Friday 27 January 2012

Friday Links

Wednesday 11 January 2012

Oh. The Greens have got a think tank.

When people talk about "providing a voice to future generations" the first thing I think of is the anti-choice lobby. Obviously this report from a Green think tank isn't that, but it's still more than a little bizarre.

And yes, I know, one doesn't have think tanks for the purpose of coming up with workable or sensible ideas, but I was hoping for better from the Green Party, as the closest we have in England to an electable left wing party, than to be encouraging this sort of thing.

If it had been some think tank for Labour or the Conservatives I'd probably just have laughed and ignored it: it's more disappointing when associated with a party I might want to vote for someday.

The basic idea is that there would be a panel of people forming a "third chamber" of the legislature, who...

[...] would have a power of veto over legislation that were likely to have substantial negative effects for society in the future, the right to review major administrative decisions which substantially affected future people and the power to initiate legislation to preserve the basic needs and interests of future people.

Right. Now, okay, there is a very definite conflict between the incentive to mainly act and plan for short-term successes that elective democracies have, and the need to make longer-term plans to have a sustainable society. Certainly, this is going to be a problem with pretty much any political system in which the faction "in charge" can change over a shorter timescale than that needed for long-term planning. (Of course, a political system in which the faction in charge can't change over short timescales has its own severe problems)

The proposed solution claims to address a weakness that despite democratic reforms, no present democracy considers the needs of future voters. So, this panel would guess what their needs might be and force legislation to be made accordingly.

Decision 1: road maintenance budgets should be cut. People in the future will all have flying cars.

How do we know that something will be a problem in the future? In general, we don't. A big hint is "is it a problem today". But if it's a problem today, then a system of government that did an adequate job of representing all currently-existing people would deal with it anyway.

Even assuming the panel have perfect access to and interpretation of all current human knowledge, their decisions are going to be based on "best guesses" about the future. Those guesses are going to be wrong at least as often than not.

And since the purpose of all legislation is its effect in the future (generations or years or days), why would the panel be significantly more likely to get "impact on the future" right than the existing legislative bodies?

The panel - and the report recognises this, though not for the reason I give - could not be elected. If it were, then we'd elect the Commons to make decisions, then we'd elect the Lords to disagree with those decisions, then we'd elect the "Guardians of the Future" (No, really, that's what they're called) to veto anything that gets past them.

So, instead, they'd be selected at random in a manner similar to jury service. It's a tempting idea - I've seen it suggested for House of Lords reform before - but it just wouldn't work. The suggestion is that anyone on the electoral register who doesn't tick the "I don't want to be a Guardian of the Future" box would be eligible for random selection. Once selected, they could opt-out at that stage, but if they didn't there would be a year of training followed by a multi-year term of office.

So, who's going to opt out (either on the register or when selected)?

  1. People who are ineligible to register.
  2. People who do not register.
  3. People who have health conditions incompatible with an intensive multi-year commitment.
  4. People who have caring commitments likewise incompatible (the report talks about subsidised or free caring services, but not everyone will want to take up that offer)
  5. People who do not wish to be in an extremely public role likely to come under considerably more public scrutiny than the average MP gets.
  6. People who don't really care about legislative politics.
  7. People with low self-confidence in their own abilities.

And so on, and so on, and so on. See also s.e. smith's Civic Engagement: Only For the Wealthy! which inspired a lot of this section.

Notice that a major common factor of all of those groups is that they are also strongly under-represented in our current political system for much the same reasons, and that they are correlated with various lacks of privilege.

Not needing to campaign and sign up to a political party might help, but not much.

The number of Guardians would be relatively small (7 to 144 is the range suggested, with 12ish being preferred) - which means that to get a representative group would be very tricky. But then, that's not a particular aim. From Appendix B, which sets out selection:

For such demographically-based pre-sortitional selection would detract from the sense, already indicated above, in which this grandest of juries must represent the voiceless, a task for us all which is in important respects equally hard, one might venture, for any of us.

Well, yes, in the "predict the future" sense. And no, in the "interests of future people who are not like me" sense. Given that representing the "interests of people not like me" is - due to privilege - one of the major failings of our current political system, a lack of concern for the same failing in a body of significant political power is disturbing.

(Now, a footnote points out that if you select by certain demographics you can bias away from demographics not selected by - I've mentioned that problem before and it's especially serious for small groups)

There's a footnote referenced in the context of the length of the commitment which is also worrying.

47 Given the vastly longer duration of the role, it would be necessary to provide the Guardians with some kind of salary. Probably this should simply be the salary they were getting in their regular job, plus generous (but not corruptly so!) expenses. Alternatively, all the Guardians could get the average national wage (as, arguably, MPs should, too). [...]

Emphasis mine. Spot the really obvious problem with that bit that significantly biases ability to be a "Guardian"... The alternative is not great either, but at least the bias it inserts is somewhat counter to current privilege.

So, okay, we have our randomly selected sample of people who will be protecting the future from the present. So, how will they decide what to do? This perfect access to the world's information won't make itself.

The Guardians would be supported both in their training and in their day to day roles as Guardians by a high- level and diverse support staff of administrators, facilitators and experts, including of course legal experts. It would be very important to stop this support staff serving the Guardians from becoming too powerful (as perhaps the civil service is too powerful in relation to government ministers, in Britain). Primarily, those serving the Guardians closely should consist literally of assistants, secretaries and administrative managers and facilitators, plus a cohort of top academic etc. advisors employed on retainers. The Guardians would have strong rights - including if necessary subpoena style rights, to call any and all further actors and experts that they wished to hear from, to help them in their deliberations.

Okay. With a very small number of Guardians, the chances of any of them being themselves an expert in a specifically-named field is small. If the selection was representative, most of them will have average mathematical and scientific knowledge. A year's training will not give them the knowledge to really tell which of two conflicting experts has the best theory. Is X safe or will it cause serious problems in 200 years? Who knows - the experts disagree, the Guardians don't have the expertise to assess the claims, and beware of experts anyway.

So ultimately they will pick the most persuasive expert. Who is probably themselves wrong.

The existing House of Commons is already perfectly capable of picking the wrong expert advice; it seems unlikely that some randomly-selected people would be significantly better. (And they need to be significantly better, not merely equal, or they're not actually helping)

That's if the Guardians can agree at all. They're randomly selected in part to avoid party political considerations. But it also means that they will have a broad spread of political views. So, how will they make decisions? By consensus.

They would be given training for example in consensus decision making, in a group with a common purpose.

Consensus decision making is really quite difficult between people with completely different core assumptions. So in practice most things would come down to a vote between different political factions. The factions might not be there already - political party membership is so rare that perhaps one of the Guardians might have it - but they'd form pretty rapidly.

See also: Tyranny of Structurelessness - even though this wouldn't be unstructured, it still applies: at least with the selection method for MPs there are so many big egos that they cancel out, whereas with a random selection a couple of charismatic big egoed people can end up very dominant.

In practice then we have a third chamber that will tend to always fall on the same side of political debates about whether a policy is a good idea or not. What side that is is selected randomly, but that doesn't really help. It also gives the "centrist" Guardians the most power as the "swing" voters - in a mini-representation of a marginal Parliamentary constituency, or a slightly enlarged version of the current US Supreme Court.

Perhaps their belief in a common objective would help. Perhaps not. There's a suggestion for their "oath of office":

"I promise to do my utmost to represent and uphold the basic needs of future people in the present. I will work to ensure that the future people of the United Kingdom are cared for by us all, and that nothing that we in the U.K. do hurts them or prevents them from being. I vow therefore to execute faithfully and to the very best of my abilities the office of Guardian for Future Generations."

Emphasis mine, again. Are you thinking "oath of allegiance to extremist anti-choice group?" I'm sure it's accidental, given the Green's general policies, but still! Especially given that the explicit point of this group is to prevent Parliament passing legislation that benefits currently-existing people at the perceived expense of potential future people, how do you stop the anti-choicers being all over that with "Look, it says right in your job description you have to ban abortion. You took an oath of office to that effect." (The draft wording is creepily anti-choice, but the group's purpose probably precludes anything without that flaw1. At least it by doing so avoids the eugenicism that previous "protect the future" movements have had, though it does still allow for invading other countries to take their land for future UK generations)

Let's back up a bit to the start of my criticism. I said "How do we know that something will be a problem in the future? [...] A big hint is 'is it a problem today'."

The major problem with our existing system of government, in my opinion, is that it does not give a voice and sufficient consideration to the concerns of large groups of people, especially those who are far from being the default human. If we can't get a system of democracy that listens to real existing people and considers them to have equal humanity and worth - and that extends to "Well, this would be great for the UK, but really mess up the lives of a bunch of foreigners who aren't part of UK democracy." of course - do we really have any business trying to represent some hypothetical future people.

Conversely, if we actually took seriously the representation of people in all their interests - including, especially, those who won't become MPs themselves - would we need specific care about future generations? What thing could we possibly do that would not significantly harm a group of people in the short-term but would harm a group of people in the long-term?

Nuclear waste has been suggested as meeting that, but I don't think it does: we know nuclear waste is dangerous now, and either we store it properly in which case future generations are going to be fine - and will have better storage and reprocessing technology to make up for our inadequacies - or we don't store it properly and some current people in the short-term will get hurt (and we don't, and they are).

You don't need an appeal to the needs of future people to do this - an appeal to the needs of certain current people would be enough. If we can't care about those current people now, we have no business privileging the needs of our hypothetical future people above both them and their hypothetical future people.

I get the motivation - embedding good environmental decisions in government decision making is going to be necessary for us to survive. But all of the environmental problems facing us today are already having massive negative effects on currently existing people. If that's not enough for us to do something, what does it say about us if future existing people is? And we have structures that care about some currently existing people already - they just don't care about all of them. But that's not a problem with the democratic structures as much as with the people inhabiting them - and their privilege. Adding more people with the same sorts of privilege won't help matters.

Footnote

1 Yes, I'm fully aware that this hypothetical anti-choicer's argument contains several major logical flaws. But it sounds superficially convincing if you loudly ignore those flaws.

Monday 9 January 2012

Political strategy mistakes

So, as previously covered, Ed Miliband wants Conservative voters, but isn't willing to defect to the Tories to get them.

How's that working out for him?

According to YouGov polling, not well. The polling is mainly about the Lib Dems loss of voters, but there's a bit about the change from May 2010 vote to current voting intention.

According to the polling, 3% of those who voted Conservative in May 2010 would now vote Labour. Yes! Clearly the strategy of drifting rightwards is paying off!

Unfortunately, also according to the polling, 3% of those who voted Labour in 2010 would now vote Conservative. So allowing for margins of error, the net move from Conservative to Labour is indistinguishable from zero. Oh.

I can't say I'm entirely surprised. The second preference polling - back when AV was in the news - always gave extremely low proportions (around 5%) of Conservative or Labour voters who were willing to give their second preference to the "other side". Those voters are naturally also the most likely to transfer their first preference from one party to the other. And there's hardly any of them.

Similarly, page 5 onwards of the poll results gives "like/dislike" figures for each party, broken down by 2010 and current vote. Only around 10% of the 2010 or current voters for each party even slighty likes (6 or above out of 10) the other party. There's really not a lot of scope for persuasion there.

Now, the right wing of Labour will say, Blair won 3 elections by drifting rightwards. And this is somewhat true ... but, on the other hand:

  1. Correlation is not causation. Pretty much any manifesto would have beaten Major's Conservatives in 1997, and given their disarray, 2001 as well.
  2. Sooner or later, voters are going to decide that if they want Conservative policies, they might actually just as well vote for the Conservatives.
  3. The Conservatives can tactically retreat and win by losing.
  4. There's a major confusion between "position" and "rate of change of position" in the argument. Even if you accept that it's true that Labour moving to the right both caused their 1997-2005 wins and was a good position for Labour to be, that doesn't mean that a further move rightwards is still good strategy.
    For a sports analogy, football players in their 20s tend to perform better than football players in their teens. So, says the Labour election strategist, the team with players in their 50s will be even better.
  5. The usual comparision is with Labour's terrible 1987 election performance. But the Conservative party have not moved so far to the left since then that there are no meaningful political positions between "nationalise everything" and "the Conservative manifesto".
    Meanwhile, there isn't a lot of viable space between the "current Conservative position" and the "current Labour position" on a lot of issues.

The polling figures

...but then, what about the polls? Lib Dem and Other support has been roughly constant, and the relative levels of Conservative and Labour support seem to go up and down as you'd expect from good and bad things for one party or the other hitting the news.

The thing that's forgotten there is that the headline polls exclude1 "don't know" and "wouldn't vote".

Good or bad news for a party will probably cause some voters to prefer or drop it - but for or from "don't know", "wouldn't vote", or a minor party. When the impact of the news fades, then they might well drift back to their previous position. Because the polls are normalised, then this effect can look like one party is gaining at the direct expense of another - but, at least with the Conservatives and Labour, this is not actually happening on any detectable scale.

The relative positions of Conservatives and Labour have changed by roughly 10 points since the election, based on current polling. Essentially none of this net movement has been direct transfers. (Indeed, currently, both parties are doing better in percentage share terms than they did at the election itself)

Footnote

1 Some polling companies instead attempt to reallocate these people instead, to improve the accuracy of their forecast. The overall effect is similar, though.

Saturday 7 January 2012

(It's still) Friday (somewhere) Links

Wednesday 4 January 2012

Beware of experts

Well, not so much "beware of experts" as "beware of definitions of experts".

It's fairly obvious that the way one recognises an "expert" in one field is not going to be the same that one recognises an "expert" in another field.

Obviously expertise and specialisation is a good thing. The problem is that identifying expertise in a field that you have little knowledge of is itself a tricky problem. So there are conventions as to what experts "look like" - some of those conventions are themselves directly rooted in privilege: they look like middle-aged middle-class white men - others are only indirectly so: academic experts tend to be found in research institutions because that's a major way that they become experts, but not everyone can get the chance to work at a research institution even if they have a strong aptitude in and interest for that particular field.

There are lots of problems with recognising expertise from the outside - a lot of my former posts on bad science reporting are caused by precisely that - and even on the inside experts will disagree with each other - but I think there's a particular issue in the way that academic expertise is identified and defined that matters.

This is: all science is basically statistical.

An expert in metallurgy is not an expert on every individual atom within a metal block (you need a different expert for the next block...) but on their statistical behaviour. An expert in economics is not tracking every penny separately but looking at the statistical movements of much larger sums of money over time. We recognise this.

An expert on psychology or health or other fields that consider the general behaviour of living creatures in particular conditions also only has statistical knowledge. We do not recognise this as much.

An atom of metal, or a penny, individually, may well not react in the overall way that a metallurgist or economist predicts that it will. This doesn't usually matter - the overall effect will still be what they predict. (And the individual variation between atoms or pennies is small)

When discussing people, however, a person not behaving in the way that the expert predicts is important. The expert's knowledge of the situation is statistical. They can tell that of 100 people, 90 will do a certain thing. The other 10 will not.

When attempts to predict are taken to the individual level they will regularly fail. The expert is an expert in general behaviour and situations. The individual is an expert in their own personal situation. So an individual's perception of their own situation - and what they should therefore do - should in general be treated as more reliable than the expert's1.

What actually happens is that the median behaviour the expert has identified - and their identification of such can be quite incorrect through their own privilege issues, of course, but even if all of that is somehow avoided - tends to become regarded as correct and deviation as incorrect. (And, further, stay regarded as correct even if the median behaviour later changes)

It also gets to a point about the difference between a claim that "X is a problem" and "X is an [adjective] problem" - which often gets intentionally ignored.

To make a claim that a particular thing affects a particular proportion of people, or that its prevalence has been changing over time or space, or that a particular counter-measure is in/effective one needs to do a very carefully designed study, probably more than once. This is really difficult to do well.

To make a claim that a particular thing occurs at all, seriously affects those it occurs to, and therefore should be countered ... you need to listen to a few people. This is also really difficult to do well, especially for people with the "anecdote is not evidence" attitude that works well in particular scientific fields and for statistical research and works very badly when exported from there.

Unfortunately, that sort of person is likely to fit right in to an academic environment where they can become a socially-recognised expert, because our expert recognition is insufficient.

We need a separation in the way that society recognises experts between "experts in statistical behaviour of people" and "experts in individual behaviour of people".

Footnote

1 The obvious exception is when the expert - who may in this context not be a socially-recognisable expert - has knowledge highly material to the individual's situation which they know the individual does not have. Of course, in most cases, not giving this information to the individual is intentionally forcing a mistake on their part.