All Episodes
May 30, 2009 - Freedomain Radio - Stefan Molyneux
34:10
1372 Austrian Economics Part 2 (from 1368)

Knowledge, truth, and the efficacy of philosophy.

| Copy link to current segment

Time Text
Hey folkies, it's Steph.
It's May the 30th, 2009.
I am out raking the lawn for the delicious barbecue folk who will be coming by.
And I had a dream last night about a podcast topic, because that is how helpful my unconscious is.
I had a dream about a podcast topic, which has been going on, about a fellow who's asking about Inflation rates and interest and Fed printing in the 1920s.
Oh dear God, it's gripping semi-erotic stuff.
And this is what I mean by a pleasant, enjoyable, productive intellectual exchange.
So he wrote back in his response.
I think it's 1371.
I enjoyed the response.
Always great to listen to another economics podcast.
Always great to do one as ably as I can, which is not to say perfectly ably at all.
At several points you said that since the downward trend at the price level was stabilized in the 1920s, that this was evidence for inflation.
But once again, a statement like this cannot be used in the field of economics because there are way too many possibilities, even in the most primitive economy, for a human to comprehend.
Ah, but what about a superhuman?
For instance, while it may be perfectly true that the stabilization was due to artificial inflation, which it was in this case, but I'm just using the 1920s example to get my point across, It may also be true that the price stabilization was due totally or in part to a decrease in productivity or a decrease in the demand for money.
A hurricane could have conceivably destroyed a large portion of the capital inventory of a given area, thus decreasing the overall productivity and halting the trend of falling prices.
My point is that the mere fact that the price level flattened does not in and of itself imply anything significant.
That said, even if the Fed printed a significant amount of fiat money and the price level declined, we cannot abandon a priori theory as empiricism would have us do.
We know, as you pointed out in your student example, that logically, and all other things being equal, monetary inflation causes prices to rise.
But all other things are never equal.
With that knowledge, we can logically deduce that either productivity increased drastically, perhaps due to some revolutionary invention, that prices fell even after the inflation Or that the demand for money skyrocketed for some reason and or something was preventing the new money from percolating through the economy, banks refusing to lend for whatever statist reason, etc.
This would be an example of empirical reasoning failing to properly interpret and invent, especially if this was the first instance in history that this occurred, i.e.
there would be no prior instances to point to for evidence.
Therefore, the Austrians wouldn't have to reconsider anything if such an event occurred.
I'm not holding my breath that it would, though.
I look forward to your responses. Well, this is delicious, juicy, wonderful brain food.
Thank you so much for posting it.
And I'm going to go on a ramble tangent, which I think will have something to do with this, but will have broader applications with any luck, too, as philosophers in their best approaches are wont to do.
So, the first thing that I would notice here is that you are saying there are other possibilities for prices no longer falling for price stabilization other than an increase in money supply.
Now, everything that you talk about here is empirical.
Something which can be tested.
So a hurricane could have conceivably destroyed a large portion of the capital inventory, blah, blah, blah.
Well, was there a hurricane? No.
So we can eliminate that.
Were banks just up and stopping, refusing to lend credit?
Well, that wasn't the case. And if it was, then it was a statist reason.
It was guns preventing them from doing so.
Which, again, we're not against the state because it prints fiat money.
We're against the state because it initiates the use of force, and that would be another instance of that.
Some revolutionary invention, productivity increased drastically.
That would actually cause deflation, in my understanding.
But of course, there was no revolutionary invention.
Demand for money skyrocketed for some reason.
Well, you can figure that out by looking at interest rates, right?
Because if demand for money increases, then interest rates will go up, and you will also have a concurrent increase in productivity.
Purchases of capital inventory and hiring and so on.
So everything that you're talking about here can be tested.
And so the question then becomes, at what point can you eliminate a large enough number of variables to come to a fairly certain conclusion?
And in the realm of empiricism, there is no perfect observation, right?
The observer... Anything that you use to observe is going to alter at some level that which you are observing.
That's a pretty fundamental principle, and I'm not telling you anything you don't know.
I'm just going to try and apply it in a way.
And just thank you again.
This is a very delicious brain food, so I'm sure the listeners have been very positive about this debate, so thank you for keeping it up.
But even in the realm of physics, right, certainly down at the level of quarks and quantum mechanics and so on, the observation alters the outcome.
If you set up a camera to record something, that camera has a gravitational pull, which will alter the outcome.
Again, I mean, I know that these are pretty small bits of gravity wells, but everything that we observe is tainted, so to speak, and it's not really the right word, but I couldn't think of a better one, is tainted by the observation.
This is true in In very extreme cases of confirmation bias, you know, people are just going out to look for things which, arguments or evidences which support their prior prejudices, right? So, I don't know, some race this is only going to watch, you know, Boys in the Hood and Juice and all that kind of stuff to confirm his racial stereotypes and watch...
I don't know, rap videos and all that kind of stuff.
And a black racist might only watch, oh, I don't know, Trailer Park Boys and My Name Is Earl and all that kind of stuff in order to confirm his prejudices or stereotypes about.
About whites. So, in that kind of confirmation bias, and Lord knows, philosophers, anarchists, libertarians, Ron-Palestinians are not immune to any of this, and certainly neither am I. It's something I always have to keep an eye on, which is why I do try to read things which are highly opposed to my viewpoint, like the trial and death of Socrates, and try to understand them, and debate with people of very differing beliefs.
And so there is the open and obvious, well, not obvious to the people who are experiencing it, but the open confirmation bias, which is something we have to keep an eye on.
But even in the realm of physics, there's no perfect observation of anything.
But there is sort of, within the limitations, good enough, right?
And that is something that is okay, right?
It's okay. And...
And so there's probability.
It's a small enough deviation, right?
So every navigational system in an airplane has flaws, right?
I mean, there's going to be some particular flaws in barometric or altitude or whatever.
There's going to be some level or set of flaws.
And the question is, is it good enough, right?
As I've mentioned before, it's why sailors use Newtonian physics and not Einsteinian physics to...
Get across the ocean, because the variance is so small that it's irrelevant to the task at hand.
Knowledge is not a platonic abstract that exists in a perfect vacuum of which any taint with the material causes it to be utterly flawed.
It's not like knowledge is baby food and empiricism is arsenic.
There is no way that the twain can meet in any reasonable way without causing The destruction of the baby food, right?
So, the purpose of knowledge is efficacy, right?
I mean, the purpose of knowledge is efficacy because efficacy brings control and control influence.
Competence brings happiness and so on, right?
So, the purpose of knowledge in my formulation is efficacy.
And efficacy is not a standard of perfection, right?
There's that old joke about pilots, you know, any...
Any landing you can walk away from is a good landing.
That's obviously a pretty extreme way of putting it, but there is some truth.
Every landing could be better. Every landing could be softer.
Every GPS could be more accurate.
My GPS refreshes a couple of times a second.
I'm sure you could write and code and launch satellites and get a GPS that did 30 frames a second.
Would that be an improvement?
Well, I would say not, myself.
Because the amount of resources you would devote to creating that would not result in any particularly improved Way of getting you to the grocery store and back, right? So again, I'm sorry about the lengthy diatribe here.
It's not really a diatribe, but to me, the knowledge peaks, right?
It's a bell curve, right? There's too little knowledge, right?
Like, I don't even have a compass, so I can't navigate, right?
And then there is...
As you get a rudimentary compass and you get the ability to navigate, then your efficacy goes up.
And then... What happens is if you continue to refine your compass, then your efficacy actually goes down because you're spending more time building the compass than actually navigating.
And for those of us who have the geeky, pallor, propeller head lifestyle, there's that old Dilbert cartoon, right?
Oh, I'm bored. I guess I'll just mess with my computer settings until something breaks and then spend some time fixing it, right?
Oh, we've all been there.
Hey, look, a new registry cleaner.
Kapow, right? But that's, of course, just developing skills in another area.
So, all of that having been said, yes, there is no perfect knowledge in this world, for sure, of course not.
But the purpose of knowledge is not perfection, right?
The purpose of knowledge is efficacy, and an excess of perfection is negative to the efficacy that knowledge is aiming to get at, right?
To take another silly example, the purpose of...
Medicine is not perfect health at all.
Perfect health would be a prison sentence, right?
Because perfect health would be that we would all live or that the doctors would prescribe to us isolation tanks, right?
So with everything sterilized and no possibility of getting an infection or a cold or anything like that because we would never leave the boy-in-the-bubble scenario, right?
So that would be an example of perfect health.
And that would be a prison, right?
That would be something which we would all view as a particular nightmare, because in order to live, we have to roam around, right?
I mean, I could ensure, or I could make it much more likely or probable, that Isabella would not get a cold this weekend by not having anybody visit.
But I want people to visit.
I like it when people visit. It's great to see everyone.
And so there is a balance, right?
And even in some areas, perfect health is a compromise, right?
So exercise is wear and tear on the joints, but it strengthens the bones, right?
So which is better, which is worse?
There's no possibility of perfection in the realm of Medicine and so on.
And in the realm of mathematics, sure, but once you move mathematics into the realm of engineering, which is where it gets really efficacious, then you get all these kinds of compromises and ethics and mistakes.
I mean, just to give one more example, I'll move on.
I hope this makes sense. In the realm of engineering, there's a phenomenon called over-engineering, which I'm sure you've heard of, where You build a bridge with such an enormous amount of redundancy that it's really, really expensive.
The reason that knowledge to be applied has this category called efficacious because, as we know, desires are limitless and resources are finite.
Therefore, everything we gain is something we lose.
I'm breaking here and doing a podcast and not doing anything else at all.
I'm giving up everything else just to talk to you.
That's How guilty and manipulative I am.
So, if you overbuild the bridge, right, you got a little footbridge, and you make it out of reinforced steel and concrete and so on, right, rather than just, I don't know, some wood and some iron, then, yeah, it's not going to fall down, but it's really expensive, and so on, right?
So, There is an ideal, of course, if you build it out of balsa wood, it's going to fall down first time someone steps on it, right?
So then you've lost the labor that you put into building the bridge because it just falls down.
And if you overbuild it, then you lose the resources that you could have applied to other things in the excess of safety, right?
Which is a deficiency of safety in whatever else you could use the concrete for that would be more appropriate to its use, right?
So, overbuilding is not quite as bad as underbuilding, but it's certainly not as good building to the right degree.
Now, what is building to the right degree?
Well, it's a rule of thumb, and there's some rules that you can put in there and so on, but basically it's just, you know, well, this is strong enough, it's, you know, double the requirements, you know, if there's a minor earthquake or whatever is reasonable to the region, it will still stand, and You sort of understand, right? So... But even still, it might fall down, right?
Or will fall down in time.
How long do we build a bridge for?
100 years? 200 years?
Who knows, right? And even if we come up with a standard that we all agree on is perfect, there's no way to actually perfectly implement that standard.
There's defects in the materials and all that kind of stuff, right?
Just innate, inherent defects in the materials, right?
So, I think you understand that there's, you know, good enough to a reasonable level of efficacy, given the constraints of other requirements...
is a pretty solid set of criteria for design.
Right now, why am I talking about design and engineering?
Well, because they're on my mind and I twist everything to that end.
Well, because when we're looking at a truth statement, like excess fiat currency printing caused The stagnation of prices relative to their former decline in the 1920s.
Well, we're making a claim, for sure.
Can we prove that claim perfectly?
All other things being equal?
Well, you're right, of course. All other things are never equal.
But that's not just limited to philosophy.
Sorry, that's not just limited to economics.
We think there's this magical divide between the sciences and the humanities, but it is a difference of degree and not of kind, in my opinion.
And you realize that I'm not saying any of this stuff is proven, it's just the thoughts that I have about it.
And Lord knows, I've had to deal with some challenges of imperfection in my podcasting, right?
I mean, I did, just as an experiment, just to see if anyone would notice, the series that I did recently on political parties, the Republicans, the Greens, and the Democrats.
Just out of curiosity, I took the Democrat one.
If you haven't listened to it, or if you listen to it again, you'll notice that there are almost no verbal tics, pauses, corrections, ums, errs, breaths, or anything like that in that podcast, because I spent a couple of hours...
Just to see if anyone would notice or care pulling out all of the extraneous verbal tics that so infest these little nuggets of whatever.
And nobody noticed.
And so that to me was a test, right?
So, 45-minute podcast, I can spend a couple of hours fixing it up, but no one notices, so that level of quality is not required.
So instead... Of fixing up the podcast, I will work to try and eliminate the purple ticks in the podcasting that I do, which is what I'm trying to do here.
So anyway, the question then becomes, what level of knowledge is good enough?
Good enough is efficacious, right?
Because once we start to require too many, right?
I mean, there could be a silly thesis, right?
That invisible space aliens were...
We're duplicating currency in their spaceship, and it didn't have anything to do with the Fed, right?
And I don't know, I guess you could take some sort of radiation monitor to all the places you think that they may have landed and see if you could discover traces of radiation and look in the background of pictures of any photographs and look for the spaceships, or I guess they would be invisible, but you know what I mean.
You could come up with some thesis that would justify Even if you couldn't find any additional Fed printing, you'd say, well, but space aliens were duplicating it, or a counterfeiting ring was going that nobody knew about, and they were perfectly counterfeiting bills, and they were never found, and that's how the deflation was prevented.
You could come up with all this kind of stuff, right?
But at some point, there becomes a reasonable level of effort versus reward, right?
And part of, of course, part of the efficacy of knowledge, I sort of mentioned that, the efficacy of knowledge is not just relative to reality, right?
Building a bridge that stands up in the realm of physics.
It's also relative to people.
Because if you have knowledge, but nobody believes you, like if I just screamed in a shrill way, well, even more shrill way, these podcasts in a barely decipherable Glaswegian accent, then the knowledge that I had gained over the past quarter-century-odd, Would kind of be for nothing, right?
Because people wouldn't listen.
Or if I did the podcast but never published them, what would be the efficacy of my knowledge?
Well, it might be useful for me, but it wouldn't be useful in, say, begging for donations to scrimp by and be an itinerant, though, rather not moving philosopher.
So the efficacy of knowledge has also to do with Your ability to get people to believe it, right?
And that's one of the reasons why empiricism is so important, because it's hard to argue with the evidence.
So, as I mentioned before, the guy who invented Ethernet, you know, became very rich.
And he then became a professor of computer science or something.
And he would have his students, his grad students, I think, over to his magnificent mansion in His students were all always saying the same thing, like, wow, what a fantastic mansion.
I wish I had invented Ethernet, or I hope I invent something as important as Ethernet so I can make all this money.
And he's like, are you kidding me?
It wasn't the invention of Ethernet that got me the house, right?
I mean, there were tons of guys who invented networking standards back then, back in the day, right?
But what got me the house...
Was the fact that I spent 10 years writing papers, delivering speeches, rolling up and down the country, conferences, you know, creating white papers, speaking at whatever industry events I could get my hands on, being an entrepreneur, hiring the right people, advertising, getting people to test drive, you know, again, I don't have to go through the whole list, but he said, that's why I've got the mansion.
Not because I invented Ethernet, because the invention is not efficacy.
It has to be communicated in an effective way for the efficacy to really kick in, I think.
Or at least if that is part of your efficacy, right?
If that is required, right? My efficacy was to make A living of sorts as a philosopher, and that required that I communicated in a way that people wanted to listen to.
And that took me on a pretty wild journey of discovering what people actually wanted to talk about in the realm of philosophy, which was themselves, which I quite agree with.
That's efficacy, right? It's Aristotelian, not Plutonic knowledge.
So, all other things being equal, you're absolutely right.
Absolutely right. But, There's still a level of probability we apply to all other things being equal.
And knowledge becomes good enough.
Because efficacy is not perfect.
To look for heaven is to live here in hell, as Shakespeare says.
The pursuit of perfection in the realm of knowledge is innately imperfect.
In fact, it's anti-efficacy.
Because perfect knowledge, that's the Requirement for those who empirically argue against the existence of God, saying, well, you haven't looked everywhere in the universe, and therefore, therefore...
But that's a ridiculous standard.
That's not what we do with anything, right?
We assume a universal law of physics and all that kind of stuff, right?
You don't get to say, the law of numbers changes on my math exam, so 2 plus 2 equals 5.
I mean, you can say it, but people would just think you were nutty, right?
So, yeah, in the realm of empiricism, there are always other factors, but to what degree are those factors valid or relevant, and to what degree do they pass the criterion of efficacy?
So, to get to the dream I had, because I know that's what everybody's dying to hear, not these arguments, to get to the dream that I had...
I dreamt I was being lectured by me, which is...
It wasn't actually...
I can get a sense of what it's like for you guys.
It's actually kind of fun. Stimulating.
Exciting. Anyway, so...
And the lecture went something like this in my dream last night.
He said, so...
If I am a doctor who does autopsies, and I find high levels of strychnine in a body, then I'm going to assume that strychnine...
It's what killed said dude, right?
Now, it could also be...
And then, so I'm going to say, well, then it's murder and strychnine is not naturally occurring in the body.
It has to be administered, blah, blah, blah, blah, blah, right?
Now, what I could also do is I could check to see if there was any potential cancerous that might have killed the guy relatively soon.
I could also try to figure out...
Whether any invisible lasers had fried deep areas of his brain, because, you know, I guess that's vaguely possible, or microwaves.
I could try and find out, I don't know, if he'd swallowed any glass shards or if he had heart damage from some ailment.
I could look into all other possible things that could have killed him or might have killed him in the very near future, right?
It's hard to prosecute a guy for murder if he shoots a guy who's just jumped off the Empire State Building, right?
Because the guy's just going to die.
You shoot him in midair, it's kind of hard to prosecute you for murder, right?
So you could look into absolutely everything.
You could also do an extensive diagnostic of his family history, look at the probabilities of ailments, try and figure out, you know...
Again, what might have killed him, what might have killed him, what might have killed him, and that's before you're just going into the bazaar, right?
Like, maybe he has an undetectable form of vampirism, and you need to start doing research into that, right?
Or maybe, oh, I don't know, you sort of get the idea, right?
You can start looking into every cause, right?
Yet still, we feel comfortable saying, Strychnine killed the guy.
And the fact that he was about to die from other causes, or maybe, maybe not, it doesn't matter.
Strychnine killed the guy, let's go ahead with the murder prosecution.
Is Strychnine killed the guy a perfect statement?
Well, of course not. Because there could be other things, right?
So, that was sort of the lecture.
And even if we say, Strychnine killed the guy, we can't say exactly how.
Down to the cellular level, right?
We can't say, well, he took the strychnine, and, you know, of course, I don't know what the hell strychnine does to a body, right?
But we can't say he swallowed strychnine, or was given strychnine, and this happened to this first cell, and then this second cell, which escalated, because all we have is the smoking body, right?
So... Even if you accept Strychnine killed him, you can't say exactly how.
To an infinite level of detail, at the cellular or even atomic level, you can't see exactly what dominoes fell over to cause his death, right?
But still we say Strychnine killed him.
So, in the realm of economics, to finally return to the point that you were making about there is an imperfect set of information and other factors can always be present, well, that's very true, right?
So, to me, saying overprinting of money by the Fed caused the stagnation in prices, to me, that's like saying Strickney killed a guy.
Could there be other things?
Well, yeah, of course there could be. But efficacy requires that we stay in the realm of likelihood probability and recognize the diminishing returns in chasing wilder and wilder theories, and that at every level of human knowledge, I mean, you could go all the way back to the Cartesian demon, right?
Which we talked about many moons ago in philosophy, right?
That it's all an illusion and everything is made up and our senses are all lying to us and it's all imperfect.
We're a brain in a jar being manipulated by a demon.
Again, that's a kind of radical skepticism that I don't at all subscribe to, right?
But you could go all the way back to that.
And, of course, for that there's no null hypothesis.
And there's no null hypothesis for it could be something else, right?
Right? Maybe he wasn't killed by strychnine.
Maybe tiny, invisible meteors penetrated into his brain, which we can't see except with an electron microscope, and they killed him just before the strychnine did.
Again, you know, I'm making up nonsense, right?
But it's important.
Because what is the level of probability...
Of there being some unknown, unverifiable cause, right?
I mean, yeah, you can look for other causes for sure, and debates can go on about that, but at some point, more and more other causes there's no evidence for, right?
There's no big other invention that depressed prices and so on.
And you've got evidence from other things, and to me, again, you see, the purpose of knowledge is efficacy, right?
The purpose of knowledge is efficacy.
And... Efficacy has nothing to do with the past whatsoever.
Efficacy has nothing to do with the past at all.
Oh, sorry about that. The neighbor just came by with a present for Isabella.
So, yeah, efficacy has nothing to do with the past.
Nothing! Nothing, nothing, nothing to do with the past.
Why? Because efficacy cannot be applied to the past, because you can't change the past.
The past is a done deal, photographed.
You can't animate that, right?
So, efficacy is only and forever to do with the future, right?
Because if you're just acting randomly in the present, you are acting, but without any kind of efficacy, right?
So, Efficacy is always and forever about the future, right?
And we learn from the past and we apply the lessons in the present to change the future for the better, right?
So, efficacy is all about the future.
So, if you take radical skepticism...
I'm not saying you are, right? I'm just saying that if we follow this argument.
If we take radical skepticism into everything to do with the past, what happens is we completely destroy efficacy with regards to the future, right?
Because we can't learn any lessons from the past, because we're always paralyzed by, well, what if, what if, what if?
Invisible asteroids, alternative banks mysteriously failing to lend for whatever reason, earthquakes that were undetected and unreported, you know, whatever, right?
And so, if we hold up the standard of perfection, we destroy the value and purpose of knowledge and theory, right?
Which is to alter the future for the better, right?
If you do a medical study, some people will live and some people will die, right?
And you can't change that, right?
The reason you're doing the study is to help people in the future, right?
Efficacy is always and forever about the future, right?
Which is why I do counsel learning the lessons of the past in a personal way, learning the lessons of the past in order to have a happier and better future, right?
And all of the statements, I mean, just at a personal level, I know that's not related to economics, but just thought I'd mention it.
All of the statements that introduce radical skepticism into the past, right?
So if you had a bad family or bad past or whatever, and then people say, well, you know, your parents did the best they could.
You have to live, forgive and forget, and don't hang on to bitterness, move on.
And, you know, they're old now, they need you.
Like, all the stuff which introduced radical skepticism into your ethical and experiential history, your memories, right, of what actually happened.
It's terrible because it destroys efficacy, right, which is to learn the lessons of the past to a reasonable degree of knowledge, right?
I mean, if you had a bad parent and then you find out later, or in the future it's found out, that...
There's no such thing as evil.
There are invisible brain parasites that cause people to act badly.
You could say that, and therefore you would no longer need forgiveness for the evil or whatever, right?
It's not evil, it's illness and so on.
But that's not anything we have knowledge of at the moment.
I don't believe it's actually possible, right?
Because an ailment is something that does not respond to particular incentives usually, but this is moral behavior.
So, the radical skepticism destroys the efficacy, the purpose of knowledge, which is to learn the lessons of the past to create a better future, right?
So, the purpose of looking at, of course, deflation in the 1920s is to prevent inflation in the 2020s, right?
That's the point.
And if we introduce radical skepticism, like, well, it could be something else, it could be anything, it could be nothing, it could be, you know, we can't ever say for sure and this and that, right?
Well, we can't ever say for sure is a self-contradictory statement, of course, right?
And therefore, it indicates ambivalence, right?
Or indicates the necessity for ambivalence and complexity, right?
Because if you say, well, we can't say anything for sure, and again, I'm not saying you're saying this, I'm just saying that if we take your argument, or the argument that you've presented to its logical conclusion, you know, all other things, nothing is ever equal, therefore we can't know anything for certain.
Well, we can't know anything for certain is a statement of certainty, right, and therefore self-destructive, and therefore needs further analysis and examination.
And, of course, the purpose of knowledge is not certainty, which is impossible in this empirical context.
Absolute, platonic, perfect certainty.
The purpose of knowledge is not certainty.
The purpose of knowledge is efficacy, just like the purpose of health is not...
The purpose of medicine is not perfect health.
The purpose of medicine is a healthy life, not perfect health.
The purpose of engineering is stuff that works, not stuff that is perfect.
Like there's that old comedian's joke, why not make the plane out of the black box that never gets destroyed?
Because the plane couldn't fly.
And therefore, you would have safety, but not efficacy.
It would be safety because it would never fly.
And therefore it would never crash and so on.
But the whole point of the plane is to fly.
So... When Austrians, and again, I don't speak for Austrians at all, right?
But it's just my understanding of it, to whatever limited degree you find useful.
But Austrians who say Fed printing caused this price stagnation in the 1920s are not talking about the 1920s fundamentally.
They're talking about things we need to learn for the future.
Are the lessons valid? I believe that they are.
Are they perfect? No, of course not, because it never is.
But the perfection is the bell curve of efficacy.
And if we learn those principles to a satisfactory degree and can use it to prevent such disasters in the future, that knowledge is perfect.
Thank you so much for listening.
Export Selection