then we need to look more closely at these guys, not less.

 Daniel Gross at points out that McCain also wants $400 billion in new tax cuts, roughly the level of the current deficit.   

 McCain seems to be a pretty nice guy; but his positions are pretty much Bush again.  And Bush’s incompetence is only half the problem.

After writing the last post, I realized that I’d written a New Yorker style review: used the book as a resource to talk about the issues it raises, rather than the book.  So how’s the book itself?  Good, as you should expect from Poundstone.  He talks about a lot of specific elections, so there’s nothing abstract about possible bad results.  And the portraits of particular electoral thinkers are interesting.  Did you know Lewis Carroll worried quite a lot about voting systems?

 My one complaint is that he never actually goes through Arrow’s proof; he barely even explains what the trouble is.  If Gödel can be explained, surely Arrow can be.

If there’s a common thread to problems with voting systems, it’s lack of information.  The more information your vote gives, the better.  Plurality systems suck because they only include one bit, who you voted for; that tells us nothing about how strong your support is or how you feel about the other candidates.  

 Alternative systems all give more information.  (From this perspective, the acrimonious debates between supporters of different systems mostly come down to the lack of yet more information: what you think other voters are doing, and whether you’ve modified your vote in response to that.)

I just finished William Poundstone’s book, which sports the clanky title Gaming the Vote: Why Elections Aren’t Fair (and What We Can Do About It).  It’s about voting systems, a subject which, curiously, seems to engage only wonks.  Despite the object lesson of 2000, most Americans can hardly conceive that anything besides plurality voting exists, much less that it might be better.

 Till recently, game theorists were almost as pessimistic.  Kenneth Arrow’s 1948 paper seemed to show that no voting system was free of paradoxes.  Whatever you picked– plurality, single transferrable vote, Borda counts, Condorcet voting, instant-runoff voting, approval rating– you can construct an example where the wrong guy wins.  Plurality voting systems are prone to spoilers or clones; more sophisticated systems fall prey to strategic voting (where voters conceal their true preferences to give an advantage to their favorite); ranking systems in general fail if voter preferences aren’t transitive.

Poundstone unveils his own best candidate last (spoiler warning): range voting.  You very likely know it already: it’s the numerical ratings used at Hot or Not or rottentomatoes.  It bypasses Arrow’s paradox because it’s not a ranking system; it easily handles spoilers and strategic voting, and it’s simple to boot.

Eric Janszen, a former venture capitalist, has an entertaining article on how the American economy has developed to run on bubbles.  If you like breezy historical overviews– this one goes back to WWI– you’ll love it.

These days, as Janszen puts it, we’re all orthodox free marketers on the way up, but Keynesians on the way down.  That is, a modern bubble starts with deregulation (and Fed complicity) and earnest explanations of how the old rules don’t apply and the favored market will keep rising forever.  When it cracks, everyone wants the taxpayers to bail us out. 

The housing bubble 

Only we’re in a far worse state now than when the dot-com bubble burst: in contrast to the 6% Funds Rate, high dollar, high taxes, and government surpluses of 2000, we now have a 4.5% rate, a weak dollar, deficits, and insufficient taxes.

 So we need some kind of austerity program to deal with the loss of $12 trillion in fictitious value, or at the least a return to sensible regulation so financiers don’t keep messing up the economy?  Nah, we need a new bubble.  The best bubbles, Janszen says, are already in place, popular, and have a favorable tax treatment.  There’s only one industry that’s poised to take off: alternative energy.  Now you know where to put your millions.

Alert reader Jean-Sébastien Girard pointed me at the Trésor de la langue française, which supports the derivation from fehu, and usefully adds that the earliest citations for feudum don’t have the d: feus, feum— which makes it clear that the Latin words derive from fehu too.  The d is thought to have crept in from the semantically related alodum ‘property’. 

Oh lordy.  Googling alodum brought me to an 1850 law dictionary, which notes that “Blackstone considers it the same as odh al, in the Northern languages… with the syllables transposed, alodh; and hence signifying entire or absolute property.”  With the syllables transposed?  That’s getting into Edo Nyland territory.

Several of the articles I read today on feudalism (see the much more important article below) mention that Latin feudum became French fief.  OK, but huh

Larousse tells me that the Chanson de Roland (1080) has feu, fiet.  Fieu was also seen.  Fine.  Then fief in the 13C with an f by analogy with bief, juif, soif

OK, bief ‘canal’ is from bedum, though it first appears as biez; juif ‘Jew’ is a back-formation from feminine juive from jūdaea; soif ‘thirst’ is 12C (earlier sei, soi) from sitis.  If analogy is happening, it’s not clear what the analogy is to; this is a pretty miscellaneous set.  (If you’re thinking sound change– no; intervocalic t or d normally just disappears.)

 But soif has a note: “f due to the false analogy with words of the type buef, bœuf, pl. bues, or of the type nois (nom.) / noif (acc.), from nix; the form in –f took over because it avoided homonyms.”

Off to bœuf ‘bull, beef’well, finally we have a word with a right to its f, from Latin bovemNeif/noif ‘snow’ comes from nivem but got replaced by neige, a back-formation from neiger ‘to snow’, from nivicare.

So apparently, in the 12C or so, French found itself with a couple of words with a morphological alternation between s and f.  Plus, er, the word juif.  This still is not telling me why the perfectly good words biez, soi, fieu got –f added to them, both in writing and in speech.

Hmm, French Wikipedia suggests that Larousse is full of it, or perhaps fullf of itf.  It derives fief from fevum, possibly related to a Germanic word vieh ‘livestock’, possibly confused with fiscum ‘royal domain’.  A French site, CNRTL, is close to this, but gives the source as fehu, related to Dutch vee ‘cattle’.  Fine, except for the -f again.  CNRTL offers that this derives from the verb fiever, itself created out of fieu.

The OED is of little use, except to point out that the –d– in feudum is hard to explain too.  It does reference fiever though.

Ha!  Another online source mentions a certain Palgrave who held that feudum derives from Greek emphyteusis.  Sure, Pal.

What a mess.  There’s a lesson for conlangers here too… if you want a naturalistic derivation, mess up your words till at least some derivations seem completely insane.  Then do it some more.

This was a shocker.  In sedesdraconis‘s LJ I found a reference to an article by Melissa Snell which reveals that there’s no such thing as ‘feudalism’.

The usual understanding of feudalism is that it comprised a hierarchical system in which near-sovereign control over land was traded for military service… essentially a way of ordering society when strong central government was not possible.  This turns out to be not the case.

The medievals never talked about ‘feudalism’; the concept was the invention of 16C French and Italian scholars, attempting to understand a 12C text, the Libri Feudorum.  Unfortunately they bungled the job.  They imported contemporary notions into the document (especially the idea that ‘fiefs’ were lands held by nobles), and they mistook the earlier writers’ own speculation for fact. 

Their interpretation solidified into received wisdom, till it was blown apart by Susan Reynold’s 1994 book Fiefs and Vassals.  Snell’s review is rather short on what’s right if feudalism is wrong.  But these seem to be the main points.

  • There was an immense amount of variation.  The worst part about the ‘feudalism’ concept is the implied uniformity.  Land could be held in all sorts of ways, and people’s ideas of rights and property were different from ours.
  • The implied class structure– serfs, clerics, knights– is hopelessly simplistic.  The armed forces were by no means limited to fighting nobles or knights.
  • Most grants of land weren’t based on any agreement to provide military service, though they might be based on service already provided.  There was generally no idea that a grant could be revoked if the grantee broke an oath of fidelity (if it was even required).
  • Far from being lawless, medieval society expected general obedience to the king.
  • Serfs’ relationship to their lord- manorialism– was really a separate concept, not considered at the time a form of vassalage.

Some additional useful information and criticism are found in these two reviews of Reynolds’ book:

There’s a lesson here somewhere about the dissemination of new insights.  This might be a very academic dispute, but I find it remarkable that I’ve never run into it in 34 years… the time since Reynolds’ predecessor Elizabeth Brown published her critique.  You’d think someone would tell the public that the textbook understanding of the Middle Ages is just wrong.

 The important question is of course… what about the conworlders?  I welcome the opportunity to rethink some aspects of Almean history.  In general we should avoid not merely reproducing the fiefs-for-service idea, but exclusively European models.  Bernard Lewis’s From Babel to Dragomans, which I recently mentioned, has some good descriptions of Islamic models.  China looked pretty different too, not really having what we think of as an aristocracy.

Next Page »