Ideology vs. Free Thought?

southsiderosie

KyussBishop wrote: »

I'm using the scientific method as a standard to judge the social sciences and what they should aspire to; Libertarianism stands apart in its deliberate rejection of the scientific method, which curtails its credibility, whereas other social sciences that still aspire to the scientific method are more credible.

Two questions:

1. What fields of social science are you referring to?

2. Why do you think this makes them more credible?

First, the process of gathering 'objective' data is itself highly value-laden - the census is a case in point.

Second, many studies that attempt to make predictions about or model human behavior far too often rely on totally unrealistic assumptions. Hence the bigger problem in much of social science is not the actual approach to understanding society, but rather the underlying assumptions involved in a given approach.

Finally, I am really struggling to see how you would falsify theories of human behavior in the social sciences. Take collective action, for example: a key question in sociology and political science is, why do people engage in contentious politics? Marxists, utilitarians, rational choice theorists, classical liberals, institutionalists, and primordialists would all answer that question quite differently

KyussBishop

southsiderosie wrote: »

Two questions:

1. What fields of social science are you referring to?

2. Why do you think this makes them more credible?

First, the process of gathering 'objective' data is itself highly value-laden - the census is a case in point.

1: In the context of this discussion, primarily economics
2: I think social sciences that take into account available empirical data, and also make predictions that can be falsified or have doubt cast on them by emprical data, are inherently more credible as they aspire to use these methods to underpin their theory (EDIT: evidence which gives them more solid grounding/credibility based on real world data).
The difference with Austrian economics from most social science (economics mainly), is how it explicitly rejects the very idea of this.

I don't disagree that there are many problems managing the accuracy of empirical data used, and how it can be applied, but the rejection of the use of it in shaping theory and even of providing a means to falsify it, inherently makes the theory lose credibility (from my point of view).

southsiderosie wrote:

Second, many studies that attempt to make predictions about or model human behavior far too often rely on totally unrealistic assumptions. Hence the bigger problem in much of social science is not the actual approach to understanding society, but rather the underlying assumptions involved in a given approach.

This is pretty much the exact problem I see with praxeology, which underpins Libertarianism; this kind of stuff needs to be tested with real world data (to at least look for problems), because the bigger and more complicated your framework of assumptions/ideas (that don't have solid empirical backing), the more likely it is there is a serious flaw in there somewhere.

southsiderosie wrote:

Finally, I am really struggling to see how you would falsify theories of human behavior in the social sciences. Take collective action, for example: a key question in sociology and political science is, why do people engage in contentious politics? Marxists, utilitarians, rational choice theorists, classical liberals, institutionalists, and primordialists would all answer that question quite differently

Well, it's not hard falsifiability, where the whole theory would come down; I think theories should make specific predictions which are falsifiable, such as many Austrian economists prediction of impending hyperinflation (which has not come about), as well as attempts to refine theory in order to try and model potential economic outcomes, in a way that can be falsified as well.

If theories try to make specific predictions like this, they are (in my view) aspiring to the scientific method, and providing useful debate; since the scientific method and use of empirical data is rejected by praxeology, it severely limits the credibility of the theory in that regard (in my view).


I'm not sure I answered that very accurately Maybe there is the impression or some confusion that I'm trying to falsify base Libertarian beliefs:
I'm not trying to do that, because if someone holds different societal values then that's that, I'm more criticizing the rejection of scientific methods in underpinning the credibility of the overall theory, regardless of whatever social values are used to justify it.

Permabear

This post has been deleted.

southsiderosie

KyussBishop wrote: »

This post had been deleted.

The problem with this view is that the theories that really reshape how people in a given field within social science think about things are the ones that aren't necessarily grounded in carefully gathered empirical data. There are plenty of theorists who present an alternative theory of how the world works that are logical, and dare I say, elegant enough to completely re-orient how we think about a given problem, and they are happy to let other people test that theory in multiple settings. Even if they don't necessarily hold over time, 'failed' theories often have the benefit of opening up unexplored areas of research that can yield real insights that refine existing approaches to understanding a problem.

To bring things back to the main idea of the thread, the fetishization of modeling human behavior in the social sciences is in and of itself an ideology that has led to gross distortions in the interpretation of social behavior and phenomena in political science and economics, and has reduced many subfields to little more than attempts to count the number of angels that can dance on the head of a pin.

Permabear wrote: »

This post has been deleted.

Amen.

Valmont

KyussBishop wrote: »

I'm using the scientific method as a standard to judge the social sciences and what they should aspire to; Libertarianism stands apart in its deliberate rejection of the scientific method, which curtails its credibility, whereas other social sciences that still aspire to the scientific method are more credible.

Again, you are asserting that libertarianism rejects the scientific method in its entirety. You have yet to explain how! You are mistaking praxeology's contesting positivism with a rejection of science entirely and ignoring any calls to expand on your position. I don't know the answer to the issue but it would be good to actually discuss it!

Eliot Rosewater

KyussBishop wrote: »

That's a silly cop-out which tries to make it a matter of 'rights and wrongs' i.e. opinions, when the claims made by the theory are a matter of fact whose credibility is subject to question by the scientific method.

Your claim was that political ideologies are ultimately subject to the scientific method. My counterclaim was that at best a subset of the beliefs of a political ideology are subject to the scientific method. Perhaps, if I'm wrong, you could tell me how the conservative and liberal views on marriage may be compared using the scientific method?

KyussBishop wrote: »

Heh, ironic much? In its rejection of empirical evidence, Libertarianism is precisely like the Popes refutation of Gallileo's ideas; it's like the Pope reasserting his own beliefs without any kind of need for scientific backing, because the beliefs are consistent with his ideological framework.

Frankly, this is just anti-intellectualism. You're saying "I'm supporting the scientific method on this thread, therefore any claims that my means of supporting it are unscientific are rubbish." That is plainly nonsense. One can support the scientific method using unscientific arguments.

KyussBishop wrote: »

Your entire argument here rests on the idea that you can only try to disprove a theory from the ground up on microeconomic effects, not by disproving its claims on macroeconomic effects, which is a completely silly proposition.

Yes - my claim is that it is probably impossible to judge the effects of one economic policy by looking at aggregate data for the whole economy of which the policy is merely a tiny component. Presenting an increasing graph of the USA's GDP for the 20th century does not provide evidence for or against the economic benefits of the minimum wage. The graph says that on an aggregate basis the economy performed positively, but it is impossible to solve the "inverse problem" and attribute that positive growth to single particular aspects. It is highly possible that the effect of the minimum wage was negative, but that the negative effects were cancelled out by other policies.

Really - isolating certain causal factors is such an essential component in the practice of hard science that I'm baffled at your dismissing such concerns as "silly".

KyussBishop

I'm not saying economics has to be a natural science, and I haven't challenged at all the uncertainty inherent in economics, and I'm not claiming an economic theory is invalid if it can't do a perfect "x-y-z" explanation of a theory, I am critical due to the deliberate inapplication (and stated rejection) of actual real world data to backing, shaping, verifying and falsifying theory.

Whatever economic framework you come up with, you need to test its assumptions, and praxeology rejects the idea of using empirical evidence to actually test the underlying assumptions of the theory; neoclassical economics fails here as well, by not properly incorporating conflicting real world data to reshape theory.
I am aware that uncertainty in data puts limits on falsifiability, and the inherent uncertainty of human behaviour in an economy also adds uncertainty, but nonetheless it is essential to base development of the theory and credibility of it, on how well it matches observation, which is best guided through empirical data (taking into account limitations in data, as it is useful regardless).

With Austrian economics, the more complicated it's underlying assumptions that it tries to put beyond the realm of (in)validation through observation and empirical data, the higher the chance of significant flaws within; hence why it loses credibility through rejection of empirical data.

Even concentrating solely in the social sciences, there is a huge distinction between the credibility of theories which use a scientific process to guide and found development, and those that specifically reject it; this process does not have to be a perfect one of developing universal quantitative laws, or describing a system with complete accuracy and perfection, it simply needs to strive towards greater credibility through aspiration to scientific standards.

southsiderosie wrote:

The problem with this view is that the theories that really reshape how people in a given field within social science think about things are the ones that aren't necessarily grounded in carefully gathered empirical data. There are plenty of theorists who present an alternative theory of how the world works that are logical, and dare I say, elegant enough to completely re-orient how we think about a given problem, and they are happy to let other people test that theory in multiple settings. Even if they don't necessarily hold over time, 'failed' theories often have the benefit of opening up unexplored areas of research that can yield real insights that refine existing approaches to understanding a problem.

To bring things back to the main idea of the thread, the fetishization of modeling human behavior in the social sciences is in and of itself an ideology that has led to gross distortions in the interpretation of social behavior and phenomena in political science and economics, and has reduced many subfields to little more than attempts to count the number of angels that can dance on the head of a pin.

Absolutely, theories which are not grounded in empirical data can still add to the discussion, and parts of Austrian economics such as its business cycle theory have been a useful precursor to subsequent more accurate theories, but it should not be a guide on how to implement economic policy when there are more well grounded alternatives.
It's perfectly possible for such theories to contribute useful ideas, but it should always be strived to form them into a more solid groundwork that can have its underlying assumptions backed (and challenged) by data.

It is the specific rejection of empirical data and falsifiability that is a problem here; I know there are problems with using both in economics, but they are still useful in guiding observational development of theory, and the very idea of that is rejected for the core of the underlying theory.
Basically, rejecting empiricism allows observational data to be ignored, when it should be used to highlight flaws, to try to explain new observations consistently and reshape a theory; if you selectively ignore data, you can make a theory say whatever you want, or deliberately warp its foundations and try to make them untouchable through obfuscation and supposed empirical immunity.

KyussBishop

valmont wrote:

Again, you are asserting that libertarianism rejects the scientific method in its entirety. You have yet to explain how! You are mistaking praxeology's contesting positivism with a rejection of science entirely and ignoring any calls to expand on your position. I don't know the answer to the issue but it would be good to actually discuss it!

It (Libertarianism that is based on Austrian economics/praxeology) rejects it at a very base level in the formation of its underlying theory, reducing the credibility of the rest of the theory; data and evidence are still used to support the theory, but only after taking the base assumptions as true without proper backing.

I'm not saying there should be a positivist perfect verification of every part of the theory, just a proper application of empirical data in backing and guiding a theory; with Austrian economics, they try to claim the praxeological base as untouchable by rejecting empiricism, which is a very convenient base for rejecting criticism of its underlying assumptions.

Eliot Rosewater wrote:

Your claim was that political ideologies are ultimately subject to the scientific method. My counterclaim was that at best a subset of the beliefs of a political ideology are subject to the scientific method. Perhaps, if I'm wrong, you could tell me how the conservative and liberal views on marriage may be compared using the scientific method?

I did not generalize application of the scientific method to politics as a whole (I did say though, that it should be used all throughout politics, where possible), neither did I say it should be applied to judge purely subjective beliefs.

Eliot Rosewater wrote:

Frankly, this is just anti-intellectualism. You're saying "I'm supporting the scientific method on this thread, therefore any claims that my means of supporting it are unscientific are rubbish." That is plainly nonsense. One can support the scientific method using unscientific arguments.

The fact that I'm promoting a rigorous aspiration to scientific principles, makes your anti-intellectual accusation absurd.

You're trying to paint my point of view in a highly black and white manner, when it is not so, and trying to justify the discarding of empirical methods by showing partial limitations of them.

Eliot Rosewater wrote:

Yes - my claim is that it is probably impossible to judge the effects of one economic policy by looking at aggregate data for the whole economy of which the policy is merely a tiny component. Presenting an increasing graph of the USA's GDP for the 20th century does not provide evidence for or against the economic benefits of the minimum wage. The graph says that on an aggregate basis the economy performed positively, but it is impossible to solve the "inverse problem" and attribute that positive growth to single particular aspects. It is highly possible that the effect of the minimum wage was negative, but that the negative effects were cancelled out by other policies.

Really - isolating certain causal factors is such an essential component in the practice of hard science that I'm baffled at your dismissing such concerns as "silly".

Your entire presentation of my point of view is a straw man, as you're saying I'm demanding a fully quantitative approach to economics, when I'm saying it should aspire to the much more general approach of the scientific method, and criticizing how Austrian economics stands apart by rejecting that at its core.

I even explicitly state that there are limitations to how much that can be followed, and that the condition is not rigorous following of the scientific method where not possible but aspiration to it, so you seem intent on continually beating a straw man here.

Soldie

KyussBishop, you have not addressed any the points made by other posters; you have just continued to repeat yourself. On the one hand you claim to acknowledge the shortcomings of applying the scientific method to the social sciences. On the other hand you claim that your own approach is more valid because it incorporates the scientific method. This is inherently self-contradictory.

We can paraphrase your long-winded posts as follows: "I accept that the scientific method is not entirely applicable to the social sciences, but Austrian economics is inherently flawed because it does not apply the scientific method". This is not a valid form of debate; it's just seeking the moral high ground. Unless you take the time to try to understand and address the points made by other posters, this discussion is going nowhere.

Eliot Rosewater

KyussBishop wrote: »

I did not generalize application of the scientific method to politics as a whole (I did say though, that it should be used all throughout politics, where possible), neither did I say it should be applied to judge purely subjective beliefs.

I'm afraid KyussBishop that that's precisely what you said (my bold):

KyussBishop wrote: »

Libertarianism is free to apply to whatever philosophical principles it wants (as it does), but the credibility of all theories are typically judged based upon the principles of the scientific method, and how well they adhere to them (or how well they aspire to them, when it's not possible to adhere).

As regards the scientific method being used in politics where possible - I doubt anyone would disagree with that. We disagree on the "where possible" part.

KyussBishop wrote: »

The fact that I'm promoting a rigorous aspiration to scientific principles, makes your anti-intellectual accusation absurd.

You're trying to paint my point of view in a highly black and white manner, when it is not so, and trying to justify the discarding of empirical methods by showing partial limitations of them.

For the anti-intellectualism, hardly. If someone came onto the thread saying that they believed in the scientific method because a three-headed god appeared to them in a dream and told them to use the scientific method exclusively, would I be incorrect to accuse them of being unacademic? You seem to think that because you are arguing for a particular application of the scientific method that you have a free pass to use all kinds of flawed arguments in your defense without those flaws being pointed out to you. For instance, labeling my argument against empiricism in economics as an excuse, and then dismissing it because it was now an excuse, was clearly a very unacademic and unscientific way to conduct the debate.

As regards my criticism being "partial" -- again, I'm simply baffled. The failure of economics to be able to isolate single causal relationships in data analysis is not a "partial" limitation: it cuts right to the heart of its ability to implement the scientific method properly. As I said, this is such a central part of the application of the scientific method in the hard sciences that I'm genuinely surprised at your attempts to brush it under the carpet.

KyussBishop wrote: »

Your entire presentation of my point of view is a straw man, as you're saying I'm demanding a fully quantitative approach to economics, when I'm saying it should aspire to the much more general approach of the scientific method, and criticizing how Austrian economics stands apart by rejecting that at its core.

I don't think I've engaged in such straw-manning. The criticism of empiricism I outlined does not merely kick in when economics is 95% empirical - it is a criticism that arguably affects our ability to make economics even 5% empirical.

southsiderosie

KyussBishop wrote: »

Your entire presentation of my point of view is a straw man, as you're saying I'm demanding a fully quantitative approach to economics, when I'm saying it should aspire to the much more general approach of the scientific method, and criticizing how Austrian economics stands apart by rejecting that at its core.

I even explicitly state that there are limitations to how much that can be followed, and that the condition is not rigorous following of the scientific method where not possible but aspiration to it, so you seem intent on continually beating a straw man here.

But why should it aspire to this? And, again, how is this not ideological? You still haven't given us any real reason why this is either a valid or even logical approach in the social sciences.

In the social sciences, theory testing often reflects attempts to adhere to the scientific method, but theory generation is quite often a logical enterprise. Put differently, you seem to be demanding a deductive process, whereas when studying human behavior, an inductive process is what often generates groundbreaking theories.

Permabear

This post has been deleted.

SupaNova2

KyussBishop, I really don’t get why its so hard for you to understand why the scientific method doesn’t apply as well to economics as it does to physics or chemistry.

As has been said already by multiple posters, in economics you cannot isolate and control variables. In chemistry you can set up two test tubes containing the same substance, change one variable, let’s say temperature and observe its effects. In economics we can’t observe the US economy under policy set X and the US economy under policy set Y side by side. We would need identical parallel universes and the ability to observe them from the point we change our variables for the scientific method to be applied in the same way it is in a laboratory experiment. I don’t think it takes much common sense to acknowledge this fundamental difference.

I am reminded of a Richard Feynman documentary where he gave his opinion of the social sciences. While he probably hadn’t thought about whether the scientific method could or could not be applied to economics, he recognized that the those in the social sciences tried to mimic the hard sciences but had no sense of rigor. And being a physicist, the reason Feynman could get that feeling of truly knowing something was precisely because he could isolate and control variables and observe the effects through repeated experiment. In economics we can’t do any of that, we can’t control and isolate variables, nor can we repeat the same experiment, absent a time machine. Population, technology, available resources, weather and billions of variables all change over time.

Memnoch

I'm going to take a slightly simplistic approach here because I think the argument is spinning out of control in a battle of semantics and jargon.

Let's take the example of income taxes. We often hear the theory that reducing income taxes is the way to spur investment, growth and to create jobs. This is in fact the policy platform of right wing parties and governments around the world and there is a huge ideological debate around this.

Now lets take the following study published recently by the non-partisan congressional research office.

http://graphics8.nytimes.com/news/business/0915taxesandeconomy.pdf

This looks at the specific issue of the top rate of tax.

The results of the analysis suggest that changes over the past 65 years in the top marginal tax rate and the top capital gains tax rate do not appear correlated with economic growth. The reduction in the top tax rates appears to be uncorrelated with saving, investment, and productivity growth. The top tax rates appear to have little or no relation to the size of the economic pie.

However, the top tax rate reductions appear to be associated with the increasing concentration of income at the top of the income distribution.

(Emphasis mine)

So it would seem that the theory that lowering taxes to benefit to the economy, job creation etc has been shown to have not worked in the US by probably the most comprehensive analysis of data we have seen thus far on the issue.

My understanding of the argument in this thread is that we should simply ignore the empericism of this study and still entertain the tax reduction theory on the basis of what, faith alone? Because you said so?

To me it is just common sense. When a theory makes certain claims, regardless of the field of study it is coming from. If those claims are supported by evidence they are stronger, if they are not, they are weaker.

Another example is that of regulation and the glass-steghal act in the United States. After the Depression we had regulation of the banking system and it seemed to work. Return to the late 90s-2000s and gradual deregulation allowed banks to engage in the same dangerous practices that caused the depression in the 1920s and lo and behold, we get the biggest worldwide recession in a long time.

In this case the THEORY that if you deregulate businesses (in this case the financial industry) to the point of giving them complete freedom to do whatever they want then they will act in a manner that is destructive to the general economy/public in the pursuit of short term greed. This has been shown to occur in the real world. It is borne out by the predatory practices of corporations in countries where regulation is weak/corrupt and not properly enforced.

Though if you listen to some libertarians you'll hear the completely unsupported conspiracy theory that the problem was not the lack of regulation but of non-economic politicians interfering in the financial process, with no evidence to support their theory.

In conclusion: This whole idea of science not being applicable smacks to me as being parallel of religious fundamentalism. I.E. We want to create a set of laws that will influence and affect the lives of others. We base these laws on theories which are based on OUR beliefs, but for which we are not willing to supply evidence or scientifically testable theory. Because science does not apply to OUR beliefs. (Oh and please don't notice the fact that the application of these theories seem to hugely benefit us even while they may be to the detriment of yourself and many others, that has nothing to do with it all, it's just economies you see, you can't apply doggone science here!)

I don't buy it. If you want to propagate your ideas and convince me they have merit, then talking about theory is fine up to a point, but I need to see some evidence somewhere down the line before I'm going to buy in and start praying to your god. Using complicated jargon and pointing to prophets who may sound erudite isn't going to be enough.

As far as the specific example of libertarianism is concerned. I think it's an extreme economic theory and I think the proponents know it, but like a religious cult they are so committed to their beliefs that they don't care if there is no evidence to support their ideas being practical or applicable in the real world (even though they will oppose many other solutions proposed to economic woes on the grounds that they aren't practical). Because THEIR IDEAS are not testable scientifically and therefore don't need to be so.

SupaNova2

My understanding of the argument in this thread is that we should simply ignore the empericism of this study and still entertain the tax reduction theory on the basis of what, faith alone? Because you said so?

No you have missed the point being made, how does the study deal with possibility that while tax was lowered, investment and growth may not have improved because of other variables discouraging investment such as increasing regulations.

Another example is that of regulation and the glass-steghal act in the United States. After the Depression we had regulation of the banking system and it seemed to work. Return to the late 90s-2000s and gradual deregulation allowed banks to engage in the same dangerous practices that caused the depression in the 1920s and lo and behold, we get the biggest worldwide recession in a long time.

Again you suffer from the same problem, yes you can take a single variable of your choosing, and see that it may correlate, but why choose this one over low interest rates and a host of interventions to encourage investment in construction over any other industry? How is this scientific, taking what ever variable you want and ignoring others, this was Feynman's point when talking about the social sciences. You don't understand the scientific method and its application if you think its ok to pick out whatever correlation suits you and assume causation. I recommend you watch this:
http://www.youtube.com/watch?v=y1RXvBveht0

their beliefs that they don't care if there is no evidence to support their ideas being practical or applicable in the real world (even though they will oppose many other solutions proposed to economic woes on the grounds that they aren't practical). Because THEIR IDEAS are not testable scientifically and therefore don't need to be so.

Well there is evidence to support their theory. Austrian's have found evidence to support their theory as far back as Tulip Mania. Rothbard has a book full of evidence that supports the Austrian story of the Great Depression. Austrian's have more than enough evidence from the current Recession/Depression to support their theory also. The problem is, Keynsians are looking at the exact same data interpreting it through a different lens and saying it supports their theory.

In the US the free marketer's will say we have had unprecedented intervention and stimulus and the economy is stagnant, what more evidence do you need. Keynesian's will say the intervention wasn't big enough and things would be worse without it. How does the data falsify either?

20Cent

The last few posts demonstrate how right Clintons quote was. Sure why do you need empirical evidence for something we don't need evidence God the markets will sort it all out.

Someone with an answer for everything, advocates changes that have never been tested never mind tried, and continuously presents dishonest evidence is not someone to be taken seriously.

That goes for all ideologues not just libertarians.

Soldie

20Cent wrote: »

The last few posts demonstrate how right Clintons quote was. Sure why do you need empirical evidence for something we don't need evidence God the markets will sort it all out.

Someone with an answer for everything, advocates changes that have never been tested never mind tried, and continuously presents dishonest evidence is not someone to be taken seriously.

That goes for all ideologues not just libertarians.

It's great to see you engaging with the debate.

Duggys Housemate

The idea that libertarianism is empirically based is ludicrous of course. I gave up on the Libertarianism in an Irish context thread because it never really answered the general questions - would everybody go to school, where has this worked? The questions were batted away, the "proof" of failure was that the present system was not perfect - cherry picked data produced. I pointed out that proving flaws in the existing system is not the same as proving your own system, and the ideologues could not handle this argument. ( it's like a homeopath saying chemotherapy is generally not a perfect system so we should use homeopathy. The first claim - even if proven - does not prove that all, or any alternative systems would work).

The Austrians have never explained their position either.

The empirical evidence supports the idea that the most productive capitalist - or any - societies in world history were Europe and America 1945-1970, a time of increasing state involvement and increased welfare - liberating people to consume.

This puts you in the middle ground but that's where the empirical evidence lies against the two competing radical theories - Marxism and libertarianism - both at their hearts simple theories - the State controls everything or nothing. One has failed but it remains an unfalsifiable theory because it's adherents say its hasnt been tried properly, or in the absence of capitalism; the other has never been applied - except for, perhaps, Haiti or Liberia - so it is by definition unfalsifiable.

I reject the idea that all ideas are ideologies. Science isn't - and we're economics a proper science the Marxists and the Libertarians would wither away like steady state theorists. It isnt, so they don't.

southsiderosie

20Cent wrote: »

"if you've got an ideology, you've already got your mind made up. You know all the answers and that makes evidence irrelevant and arguments a waste of time".

20Cent wrote: »

The last few posts demonstrate how right Clintons quote was. Sure why do you need empirical evidence for something we don't need evidence God the markets will sort it all out.

Someone with an answer for everything, advocates changes that have never been tested never mind tried, and continuously presents dishonest evidence is not someone to be taken seriously.

That goes for all ideologues not just libertarians.

20Cent, how exactly do you 'test change' in the social science? You can't run simulations of riots in STATA.

Also, can you please point to ONE school of thought in ANY branch of the social sciences that is universally accepted, and can subsequently predict human behavior or events at least 95% of the time (i.e. to a point where it is statistically significant)?

Finally, please explain to us how the anti-libertarians/pro-falsifiability in the social sciences folks on this thread are not behaving in a manner that is totally consistent with being an ideologue, as defined on this thread?

Duggys Housemate

I can take that. :-)

Social science is not science - and this includes Economics, with the exception of some trivial cases. All of this means that we have to muddle along. If Keynsianim seems to work, use it. When it goes too far try some market discipline, when that goes too far - and the banks become too powerful and the banks fail - try something else. The algorithm is a learning algorithm , we could nationalise all industry or privatise the state - in absence of any real evidence that these radical solutions ever work that would be madness.

Eliot Rosewater

Duggys Housemate wrote: »

The empirical evidence supports the idea that the most productive capitalist - or any - societies in world history were Europe and America 1945-1970, a time of increasing state involvement and increased welfare - liberating people to consume.

I'm surprised that people find the "correlation does not imply causation" point so difficult to appreciate. It could be the case that large economic growth was caused by other elements such as population increases and technology developments, and that in the absence of the increased state involvement economic growth would have been even higher. Economic growth is the sum of thousands and thousands of factors; its is mathematically impossible to "reverse the sum" and claim that particular factors were positive.

I find it ironic, because cherry picking correlations and claiming they were the causes of good things has no logical or scientific justification -- and yet people who argue against this line of thought are precisely the ones labelled "ideologues".

KyussBishop

Going to respond with one post rather than quote-for-quote with all those responding to me, so excuse the inevitable parts that I miss replying to (and excuse the length also):
Ok to clear up, my arguments have made it seem like I am arguing that economics should try to become like the natural sciences, in making 'x-y-z' directly experimentally testable statements that return the same result every time, to have a perfect deductive consistency as southsiderosie said.

This is not what I'm arguing, though yes my past arguments have read exactly like that despite my (many) attempts to clarify what I was saying; so backtracking: I'm not saying economics can become a natural science, one that can be formulated through deduction, but I am saying that it should strive to apply tools from the scientific method (including use of empirical evidence, and experimentation where possible) and should be judged by the rigor with which it applies that (again, I'm not insisting on perfect application where it is not possible).

This will necessarily mean discarding tools that can only be applied to a deductive theory, but striving to make as full use of the remaining tools as possible (also, my arguments are not about how to generate theory, but on rigorously backing it and testing it).


One of the most important parts of that set of tools is falsifiability (I'm probably going to have a hard time putting this clearly, as I don't have a lot of time to spend at writing this); necessarily, not all behavioural statements (which a lot of economics depends upon) can be made into easily falsifiable statements, and not all of them can be backed with empirical evidence.
This is a necessary thing to accept with the social sciences, simply as a practical matter for being able to develop theories which have to be based on concepts that are not supportable in a deductive way; praxeology goes some steps further than this though, and states that empirical evidence does not apply to it, thus it can't be falsified with empirical evidence.

This might be true for individual parts of praxeology (who can falsify the near-useless term 'humans act'?), but praxeology is a whole framework of claims, multiple levels of assumptions, and the higher that framework goes without empirical backing or accepting falsifiability, the more precarious it is and (due to that) the less credible it is.
That is the key point of criticism, that it claims that empirical evidence can not be used to falsify it, thus if there is evidence that is contradictory to it, that can be used as a justification to ignore that evidence.

That is why it loses credibility, it takes that extra step in saying empirical evidence cannot be applied to it, but that is false; mainstream/neoclassical economics also depends upon some similarly rickety frameworks, but (as far as I know) does not take that extra step in claiming empirical evidence cannot apply to it.
Note the very important distinction here, that I am not saying that theories need to be built up in a deductive way according to the scientific method (which seems to be the primary impression people are getting from my posts, which is fair enough as I didn't make myself clear enough, with my statements being contradictory); I'm dealing with a very specific aspect here, which is not as general as people are thinking.



To give an example of where empirical evidence has in the past applied to such a framework, you can look at a part of the Behaviourist school of Psychology, which made the assumption that you can only understand peoples behaviour as a response to external stimulus (abstracting away many internal cognitive processes); on the face of it, this is an observation on human behaviour which doesn't seem easy to experimentally verify/discredit.

This assumption was shown to be in conflict with many experiments though (and effectively disproven); a study on piano players and the speed with which they play, showed that for good players their fingers moved faster than the time needed to mentally process the sensory results of each action, and then deliver a response based upon that (i.e. each keystroke could not be a reaction to the sensory input from the previous keystroke).


Another example would be Monetarism and the claims it makes, such that inflation is caused by the money supply growing faster than the economy; this was effectively disproven by Thatcher's attempts to utilize this idea to shape economic policy, yet you can't say "that didn't prove anything, because it wasn't a controlled experiment"; Monetarism (rightly) lost credibility as a result of this.

I think credibility is the key point here; the falsification of Monetarism is not a perfect falsification (such as disproving a mathematical hypothesis in deductive terms), but it is certainly more than enough to kill its credibility.
It should be noted that neoclassical/mainstream economics (and well...economics as a whole, almost) is highly subject to this kind of problem in judging the credibility of theories and their underlying assumptions.

The solidity of the underlying assumptions of any theory are extremely important, as well as its ability to make falsifiable claims (that does not mean every stage has to make falsifiable claims though), and if empirical evidence can be found which contradicts those underlying assumptions/claims, it is not credible to just say that empirical evidence in general does not apply, and should not impact the credibility of the underlying theory.


I know I'm repeating a lot of this as like my previous posts (hopefully it won't spin the discussion around the same roadblock, am mainly trying to clarify what I was saying), but I've done my best to resolve the main complaint against my posts; my primary argument is separate to that main complaint, and has not been addressed yet (mainly due to the confusion about my posts, which admittedly I've not been clear enough about).

Duggys Housemate

Eliot Rosewater wrote: »

I'm surprised that people find the "correlation does not imply causation" point so difficult to appreciate. It could be the case that large economic growth was caused by other elements such as population increases and technology developments, and that in the absence of the increased state involvement economic growth would have been even higher. Economic growth is the sum of thousands and thousands of factors; its is mathematically impossible to "reverse the sum" and claim that particular factors were positive.

I find it ironic, because cherry picking correlations and claiming they were the causes of good things has no logical or scientific justification -- and yet people who argue against this line of thought are precisely the ones labelled "ideologues".

I think we all understand that correlation is not causation. This doesnt prove your argument - you have to disprove the correlation, and prove the other factors which may have caused the effect , in this case economic growth. I could have argued instead - were the ideologues on this thread Marxists - that the economic growth of the West in the same time period disproved Marxism. Would we accept that the Marxists could then argue that "causation is not correlation". Would you? Or do you think they should

A) explain why the economic growth happened
prove their economic theory.

The same applies to you.

Eliot Rosewater

Duggys Housemate wrote: »

I think we all understand that correlation is not causation. This doesnt prove your argument - you have to disprove the correlation, and prove the other factors which may have caused the effect , in this case economic growth.

One cannot disprove a correlation, if the mathematics is tight enough. I think, perhaps, you mean disprove the causation, in which case your statement is not right. The onus is on the person making a claim to support it, not (initially) for those on the receiving end to "disprove" it. Otherwise we could go around making statements like "there are teapots orbiting Saturn", happy in the knowledge that no one could specifically disprove that sentence.

Duggys Housemate wrote: »

I could have argued instead - were the ideologues on this thread Marxists - that the economic growth of the West in the same time period disproved Marxism. Would we accept that the Marxists could then argue that "causation is not correlation". Would you?

In general, yes. Though the empirical evidence is quite more useful when discussing Marxism, given that the period contained an overall Marxist area and an overall non-Marxist area. The problem of reversing the data doesn't quite occur, because the economic performance of the US is roughly the sum of "market things" and the economic performance of the USSR "non-market things" - allowing us to make a "market" versus "non-market" analysis. It would be the same if there was a country run strictly on Austrian economic principles - we could then look at the data to see how it was performing relative to, say, Ireland.

Duggys Housemate wrote: »

Or do you think they should

A) explain why the economic growth happened
prove their economic theory.

The same applies to you.

They should do this as well. Any mention of empirical data should be accompanied by a causal theory. It's no good saying "the Glass-Steagall act was partially repealed before the recession, and hence this repeal caused the recession". Similarly, it's no good saying something like "the decline in piracy is correlated with the rise in global temperatures, therefore the relative absence of piracy is heating the planet". Then, when you've developed a causal theory you have to deal with the fact that others are going to develop their own causal theories.

20Cent

southsiderosie wrote: »

20Cent, how exactly do you 'test change' in the social science? You can't run simulations of riots in STATA.

You can test, compare and analyse lots of things. Its not physics and we don't get 100% proofs but things can be modeled and examined. Compare a situation before and after some action, compare a country with tight financial regulation to one with little or none. We can measure the consequences of various policies to a useful degree. Life expectancy, exports, loads of stats are produced daily which can be used to make rational decisions. I'd like to see solutions shown to have worked in the real world rather than ideas deduced from some preconceived axioms.

southsiderosie wrote: »

Also, can you please point to ONE school of thought in ANY branch of the social sciences that is universally accepted, and can subsequently predict human behavior or events at least 95% of the time (i.e. to a point where it is statistically significant)?

Where does some data become useful? 95% certainty, 50% 40% 10%? If there is a 20% chance your car breaks will fail do you keep driving until it reaches 50 or 95%?

southsiderosie wrote: »

Finally, please explain to us how the anti-libertarians/pro-falsifiability in the social sciences folks on this thread are not behaving in a manner that is totally consistent with being an ideologue, as defined on this thread?

Wanting some kind of evidence is an ideology? If you say so, whats it called?
I'm not talking about just libertarians anyway I think all ideologies inhibit free thought.

Permabear

This post has been deleted.

SupaNova2

Duggys Housemate wrote: »

Would we accept that the Marxists could then argue that "causation is not correlation".

Marxists could try argue that, not sure they would convince anyone, since the closest we have ever had to real life test tubes side-by-side are South Vs North Korea, East vs West Germany. The fact is that in both cases you had populations and geographical areas so similar in terms of technological capability, culture, available resources etc, but had drastically different results when economically and politically they took radically divergent paths.

Permabear

This post has been deleted.

southsiderosie

20Cent wrote: »

You can test, compare and analyse lots of things. Its not physics and we don't get 100% proofs but things can be modeled and examined. Compare a situation before and after some action, compare a country with tight financial regulation to one with little or none. We can measure the consequences of various policies to a useful degree. Life expectancy, exports, loads of stats are produced daily which can be used to make rational decisions. I'd like to see solutions shown to have worked in the real world rather than ideas deduced from some preconceived axioms.

But whether or not you think something has 'worked' is in and of itself a value judgement. For example, is having a high level of exports always a good thing? Not everyone would agree that it is.

It is also ironic that you keep asking for real world examples - not only have you not been willing to come up with any when explicitly asked for some, but you have pooh-poohed everyone else's on this thread. Why even bother posting if you are not going to actually take other posters' responses to you on board?

20Cent wrote: »

Where does some data become useful? 95% certainty, 50% 40% 10%? If there is a 20% chance your car breaks will fail do you keep driving until it reaches 50 or 95%?

This is getting ridiculous. On the one hand, you claim that you want people to take advantage of data, but on the other hand, you don't want to hold that analysis to any kind of standard - 95% certainty is the benchmark for statistical testing. Perhaps I shouldn't be surprised, since your earlier comment about 'examples' suggests that you are just interested in cherry-picking to confirm your existing biases anyway.

20Cent wrote: »

Wanting some kind of evidence is an ideology? If you say so, whats it called?
I'm not talking about just libertarians anyway I think all ideologies inhibit free thought.

This all seems to be a game of semantics. The overarching narrative here reads as ideology = beliefs I disagree with and values = guiding principles that shape my behavior (and allow me to feel superior to those dreadful ideologues).

Dr Galen

20Cent wrote: »

You can test, compare and analyse lots of things. Its not physics and we don't get 100% proofs but things can be modeled and examined. Compare a situation before and after some action, compare a country with tight financial regulation to one with little or none. We can measure the consequences of various policies to a useful degree. Life expectancy, exports, loads of stats are produced daily which can be used to make rational decisions. I'd like to see solutions shown to have worked in the real world rather than ideas deduced from some preconceived axioms.


Where does some data become useful? 95% certainty, 50% 40% 10%? If there is a 20% chance your car breaks will fail do you keep driving until it reaches 50 or 95%?



Wanting some kind of evidence is an ideology? If you say so, whats it called?
I'm not talking about just libertarians anyway I think all ideologies inhibit free thought.

Not for the first time, you are going to be reminded that this is the Political Theory sub-forum.

Really, when I read your posts, I see a lot of noise and not a lot of signal tbh. Other posters on all sides are making a pretty great stab at having a proper discussion here. You are going to need to step up to that plate or leave the thread.

Cheers

DrG