Showing posts with label science. Show all posts
Showing posts with label science. Show all posts

Saturday, 28 July 2018

The core flaw in our democracy


Our understanding of what politics in a democracy should be like is sadly lacking. In fact, the yawning chasm between how we act and how we should act is big enough to swallow the world, leaving nothing but a sad memory of a nice place behind… and is likely to do so.
Everyone has heard of Machiavelli, made famous through his (tongue-in-cheek) advice to his patron about sly deceptions and the pursuit of power over all else. That’s how politics in undemocratic and chaotic societies works; there are no rules and power is everything. That’s our history.
In the modern world we have tamed that conflict. Instead of the barbaric power-mad conniving of our overlords we let rival factions compete peacefully for votes. Who wins power is decided by who persuades voters the best. This has its subtleties: do you persuade people to think you have the best position, or do you concentrate of persuading those who already think so actually to make the effort to vote? Do you hope that by energising your base they go out and convert others for you?
This is fantastic progress. But it’s still basically the same idea: rival groups compete for power through a mixture of trickery, generous promises to the wavering and flowery rhetoric that wins broader support. Of course, the generous promises come directly from public taxes without going through the intermediate step of becoming our leaders’ personal wealth. That wealth, and the wealth of their supporters, pays for the promises to be made – it pays for the research to find out what to promise, and the advertisements to let everyone know. We pay for those promises that are kept.
What hasn’t changed from our barbaric past is the focus on rival factions. What else should we be thinking about? Well, truth and facts seem to have got lost in all this. Yes, the supposition is that truth is more persuasive for the voters, but all the evidence is that this is, well, not true. Does this persuade you? If not, it merely proves my point that truth doesn’t persuade!
People derisively dismiss technocratic government, which they summarize as allowing experts to run things; it’s a matter of faith that democracy is best. It protects us against tyranny and complete meltdown of society, because the people can trust themselves to vote in their interests – and if they don’t, at least they chose the result deliberately. Yet at the same time we put our trust in the science that has achieved modern society: our food production, our communications technology, our transport…
Where does the science of government stop and the legitimate arena of political conflict begin? At the moment, we assume that wherever the politicians choose to go is a legitimate place for them to bring their conflict. Just like when Roman gladiators left the arena and started hacking at the spectators, everyone cheered, munching their dormice and leaded wine and … no, wait. They had guards to send any straying slaves back to where they belonged. Actually, comparing modern politicians to gladiators is laughable. It makes them seem much braver and more macho than they are. But they are public servants, and modern politics is as much about the spectacle and as much about revealing truth or persuading people. Perhaps a sports simile is more appropriate? They’re like footballers, and we should expect the ‘game’ to be played on the pitch and according to the rules.
In fact, it’s very like a sport, in that most people have their favourite side, and the most important thing to them is that that side wins. If one of their players dives for a penalty, it’s clever gamesmanship; if an opposition player does it, it’s dishonourable and dishonest. The rules should apply strictly to others, but are a guideline for us.
We invoke modern concepts of ‘the marketplace of ideas’ to suggest that the wisdom of the crowd is best and democracy will always reach the truth. And yet apart from these occasional passing references, we ignore the truth as a goal. Maybe a crowd could collectively determine what is true (a dubious claim in itself), but not when the crowd isn’t even looking to buy that particular product! The goal is always for our side to win! Truth matters only because lies and lack of understanding are a convenient way to mock and belittle the opponents. Truth isn’t the goal; it’s just one way to make the other side look bad.
Imagine if science worked that way, not only in academic research (where people are people and there is often rivalry) but also more widely? Imagine if half the passenger flights in the world were planned as if the world were flat, mostly driven by a few powerful organizations that demanded it. Imagine that all the discussion over which holiday to go on focussed on rival advertisements claiming that their location was sunnier, and occasionally on mocking the other airlines because a spokesman hadn’t worn a tie. Imagine if public broadcasters had all the airlines on some sort of Watchdog programme, and gave the flat-worlders as much airtime as anyone else, or possibly more.
Would that be a sensible way of doing things? And if holiday-makers ended up in the wrong places all the time, and the airlines with the most flat-world flights were the most sympathetic, offering discount return flights for people stranded, would you buy more tickets from those airlines?
A lot of government is exactly the same. We know that parties’ policies won’t work, or won’t achieve their goals, or cannot be combined with some of their other policies. But we value the assertion of intent more than the truth: we value the statement of identity more than expertise.
Identity politics; tribal politics; democratic politics – these are all the same, and they are all based on the same barbaric notion of politics as conflict and beating our opponents. Plato, 2,000 years ago, tried to imagine a different mode of governance, of philosopher kings who governed with wisdom. We mock his ideas because we have seen countless seemingly benign tyrants become tyrannical, but a core part of his observation remains as valid now as it was then: it is truth and justice that are the foundations of society, and a mob isn’t the best way to find either.


We should be co-operating, as scientists do, to find the best ways to run a country. Our political parties should be exploring different possibilities, sharing data and coming up with results… and we should reward those who have the best findings, not those with the most appealing conclusions. The space for genuine conflicts of opinion is minimal compared to what can be understood and known for certain. Not all opinions are equal, even if those who hold them are.
If we don’t value expertise; if we give platforms to liars and charlatans time after time without asking them to recant and repent; if we care about entertainment instead of truth and complex debate, then we might as well be in a dictatorship. After all, neither system will provide what is best for us, but at least a dictatorship doesn’t involve the soul-corrupting self-deceit that currently defines us. Without a drive for truth, we are just semi-wild apes following our leader because he has covered his rival with more mud.

Wednesday, 28 April 2010

Bayesian inference

http://en.wikipedia.org/wiki/Bayesian_inference
'One of the crucial features of the Bayesian view is that a probability is assigned to a hypothesis, whereas under the frequentist view, a hypothesis is typically rejected or not rejected without directly assigning a probability.'
Most people have a truth and change it when enough evidence has accumulated. Some people learn to treat truths as more or less probable, but have to do this consciously. This leads to paradigm-shifting, because people stubbornly resist new evidence until it's really overwhelming.

Some experiments on belief revision have suggested that humans change their beliefs faster when using Bayesian methods than when using informal judgement. If I say I am intuitively Bayesian, I am claiming that I more naturally accept a certain level of uncertainty in statements of 'fact'; that my informal judgement typically involves subconscious assignation of probabilities of truth to certain statements rather than a binary system of accepting or rejecting it.
I'm sure that most people do this to an extent. But studies show that it's certainly not an everyday mode of thought.I believe that this approach to science is very important. Scientists currently test whether a hypothesis is true by seeing if the data pass an arbitrary limit of probability. If, given that the hypothesis is not true, the probability of getting the data is too small, they conclude that the hypothesis is true. The Bayesian method would be to assign a probability of truth to that hypothesis, using the data to measure that degree of probability.

Monday, 19 April 2010

The link between junkies and politicians

http://www.ted.com/talks/elizabeth_pisani_sex_drugs_and_hiv_let_s_get_rational_1.html

The points I want to take from the talk are these:
Politicians and junkies do stupid things: politicians make stupid policies and junkies do things that help them get HIV.
People respond more favourably to 'making the world a better place for starving children' than to 'making the world a better place for junkies'.
If junkies are going to be punished for those things they can do that reduce the risk to themselves then they will balance the risks of being caught and getting ill and take more risks.
Both have politicians and junkies have rational reasons for doing what is, practically, worse.
Politicians respond to people's desires to be tough on drugs and crime by punishing drugs users, and using any evidence of drugs use as an excuse to prosecute. They do this even though it costs the nation more money in health-care and harms more people's lives.
Junkies respond to these induced risks of having drug-using equipment by sharing equipment. They do this even though sharing equipment means that roughly half of them will get AIDS.

It's an interesting enough observation if we stop there and simply suggest that we take more notice of sensible solutions like needle exchanges, rather than ignoring them.
But I like this story because it brings attention to a problem that is dawning on me, and has been forming in the world for a while now, and that is the conflict between science and politics.

Another good talk on the subject was here:
http://www.ted.com/talks/michael_specter_the_danger_of_science_denial.html

Michael Specter talks about a number of instances of 'science denial' which have had measurable negative consequences. The aversion to vaccines, fear of genetically modified crops (which is foolish, unlike quibbling with the direction that GM research is taking) and, of course, denial of climate change, all have direct consequences that add to the sum total of suffering in the world.

Elizabeth Pisani has found yet another one of these instances of science denial, and hers is a fine example of politically-motivated science denial. This is not a scare story about vaccines spread by stupid or cynical journalists and believed by the gullible and ignorant. This is the deliberate prioritising of political expediency over scientific fact, and it is this aspect of science denial that is as worrying as any other.
We can all agree that gullible people need to be educated, and we can all agree that media scaremongering in order to sell copy is morally dubious (at best). But I am certain that there are people out there who truly believe that science should be subservient to politics.
I have heard someone say that he has 'no problem with science or scientists, until they try to dictate what we should or should not do. That's meddling with politics, and they don't belong there.'
I'm very keen to stress that science is a process, not a belief-system, and that it makes no moral demands or statements. One cannot call scientific knowledge evil; one can merely say that it is being abused. However, part of this labelling science as a process is describing what it does. Science is the best process by which we ascertain truth. We have no better method for finding out what is true and what is false.

Given that science gives us what we know to be true, let's look at the idea that science does not belong in politics again: this idea is telling us that the truth is unimportant in running our countries. I find this idea incredible. Do we really want politics based on lies, rumour, scandal and opinion? I know that's what we have, but are we happy with it?
Voting figures and opinion polls suggest that we're not.

I see here a fundamental conflict between the current politics, the ideal of democratic engagement and the ideal of enlightened rule. Obviously the current political system is neither enlightened nor democratically engaged, and people who are empowered by the status quo try to 'play off' the two opposing ideals. Democratic engagement at the moment would mean less science involved in governance, because it would hand power to the large media corporations who promote and sway opinion. Democratic engagement is a measure of opinion; if it were perfect each and every person's opinion would matter. Enlightened rule contrasts with this because opinion does not matter: only truth matters. Science tells us what is most likely to be true, and so where science gives us an answer we would do best to heed it.

The simple problem is, as Michael Specter says, sometimes you are not entitled to your opinion. Where science gives us an answer then democracy is defunct. If we are to aspire to being well-governed and democratically governed then we must accept limits on those things on which democracy can pronounce. We must have a constitutional recognition (written or unwritten) that no amount of desire or stubbornness can undo scientific findings.
Science is rarely concrete, but when it reaches a conclusion, that conclusion is the most likely to be true. Democracy must cede to science those areas of life that science can govern. We do not vote on the best way to treat high blood pressure, and we should not vote on the best way to prevent AIDS. We cannot have people staking out for democracy every decision that a government can take.

I see here the ghosts of science denial in all its forms, trying to keep science out of politics. It is not only those who truly believe that it is morally right to value majority opinion over truth, but those who stubbornly believe in falsehood and will side with majority opinion to protect it from truth.
I can imagine that a large number of fundamentalists would be upset were science to be welcomed into government. Science is the utter opposite to fundamentalism. Where fundamentalism holds certain facts to be universal and true, science questions everything, and no theory in science is unquestionable. That is not to say that one can simply discard a theory; to question is not the same as to ignore. Where fundamentalism provides a solid foundation of unmoveable knowledge, science provides a shifting sea of grains of data which form great pyramids of knowledge, but can occasionally fall apart.
It is the stubborn adherence to dogma of any sort that is the greatest rival to science. Religion springs to mind as the most obvious example, but free-market fundamentalism can be as dogmatic as any religion.

Stubbornness is inherent to human beings. We are stubborn creatures: economists call it the sunk-costs fallacy, psychologists no doubt have a name for it too (I remember going to a talk about research in which people stuck with their previous option in a sequence of choices 70% of the time. The best solution would have been to change every time). We can't fight stubbornness sub-consciously. We must have it laid out before us: dogmatic faith is no way to run a country.

Running a country is a practical task. It is hard enough to do well without having reality questioned by any sort of faith.

Taking stock of a public figure: market-maker or made by market failure?

       I’m torn over whether Gary’s Economics, and Gary Stevenson himself, are a good thing for or a symptom of the problems society faces...