27 October 2019

Infinite village

Humans have a limited ability to maintain social connections; it seems to be about a hundred and fifty people in the abstract (Dunbar's Number) but being authoritarian -- defining and enforcing strict social roles, so you don't need to think much about your relationship with other people -- can make it easier to get the group size up, toward the cognitive limit.

Other people have since produced different estimates; various other people are pretty sure paleolithic human groups were nutritionally limited before they were cognitively limited.  You need agriculture to get the social grouping size large enough to run into the cognitive limits on band sizes.

So it's a cool factoid.  What makes it interesting in context is that you can look just a little more into Dunbar's work and find out that the cost got quantified in terms of social grooming time.  Maybe not literally removing bugs, but spending time maintaining the social connection. For the maxed-out group size, you get to about two-fifths of all your time being spent on (the technical sense of) grooming behaviours.

Social media is a social grooming amplifier; it creates a belief in connection.

It doesn't create the actual connection, or empathy, or any kind of detailed knowledge, but it does create a belief in group membership, emotional group connection, and a common narrative label for events.  Which means being in control of a social media platform gives you a ridiculous amount of power; you get to pick the questions.  (Maybe not the answers, strictly, but picking the questions is more than enough.)

19 October 2019

The label goes on the responsibility

So for just years now I've been very uneasy about the taxonomy approach to "we're implicitly enforcing a prescriptive norm and should stop" social problems, but all the explanations for why this isn't going to work (from the perspective of anybody who can't fake the norm) need a patient listener, twenty minutes, and a white board.

That doesn't much resemble helpful.

So -- when you say someone is disabled, or has special needs, or come up with a vast nuanced taxonomy of gender or orientation, the labels are being applied to specify distance from the prescriptive norm.

(It is prescriptive; the more actual science gets done about what people are like, the more obvious this becomes.  The prescriptive norm is a political tool to articulate power across a social group.)

That's never going to delegitimize the prescriptive norm.  More to the immediate point, it doesn't attach to the individual with the responsibility; the point to "special needs" is not that the individual differs from the norm, but that someone is responsible to help them do the thing.  The point with accessibility is not that you want more people to be able to do the thing, it's that someone who owns property is responsible to do whatever is necessary to make some list of material things require no more than a specified level of effort.  (And where are the city inspectors using a force gauges to tow little carts with accelerometers, I ask?  Let's quantify the ramps and the curb cuts; while we're at it, add some reach sensors and quantify the swept volume of the doors and ramps and curb cuts.  And the point A to point B energy requirements.  Measure things, don't wave the sacred name of inclined plane at the problem.)

So, anyway; the appropriate labels aren't "disabled" or "special needs"; the appropriate labels (for structures) are "fit for purpose" and "unfit for purpose" and every commercial building should get a pass-warn-fail card just like restaurants do.  In detail.  And if you fail, you can't operate. Her Majesty's servants come and change your locks and turn off the power.  (You're not a legal commercial utilities customer if your building is unfit for purpose.)

The appropriate labels for the education system (in the very broad "become able to participate as a citizen" sense) are "effective" and "ineffective"; can the student do the thing?  Well, ok, what's the next thing?  The measures here are inevitably statistical; there's a lot of things.  List of things, rate of acquisition of ability to do the thing, persistence of the ability, all that stuff must be extremely public at the scale of regional statistical abstraction.  People's jobs depend on there being a positive trend in the statistics.

Sure, some people won't ever be able to do the thing.  (Don't put too much weight on that; anything anybody gets a regular paycheck for is something you very likely can't do at all or that well.  The thing you can do isn't anything like general.)  That's not the point.  The point is that we don't get an effective result by declaring "this is normal; be normal".  We might be able to get a more effective result by presenting a stack of capability and saying "we want you to be able to do all these things"; we can measure that, we can attach the label to the responsibility (ministers of education, school boards...) and we can be descriptive ("when we say thing, we mean..") rather than prescriptive. 

For the most part, the violence inherent in the system derives from the prescriptive; making the prescriptive is work, and damaging.  It takes realized threats to make it stick.  If we want peace, we can't have prescriptive.

16 October 2019

Proportional representation

This is a hard problem.

There are two constitutional guarantees specific to Commons representation; PEI gets 4 seats, as a condition of joining Confederation, and Quebec gets a quarter of the Commons seats, as a condition of the resolution of the great separatism-and-patriation-of-the-constitution debate  that consumed 1970 through 1995 in Canadian politics.

Then there's Section 3 of the Charter; it is held to guarantee not only a right to meaningfully participate in the electoral process, but that constituencies should have roughly the same number of voters.  This pretty much requires a strict counting rule for seats under proportional representation.

Then there's proportional representation itself; the usual result from proportional representation is a coalition government dependent on a small number of representatives holding extreme views.  The extreme views function to drive the agenda.  There is no actual fix for this in a pure proportional representation system; it's systemically inevitable.

The other three things about strict proportional representation is that you stop having a riding and a specific MP who represents you, on the one hand, and you are electing from a trivially corruptible ordered party list, on the other.  (Exchange of favours for list position in a party list is nigh-impossible to prove.  It makes it relatively easy to buy a political party and it makes it incredibly easy to shut out dissident voices.)  Those are both bad.  The "can't have a by-election" (so resignations for gross moral turpitude, taking bribes, etc. don't cost the party seats immediately) is more annoying than bad.

Oh, and the "while we've got the constitution open..." aspect is a problem; someone is going to try to get a guarantee of private property or similar greed-head thing in there, just to make any meaningful electoral reform impractical.  The incumbents never want reform.  Nobody now in power wants a system that guarantees they will never again have a majority government.

So, what to do?  Getting a majority government with 34% of the exercised vote is obviously not just or democratic.  So as citizens we have a compelling reason to change this, but to get it changed we'd need to agree on something to change it into.


Everybody has that Section 3 Charter guarantee of meaningful participation; we can't do anything specific about a proportional representation party list system's trivial corruptibility that involves removing people from it. We can't do anything that works by subscription or produces any additional barrier to exercising the franchise; no setting up representative groups.

Commons is proportional representation.  There's a set "divide the votes by 100,000" mechanism with rounding rules and your party gets the resulting number of seats.  (The first seat requires the whole 100,000; the second seat comes in at 151,000 or 160,000 or whatever the rounding rules say.) The number of seats in the House is NOT fixed.  It depends on the number of votes cast.  (That 100,000 is subject to negotiation; think of it as $SEAT-SCALING-CONSTANT.)

ANY member of the party can put their name in the hat for a seat.  The leader of the party is guaranteed position one on the party's seating list.  "The remaining names are draw randomly by dispassionate officials from Elections Canada" is very tempting but this would be almost impossible to defend; do not give the political parties intense motivation to corrupt Elections Canada.  It nearly has to be a party-members-vote; you can vote for everybody on the list of volunteers if you want, or not vote for anybody on the list if you want.  The total number of votes received determines seating order.  Only when there are ties does something random get used.

There are NO Commons by-elections; is someone dies or resigns, the next person on the party list as of the election that seated this House of Commons gets the seat.

SEAT-SCALING-CONSTANT is different between PEI (to guarantee four seats), Quebec (to guarantee a quarter of the seats), and the rest of Canada.  The Quebec-and-PEI values of SEAT-SCALING-CONSTANT for a particular election are only known once all the votes have been cast; none of this Western Canada "we already know who forms the government, why bother voting?" nonsense we have now.

That takes care of the Commons.  There isn't an obvious Charter violation in there; it's better than Section 3 with respect to first-past-the-post, since your vote will more certainly elect someone to represent you.

The current Senate's this awful imperial relict thing; it's even worse than the provinces as remnants of 18th century communications technology.  So it goes as it is.  We need a way to represent dirt; people need a specific representative whose job it is to advocate for them irrespective of party affiliation.

Trouble is, we need a constitutional way to represent dirt.  The way I'd prefer is to say that area counts less the more people live there.  So about a million people is one senate riding (we don't have the other kind anymore); Toronto and Montreal are a bunch of senate ridings.  Senate ridings are either cities (net outflow of taxes, 3 seats) or regions (net inflow of taxes, 2 seats).  The Elections Canada folks take this tax flow distinction into account when drawing the riding boundaries and maintaining the riding status; if the net flow of taxes changes per census, so does the senate riding's status.

This won't pass a section 3 challenge; a region has 2 votes with potentially the 40,000 people of Nunavut rather than the 3 votes of the million people in city-riding of Mississauga.  Could this be directly added the Constitution?  Sure.  Would it provoke Section 3 charter challenges?  Yes.  Can it withstand the Section 3 challenge on Section 1 ("necessary to a free and democratic society") grounds?  I think it could; we're giving people who certainly don't have one via the Commons a more meaningful political participation and we're not directly disadvantaging the more populous regions because they've got the Commons.  We're doing a checks-and-balances thing but being Canadian we probably ought to call it "many voices".

So we've got 100 senate ridings; Elections Canada draws the boundaries based on combinations of population and area.  No more than a million people population per riding; however much area after the cities (that net-outflow of taxes thing being used to set boundaries) are accounted for to make up 100 senate ridings.  Provincial and territorial boundaries are explicitly ignored when creating senate ridings.

These 100 ridings elect about 250 Senators via single transferable vote.  Senate terms are long (9 years? 15 years?) and staggered; if the terms are 9 years, 1 of a city-riding's senators is up for election every three years.  (The first elected senators have to draw lots for who has the initial short term.)  Senators very explicitly collectively and jointly represent absolutely everybody in their riding and are institutionally expected to keep no secrets from their fellow-senators in the same riding.  They have no party affiliation and may not be members of a political party while holding senatorial office.  Senatorial elections are on the fixed rolling schedule; they have nothing to do with Commons elections.  (The PM necessarily arises from the Commons.)

A bill can originate in the Commons or the Senate; to become law, it must pass both by simple majority.  Neither can override the other.

That's too simple to be an actual detailed proposal and way, way too complicated to sell. This is the big problem with getting rid of first-past-the-post.  Any reasonable alternative is complex and it's hard to build a constituency for it.  Since it requires opening the constitution, it's got to have a constituency.

10 October 2019

Some observations on the carbon tax

A carbon tax now, in 2019, it too little, too late; it's the kind of hinting-and-nudging tweak-the-feedback social engineering that would have been useful in 1980. (A carbon tax today is entirely like telling a heart attack victim they should have taken up daily walks twenty years ago.)  Today, the appropriate policy on the user side is draconian fossil carbon rationing with a five year goal of a zero ration for all users.  (If you [cut by 80% every year for five years] made this year's ration 80% of last year's ration, you'd still be over 30% of the original ration in year six.  This is no, no, we're going to zero.)  The appropriate policy on the source side is "stop that", Ultima Ratio Regum.

In the general case, emissions taxes -- if it isn't actually, observably, measurably reused, recycled, or rotted back into a thriving[1] biosphere, it's an emission -- are a good thing.  They provide pressure to close the loop.  We could do with some.  The problem is the idea of sufficiency.

Sufficiency would mean full industrial mobilization, economic rationalization, and just generally acting like there was a twentieth-century style full mobilization war on to replace the economy really fast.  It would have side effects that involved causing the current crop of rich and powerful to stop being, and the suburban land use pattern to go away.  These things are not inside the domains of good or bad; these things are in the domain of necessity.  If we want to have a machine civilization, we need to do at least that much.  (Giving up machine civilization during catastrophic climate change is not much like a sensible plan.)

It is to my mind extremely telling that the arguments against the carbon tax are that it represents an unbearable financial burden on working Canadians.

One the one hand, this is really terrible accounting; when the roof leaks, the sooner you fix it, the less it costs.  This is exactly that sort of situation, and we've waited until there's leaks down into the basement and rot in most of the metaphorical joists.  This won't be cheap; waiting won't improve matters; one way or the other, we're going to pay.  It would be sensible to try to get something we want.

On the other hand, absolutely no one has pointed out that the problem is not that taxes are too high -- given everything that we really must be doing immediately if not sooner, taxes are much too low and much too insufficiently progressive -- but that wages are to low. It's as though there is a terrible curse upon human knowledge and discourse, that there can be no knowledge of value -- that ratio of benefit to cost -- only of prices, and false -- meaning someone set them, there isn't a consensus on worth -- prices at that.  This is what's going on with labour; the idea that you could, perhaps, pay people more is maybe not unthinkable -- how could you tell? -- but it's completely unsayable.

(And the NDP are still to the right of Brian Mulroney's Progressive Conservatives, because the NDP still aren't saying "raise wages" or "spend on industry".)

[1] None of biological diversity nor biological disparity nor five year population averages of key species are decreasing.

Successor States

History is continuous;  you can point to things, but it's long tangled skeins, rather than discrete blocks.

For practical purposes, here early in the 21st, we can look back at three great braided cables; the decision in the English marcher-state to base their economy on the products of loot (that is, you do some processing for value-add before shipping the loot); the decision by Winston Churchill to move the Royal Navy's dreadnought battleships to oil-fired boilers; and the decision to address the looming food shortage of the 1950s through opening the agricultural nutrients loop and adopting the mass use of pesticides.

This can descend into a lot of historical neepery; yes, a triangle trade where one leg is the rum slave labour produces from sugar slave labour produces from Caribbean sugar plantations is loot, anything you get from slavery is loot.  Power rests on maritime control because a machine economy hasn't got the possibility of autarchy and once you depend on trade strait control is power.  Navies give you strait control; oil-fired boilers really are superior to coal (three shifts of stokers are good for about 18 hours, tops, and then it's days before you can go full speed again; oil, even horrid sticky bunker oil, is pumped, and you can run at full speed until you run out of fuel) but the British Empire didn't have much oil, so it had to go get it.  The result is the 20th century's wars for control of colonial possessions becoming wars to control oil and the rise and hegemony of the United States, the one and only Oil Empire we're ever going to see.  Open-loop agriculture runs out of nutrients to add (see, for example, phosphate); it makes a horrid mess anywhere downstream; it destroys soil.  The pesticides drive mass extinction on the scale of a geologic epoch.  (no bugs is no birds, no bats, no pollinators; no pollinators is back to the Aptian in the Early Cretaceous, more than a hundred million years ago.  Nothing now living is adapted to such a world.)  It certainly wasn't the only way or the best way but it fit in well with an economy based on loot.  (That is, an economy built around guaranteeing you can keep the loot.)

Fossil carbon extraction is going to stop.

It may stop because we have a collective rush of sense to the head, or it may stop because agriculture breaks, taking industrial civilization with it.  But is is going to stop; sure as death, sure as fate.  No human thing can keep it from stopping.

No human thing can keep the temperature from going up two and a half degrees against baseline by 2100; that's the "everyone does everything immediately" climate change mitigation scenario from the IPCC, the very most we can now hope to do.  Somewhere in there, agriculture breaks.  ("rain in due season" is what agriculture depends on, that and six inches of dirt; climate change can be understood as water chaos; wrong time, wrong amount, too much, too little...)  When agriculture breaks, the last ten thousand years of civilization breaks with it.

The mass extinction -- the loss of pollinators, the loss of sufficient soil diversity, the whole "enough blocks out of the trophic web and it does the jenga thing and collapses" -- and the open-loop nitrogen, phosphorous, and carbon cycles in the soil both each suffice to break agriculture all on their own.

So; we aren't keeping the civilizations of the Holocene.

Present politics, politics in our time, politics in as much time as there shall be to come; that's going to be about defining the successor states of human social organization.  That's going to start with "how are we going to eat?" and that does not presently have an answer.

It would be useful to insert "we need an answer about food" into politics, but that doesn't suffice the problem, because the social organization you get is what can keep the other social organizations from taking it over.  You have to win all those fights to copy yourself into the future.  So it becomes "how are we going to eat while creating ourselves into the future in the face of all opposition?"

We shall get an answer; all the current answers approximate to "you won't."

That could really do with going into politics.

03 October 2019

Morals are post-facto rationalizations

Thing is, the ones you got taught as a wee small child were probably rationalizations for something four generations back.

It makes it hard to think effectively.  There's a nigh-overwhelming desire to get everything into a moral frame, and aside from the difficulty of thinking effectively, this makes a couple-three things difficult.

It cripples people up with an expectation of being good; to be moral is to be good, after all.  Only good is always contextual and if you're honest you'll notice you don't know if you're good or not and then you run into the axiomatic expectation of childhood that you will be good and then you're way off somewhere in "the best lack all conviction".

People are social; people are almost eusocial, and we'll pack-bond with anything.  Only there's a numeric upper limit under a thousand somewhere.  (Probably way under; couple hundred.)  The desire to have congruent feels is antithetical to the rule of law and good public policy, because those specifically abjure congruent feels in favour of quantified measures.  There's a thought experiment; talk about your political goes in dispassionate language.  No "want", no feels language -- absolutely no references to love! -- just material results.  Explain why the material result is desirable in purely material terms.  This feels really wrong; it's amoral, it's inhuman, there's a lot of disparaging words for this one.  It's also what's required for effective public policy, because we shouldn't care about how we feel about it, we should care about what it materially accomplishes.

There's no moral way to resolve a moral impasse.  Historical efforts resulted in rivers of blood; the minimum for an effective, permanent resolution of moral disputes is something like the Albigensian Crusade.  If you want to have a functioning society, you have to notice that you need dispute resolution mechanisms that don't involve the river of blood; people have to be willing to lose sometimes.  This is where the social machinery for identifying facts comes from; you can agree on the material world, if you not your moral responses thereto.  ("Tolerance" isn't a stable approach; though it's looking like facts aren't doing all that well, either, great material benefits notwithstanding.)