Evidence-based politics: would such a thing ever be possible?

[teuchter]

And so we get to reductio ad absurdum, because interacting as social beings allows us to be more than the sum of our chemical constituents and their interactions.

I'm not sure about what "more than the sum of" really means.

I'd rather just say it's quite a big sum.

 

[ymu]

But I'd argue that so does human society (on a global scale). Effectively just a bunch of chemicals reacting with each other in the end.

This is fantastically reductionist. Determinism at the level of knowing what thought will pop into my head, what words will be included in a speech and how the media will respond, exactly how a government will respond to a crisis - it's just idiocy. Sorry, but it is. Randomness is an inherent part of our physical universe, and has been known ever since Einstein was proven wrong when he declared that God does not play dice.

However, I can see that those reactions are possibly getting more and more complex over time, and that there are plenty of reactions that haven't happened yet but could in the future.

Not necessarily more complex, but different. Take the obvious example. The printing press changed the human world by making books (not to mention pamphlets) affordable for ordinary people, thus encouraging literacy to become much more widespread, and depriving the priesthood of much of their power because they were no longer the sole source of information about what was going on in the world.

That's not a change you can analyse out. Ever.

 

[ymu]

I'm not sure about what "more than the sum of" really means.

I'd rather just say it's quite a big sum.

Wut?

 

[teuchter]

This is fantastically reductionist. Determinism at the level of knowing what thought will pop into my head, what words will be included in a speech and how the media will respond, exactly how a government will respond to a crisis - it's just idiocy. Sorry, but it is. Randomness is an inherent part of our physical universe, and has been known ever since Einstein was proven wrong when he declared that God does not play dice.

If physicists tell me that true randomness exists, fundamentally, then I am not going to argue with them. I can accept that. My understanding is that that is on a subatomic scale, although of course it has consequences for everything at larger scales too.

And given that, of course no-one can predict something at the level of what thought will pop into your head at a given moment. It doesn't mean that we can't make predictions about trends on a slightly larger scale though, given the right information.

I do totally take your point though, that knowing what the "right information" is exactly, is pretty difficult.

 

[teuchter]

Wut?

Just that when some people say "more than the sum of its parts" it suggests there is the introduction of something "else", some kind of supernatural agent or whateve, rather than simply immense complexity. Not that I think that's what VP was meaning in this case.

 

[ymu]

Just that when some people say "more than the sum of its parts" it suggests there is the introduction of something "else", some kind of supernatural agent or whateve, rather than simply immense complexity. Not that I think that's what VP was meaning in this case.

It just means that the combination of two or more things has properties that are not possessed by any of those things alone. You're not simply adding functions - like vodka and coke - you're creating additional functionality by combining them.

Human beings are not the same as the pile of atoms you could theoretically deconstruct them into. If you cannot predict what thought will pop into my head, or when and where I will trip over my shoelace, you cannot predict everything about the (ever-changing) collection of atoms that makes up me, or anyone else. If we were all just a pile of atoms, you could.

That's all.

 

[mattie]

It just means that the combination of two or more things has properties that are not possessed by any of those things alone. You're not simply adding functions - like vodka and coke - you're creating additional functionality by combining them.

Human beings are not the same as the pile of atoms you could theoretically deconstruct them into. If you cannot predict what thought will pop into my head, or when and where I will trip over my shoelace, you cannot predict everything about the (ever-changing) collection of atoms that makes up me, or anyone else. If we were all just a pile of atoms, you could.

That's all.

A cocktail is a mixture (or solution), not a system.

By the same token, you couldn't predict a car's performance by the properties of carbon or iron, despite the car being made of these and other elements. It's not at that level that the emergent behaviour is entrained or measured.

What you can do is understand how these elements are fitted together - carbon and iron into steel, steel into components, components into sub-assemblies, sub-assemblies into the car. From this the behaviour comes. A crank rod does not do very much on its own. A synapse doesn't do much on its own, but what it does is predictable and can be modelled. The problem is, the system it's part of is a massive, hugely, overwhelmingly iterative system with massive recurrence and with adaptability and decay. A system nonetheless, made up from comprehensible parts.

 

[bhamgeezer]

A wholly deterministic universe is a tyranny.

http://pacherie.free.fr/COURS/MSC/Strawson-Freedom&Resentment.pdf

No it's not

 

[Knotted]

Even if you can effectively research the consequences of policy (and I think this is quite possible a lot of the time), this does not necessarily inform you as to whether it is a good policy. Not everybody's interests will be served. Politics will remain important regardless of how well researched the policies are. We live in a partisan world. If we lived in a world where the common interest was easy to define, then I think yes we would have evidence-based politics. As it is we can only have evidence-informed politics.

Should note we get a lot of evidence-misinformed politics. You get a lot of politically motivated bullshit.

 

[Shevek]

This is a different point to the one already raised but I think addresses the initial question about evidence based politics.

I vaguely recall having a conversation with someone/overhearing someone talking about evidence based politics. Their essential argument was that they were sick of political bickering and speculation and they thought political decisions ought to be based on evidence and research.

I agree that decisions should be partly based on evidence but doesn't the injuction to decide everything by 'evidence' assume that there is some sort of agreement within society on what is the best way to run things.

Would you for example find evidence of how best to run the current status quo or would you seek evidence to support an anarchist society.

I don't like this idea that through process of collecting evidence you can choose between wildly different political philosophies. It seems to me that there are valid areas for disagreement. Its not a case that you can 'prove' that capitalism is better than communism or that trotskyism is better than anarchism or vice versa. These are contentious living political issues.

I do think evidence has a place but I think there is too much essential disagreement about the direction of society for it to be decided by research.

I don't know if I have expressed myself very well.

Shevek

 

[teuchter]

Human beings are not the same as the pile of atoms you could theoretically deconstruct them into. If you cannot predict what thought will pop into my head, or when and where I will trip over my shoelace, you cannot predict everything about the (ever-changing) collection of atoms that makes up me, or anyone else.

OK, maybe I couldn't ever predict everything but that doesn't mean I can't predict something useful.

It's not like it's an all-or-nothing type of scenario.

Even if we can just increase the confidence with which we can make a prediction, based on observation, that is still useful isn't it?

I'm trying to think of practical applications as far as politics is concerned. Take the "trickle-down effect" which certain types of economic approach are based around. It doesn't seem like anyone can really prove that it does or doesn't exist, or quantify it usefully at present. But if someone could prove with a certain level of confidence that it had a certain level of effect in a certain context, based entirely on observation, then that would be very significant information to have available.

Is it totally far fetched to suggest that given a larger quantity of case study than we have available at present, we could at least state with a certain level of confidence that it was or wasn't a significant phenomenon?

We still might be a long way from understanding secondary less obvious implications but at least we'd be better informed than we are at present.

 

[teuchter]

Even if you can effectively research the consequences of policy (and I think this is quite possible a lot of the time), this does not necessarily inform you as to whether it is a good policy. Not everybody's interests will be served. Politics will remain important regardless of how well researched the policies are. We live in a partisan world. If we lived in a world where the common interest was easy to define, then I think yes we would have evidence-based politics. As it is we can only have evidence-informed politics.

But at least we could be arguing just about defining the common interest, or the intended goal, rather than the means of achieving it.

 

[Shevek]

in answer to the OP isn't evidence based politics a bit utilitarian?

You can collect evidence but whose interests do you seek to serve?

 

[teuchter]

Its not a case that you can 'prove' that capitalism is better than communism or that trotskyism is better than anarchism or vice versa.

Well yes, you can't prove one is "better" because "better" is subjective depending on who is defining it.

But, say you could "prove" that one system created a more equal distribution of wealth, but another created more wealth overall. And, even, quantify these differences. At least you would then cut out a lot of argument about what each system can offer, and then you could concentrate on arguing about which end result is more desirable instead.

 

[Shevek]

Well yes, you can't prove one is "better" because "better" is subjective depending on who is defining it.

But, say you could "prove" that one system created a more equal distribution of wealth, but another created more wealth overall. And, even, quantify these differences. At least you would then cut out a lot of argument about what each system can offer, and then you could concentrate on arguing about which end result is more desirable instead.

I don't know if it would be as clear cut as that but I see what you mean. More information on which to make choices is a good thing but as you say there are still choices to be made between different systems and policies.

 

[teuchter]

How about the old "would communism ever actually work" question.

It's easy to list various communist regimes that have collapsed or become corrupt in some kind of way.

It's also easy to give reasons for each of those failures, and say that those reasons aren't fundamentally the result of the communist system as such.

Or make complicated arguments about different types of communism and how the failure of one can't be extrapolated to the failure of others.

But, as history goes by (supposing that further attempts at communist-type systems are made, which of course may not be the case) and you have more and more "case studies", the evidence might start to look strong enough in one direction or the other to say fairly confidently that it's never really going to work, or that it will probably work under these certain conditions.

But, I do take the point ymu made earlier that there may be too many hidden influences that do not get properly considered, therefore making any conclusion inaccurate. The question is whether it would be any less inaccurate than any conclusion we can come to at this point in history.

 

[kyser_soze]

And so we get to reductio ad absurdum, because interacting as social beings allows us to be more than the sum of our chemical constituents and their interactions.

The thing is tho, this is a faith statement. Our actions as social beings could be just as determined by our chemistry as our internal processes.

Ultimately this comes down to belief in indeterminacy or belief in determinate (and therefore measurable, even if we can't do it yet) reality.

 

[teuchter]

The thing is tho, this is a faith statement. Our actions as social beings could be just as determined by our chemistry as our internal processes.

Ultimately this comes down to belief in indeterminacy or belief in determinate (and therefore measurable, even if we can't do it yet) reality.

What do you mean by indeterminacy? Do you mean something more than that there is stuff going on that is truly random?

 

[kyser_soze]

The second sentence doesn't parse mate. Are you saying 'Do you mean there is something more going on that makes stuff random'? I.e. a process happenning in an identical environment 1bn times should be the same on the 1bn+1 iteration, but *something* makes it change (which obv changes the environment, and thus resest the counter to zero), or that there is a chance that the 1bn+1 iteration will be different with no additions to the environment?

 

[teuchter]

Sorry, it wasn't a very good sentence.

What I mean is -

As far as my shaky understanding of current physics goes, there is stuff that goes on at subatomic level which seems to be truly random; ie there is absolutely no way to predict whether A or B will happen (although I think I'm right to say that we can say what the probability is - eg. we know that B will happen 50% of the time).

I can go along with that, and therefore see that it can be argued that nothing can ever be predicted with 100% accuracy even if you have all the information it's possible to have.

But that doesn't mean we have free will.

When you say "indeterminacy" do you mean something else going on that somehow means we do have "free will"?.

 

[kyser_soze]

Yes, it doesn mean we have free will, altho it's not something else going on (i.e a god or little homoculous in my head ) - it's the impossibility of predicting that 1bn+1 iteration of a set of human interactions, irrespective of both your level of information and the models you use to use that information.

 

[ViolentPanda]

Just that when some people say "more than the sum of its parts" it suggests there is the introduction of something "else", some kind of supernatural agent or whateve, rather than simply immense complexity. Not that I think that's what VP was meaning in this case.

My point is that our brain chemistry and it's interactions aren't determinant of the limits of our experience in anything more than a mechanistic sense. They limit our ability to process sensory data, but they don't limit our ability to think creatively, to experience novel sensations or to "improve" ourselves. By existing, living and doing we transcend being merely an electro-chemical machine.

 

[teuchter]

By existing, living and doing we transcend being merely an electro-chemical machine.

See, when you use a word like "transcends" it suggests that there is some threshold that we have passed. What is that threshold?

Does everything "living" transcend this mysterious threshold? Not that there is really a very strong definition of "living" anyway.

Amoebas, viruses, fruitflies, cabbages, dogs, monkeys ... which of them transcend being merely an electrochemical machine?

The way I see it, we're all just points on a sliding scale of complexity.

 

[butchersapron]

You set up a false question (like you also did earlier with 'complexity) then ask people not only to answer it bit to answer for your extention of it to other areas.

 

[teuchter]

You set up a false question (like you also did earlier with 'complexity) then ask people not only to answer it bit to answer for your extention of it to other areas.

"False question"?

 

[butchersapron]

False concept then. 'Threshold' in the above example (ignorance of emergence aside). It's just a trick. It's not a substantive point.

 

[teuchter]

I'm fairly well aware of the concept of emergence. The question is about whether it becomes effectively impossible to usefully predict the behaviour of a system once it exceeds a certain level of complexity, or whether it just becomes more difficult.

I don't think it's true to say that the appearance of some kind of emergent property in something suddenly makes it impossible to make useful predictions about that thing.

 

[Pickman's model]

I'm fairly well aware of the concept of emergence.

good. but do you know anything about it?

 

[butchersapron]

I'm fairly well aware of the concept of emergence. The question is about whether it becomes effectively impossible to usefully predict the behaviour of a system once it exceeds a certain level of complexity, or whether it just becomes more difficult.

I don't think it's true to say that the appearance of some kind of emergent property in something suddenly makes it impossible to make useful predictions about that thing.

And there you do it again. You decided that the issue must be viewed under the sign of complexity and then off you go. (leaving aside your misuse of 'predict' and 'system'). No matter what you're just going to try and compare human society to a system because that's where the logic of the OP leads.

 

[teuchter]

I'm not comparing human society to a system. I'm saying it is a system. Go on and tell me why it's not.