Saturday, August 7, 2010

Intellectual Property's Ungrateful Hitchhikers

I'm going to discuss an ethical and decision-theoretic intuition that underpins my support for intellectual property rights, and which seems to be absent, or unintuitive, among anti-IP libertarians. (See the discussion linked in yesterday's post for lots of good examples.)

But first let's consider a puzzle in decision theory. This one is known as Parfit's Hitchhiker and, as best I can tell, comes from Derek Parfit's book Reasons and Persons, though the term "hitchhiker" didn't come up in a search of the book.

It goes like this (well, my version does anyway): Assume you're lost in the desert, with nothing of value on you. You're approached by a superpowerful, superintelligent being we'll call Omega. It is willing to take you back to civilization and stabilize you -- but only if you will withdraw $5 from your bank account and give it to Omega once that's over with. (Yes, such a being might have reason to do this.) It has no enforcement mechanism for if you don't pay.

But here's the catch: Omega can scan you in detail and find out if you're really intending to give it the $5 when you're safe, rather than -- I don't know -- reasoning that, "Hey, I'm already safe, I've already got what I need and all, and you know, this Omega thing is powerful enough anyway, I think I'll just keep the $5." And if it finds that you wouldn't give it the money upon reaching safety (i.e. you don't have a decision theory that outputs "pay $5 to Omega" given that you are safe), then it just won't take you back and you can die in the desert.

At this point, a lot of you might be recoiling in horror: "What? Keep a measly five dollars when this thing saved my life? Are you ****in' nuts?" Yeah -- you're the people with the intuition I was referring to at the beginning -- the one that I have, and the anti-IP libertarians don't seem to. More about that in a minute.

Those of you who didn't recoil in horror may be thinking something like, "Whoa whoa whoa, I don't like dying. See, I would just make a contract -- or heck, even a simple promise -- that I will give Omega the $5. Since I feel honor-bound to abide by my promises, of course I would pay, and wouldn't have such diseased thoughts" as I referred to above. But I didn't make it that easy: note that Omega doesn't ask you anything and can't even receive your messages. Its decision is based entirely on what you would do, given that you know the details of the situation.

Here's the neat thing to notice: you will never find yourself in a position to be deciding whether to take that final step and give the Omega-like being $5 unless you adhere to a decision theory (or "ethic", "morals", etc.) that leads you to do things like "give Omega $5 for rescuing you at least in those cases where it rescued you conditional on expecting you to give it that $5" even when you already know what the Omega-like being has decided, and that decision is irreversible.

(I know, I know, I'm doubling up on the italics. Bear with me here.)

Conversely, all of the beings who come out alive have a decision theory (or ethic, etc.) which regards it as an optimal action (or an "action they should do", etc.) to pay the $5. Omega's already selected for them!

Now at this point, those of you who don't have the recoiling intuition I referred to, or are still worried I'll derive implications from it you don't like, may insist that this is a contrived scenario, with no application to real world -- you can't make your decisions based on what capricious, weird, superpowerful agents will do, so why change your decision theory on that reasoning?

And there is something to that belief: You don't want to become a "person who always jumps off the nearest cliff" just because there's some rare instance where it's a good idea.

But that's not what's going on here, is it? Omega makes its decision based upon what you would do, irrespective of what decision process led you to do it. So for purposes of this scenario, it simply doesn't matter whether you decide to pay that $5 because you:

- feel honor-bound to do so;
- feel so grateful to Omega that you think it deserves what it wanted from you;
- believe you would be punished with eternal hellfire if you didn't, and dislike hellfire;
- like to transfer money to Omega-like beings, just for the heck of it;
- or for any other reason.

So, then, is it normal for the world to decide how it treats you based on (a somewhat reliable assessment of) "what you would do"? Yes, it is, once you realize that we already have a term for "what you would do": it's called your "character" or "disposition" (or "decision theory" or "generating function").

Do people typically treat you differently based on estimations of your character? If you know where they don't, please let me know, so I can go there and let loose my sarcasm with impunity.

So, to wrap it up, what does Parfit's Hitchhiker have to do with intellectual property? Well:

- Omega represents the people who are deciding whether to produce difficult, satisfying intellectual works, conditional on whether we will respect certain exclusivity rights that have historically been promised them.

- The decision to rescue us is the decision to produce those intellectual works.

- The decision to pay the $5 represents the decision to continue to respect that exclusivity once it is produced "even though" they're "not scarce anymore", and we could choose otherwise.

The lesson: if you don't believe that the Omegas in your life "deserve", in an important sense, to be paid, you won't find yourself "rescued". We are where we are today because of our beliefs about what "hitchhikers" should do, and we miss out on rescues whenever we decide to become ungrateful hitchhikers. (Edit: that should probably be phrased as "... whenever we decide that it's right for hitchhikers to be ungrateful.")

(Note: this post was heavily influenced by Good and Real, Chapter 7, and by this article on Newcomb's problem.)

54 comments:

Stephan Kinsella said...

I empathize, man. I was where you were for several years--in that stage of desperately trying to find any good argument for IP, even to the point where you have to scrape the bottom of the barrel. It won't last forever. When you exhaust them all, then you'll finally see, "ahhhhh" I see why I can't come up with a good argument for it! :)

Silas Barta said...

I don't understand, Stephan_Kinsella. If you haven't found any good arguments, then why haven't you made up your mind about whether you support EM spectrum rights, and how to reconcile that with your position on IP? I mean, if you don't know your answer yet, that certainly sounds like you haven't exhausted all the good arguments, don't you think?

By the way, what exactly is your answer to the Parfit's Hitchhiker problem I gave?

Silas Barta said...

Note: there are two hyperlinks in the previous post; if you click on it make sure you notice them both. The way the columns are set up, the second link appears on a different line but appears to be part of the first, so it looks like one, but it's two.

nfactor13 said...

Isn't it more useful to deal with situations that are not one-shots like this? No doubt there are some artists who feel that they should be given these exclusivity rights by everyone. But there's nothing preventing such arrangements when people feel it's mutually beneficial, and it doesn't require IP law to be realized.

If you write a song, it may be that very few, if any, want to pay you for a recording of that song. But you may find that people are willing to pay you to perform live at private gatherings. People may want to pay you to hear you in concert. People may want to hire you to write a personalized song for a wedding. Just because you prefer to write songs and sell them in a particular way doesn't mean everyone else needs to agree to pay you.

Imagine you heard someone whistling a tune you'd never heard before, and then you decided it was a great tune and wanted to whistle it, too. If the guy turned to you and said, "pay me for that song, or I swear, I will never whistle another song out loud as long as I live," I think your best rational response would be to back away slowly and make no sudden or threatening gestures.. not to pull out your wallet. :-)

nfactor13 said...

Another suggestion. Why use a fantastical example when there's a far more common one available? I'm thinking of the rationality of tipping. Most people do it, even in situations where they have no expectation that they'll return or encounter the same people ever again. I think it's rational, but I would find it odd to advocate a tipping law to ensure that people support wait-staff for the service they provide. I doubt you fear we may end up not having anyone willing to work as a waiter/waitress if we don't have tipping laws.

ianpollock said...

Hm, I will have to think about this, but it does seem to be a powerful argument for intellectual property.

Stephan Kinsella said...

Ianpollock: it's not "powerful" at all; it makes no sense. And in any case, how is the argument *libertarian*? Is this what libertarianism is about? Stopping hitchhikers?

Silas Barta said...

@Ian: Thanks for the comment! I probably would have submitted this to LW if not for the focus on a political issue and the fact that its rationality points aren't much different from the article I linked at the bottom.

@nfactor13: Three points:

1) Your reply unnecessarily focuses on the legal and enforcement aspects of the decision to pay/respect IP. But note that the point applies completely independently of the legal enforcement. What ultimately matters is whether people respect IP, regardless of their reason for it. Certainly, setting the laws to require people is one way to promote such respect (and it can do the opposite, if the IP laws become abusive!). Still, all such factors are "screened off" by the *actual* respect they show and whether potential producers believe this is credible. If you agree that "everyone should respect some measure of IP", how do you think you really differ from IP opponents?

2) You've shown alternate ways that people can try to capture a portion of the value they've produced. But, like I argue in the IP calculation argument (point 3), these methods are *not* selling the same thing, and therefore not rewarding the same kind of valued activity, as when they can sell the rights directly. A system in which creators can *only* make money in ways that don't require respect of IP will find it self rescued less -- it will be less and less in the position to decide whether to pay for a benefit that's "already happened anyway".

3) This tipping scenario does highlight some of the salient points. A certain form of food service business only exists in the first place because of a reliable, and correct, estimation of the collective "character" of the local population, in the eyes of those offering restaurants with waitstaff. For this to exist, a large portion must find it unthinkable not to tip, "even though" they've already been served.

The same kind of mind is selected for by Omega in Parfit's Hitchhiker, and by potential creators of difficult, satisfying intellectual works.

Silas Barta said...

@Stephan_Kinsella: You're still not answering the question: would you pay Omega? Why or why not?

These questions aren't some separate compartment; sometimes the issues connect in a way that isn't immediately obvious. Just like the similarity of EM spectrum rights and IP -- which, I note, you also haven't decided how you reconcile.

nfactor13 said...

"What ultimately matters is whether people respect IP, regardless of their reason for it."

I think what ultimately matters is that people have 1) multiple reasons for producing artistic works, and 2) people have multiple reasons why they might pay an artist that could include, but are not limited to, their beliefs about IP. There are also many ways to encourage the production of artistic works that do not involve IP in any way. The example you're using presents it as very binary: either you value the rescue enough to agree to the price demanded, or you don't get a rescue at all. But there are plenty of in-between options, especially when trying to put a value on something so subjective as artistic taste.

"If you agree that "everyone should respect some measure of IP", how do you think you really differ from IP opponents?"

Not sure what you're referring to here.

"these methods are *not* selling the same thing, and therefore not rewarding the same kind of valued activity, as when they can sell the rights directly. A system in which creators can *only* make money in ways that don't require respect of IP will find it self rescued less"

My point is that this is not inherently a bad thing. It seems you're saying that IP law will maximize a certain kind of creative activity, and that might be true (though, as you say, it may just be maximizing a particular product or type of product), but I don't know why that should matter. Why is maximizing the kind of activity that IP law would help a goal that I should support? Why is the kind of creative activity that does not require IP not the better one to maximize? (And clearly it won't be produced as much if IP law makes it easier to produce other things that require IP law to be profitable. Or do you disagree with that?)

"For this to exist, a large portion must find it unthinkable not to tip, "even though" they've already been served."

However, this is not universal. There are cultures where tipping is either pre-determined or doesn't happen at all, and they still manage to have waitstaff jobs that people are willing to take. And I think that's fine, but it just so happens that I live in a society that values tipping, and I myself see the value in it. I don't think everyone needs to agree with me on that, though. It's not clearly the best choice possible for everyone (at least, I don't see it that way).

"The same kind of mind is selected for by Omega in Parfit's Hitchhiker, and by potential creators of difficult, satisfying intellectual works."

Just to reiterate: it may be that too many people are doing difficult, satisfying, intellectual work. In fact, I think in some areas of society, that's clearly the case. So even if you're correct, you still haven't established why it's a bad thing. Somewhat of an is-ought gap, and I'm not entirely convinced of the 'is' step anyway. :-)

I appreciate your comments, though, and thanks for taking the time to answer my questions so clearly. I'm still thinking these things through, so I appreciate hearing other views.

Silas Barta said...

@nfactor13:

I'm glad you found my comment helpful, and it's my pleasure to offer my thoughts on your last one.

1) Regarding the not-so-bad alternate profit models for intellectual works, and why I think the IP-centered ones are important: The reason is that's it's a Pareto-improvement for people to make profitable intellectual works predicated on (respected) IP rights -- even relative to those other models. That is, the creator and the people willing to pay both benefited. No one lost out.

If there were a method that were Pareto-superior even to these, they could exist even alongside IP. But with IP, we get these works that require it, without losing out on those that don't. You correctly recognize this isn't an either/or thing, but then, the Parfit's Hitchhiker easily generlizes to the continuous case: the *more* we respect IP, the *more* of these marginal cases (which are Pareto-improvements) we see.

(There is, of course, the issue about less specific creations, like inventions, which others can reasonably claim they discovered independently. I agree that such independent inventors should also have the IP rights in such cases, as so most IP proponents. But I admit that aspect can get tricky.)

2) "If you agree that "everyone should respect some measure of IP", how do you think you really differ from IP opponents?"

Not sure what you're referring to here.

This was about your position going in the direction of, "IP is great, but it shouldn't have the force of law behind it." That is, you seemed to be saying that, even if it might be a noble thing for everyone to collectively decide, without being forced, to respect (reasonable) claims of IP, that doesn't imply there should be laws about it.

My response there was saying, once you agree with this much, you're already in agreement with IP proponents -- they think people should respect IP without being forced too!

3) Regarding tipping: It's true that there are alternate restaurant model equilibria. However, the public's acceptance of tipping "against all reason" does affect the final form of the service, and the ability of restaurants to extend a certain amount of "credit" to patrons depends on this acceptance. It's not that something bad will happen if you don't tip; it's that when people generally "don't see" why to tip, restaurants will have to be less trusting than otherwise *for that particular society*, even if norms can make up for it in another context.

Okay, this reply may be a bit rushed, but I hope you found it responsive.

nfactor13 said...

"The reason is that's it's a Pareto-improvement for people to make profitable intellectual works predicated on (respected) IP rights -- even relative to those other models. That is, the creator and the people willing to pay both benefited. No one lost out."

Do you think what you've said applies to the fashion industry? (meaning, would imposing IP law on the fashion industry benefit at least some and no one would lose?) I'd be very interested what benefit you see for that industry to have IP law involved (and alongside, why you think it wouldn't cause any harm to anyone involved).

"If there were a method that were Pareto-superior even to these, they could exist even alongside IP. But with IP, we get these works that require it, without losing out on those that don't."

I'm unclear on what the Pareto criteria you're using are. In the first part you talk about the creator and those willing to pay as beneficiaries and no one losing, and here you're talking about the total works that need IP going up and none that don't losing out. But those are totally different things. It's quite different if you're only measuring the number of 'works' in a sort of Platonic existence without regard to the way those works might exist, the forms they might take, the multitude of uses and interesting combinations they might be put to.. things that IP law will generally disallow if it's to have any meaning. Why tally the number of 'works' to decide on this Pareto optimality?

"That is, you seemed to be saying that, even if it might be a noble thing for everyone to collectively decide, without being forced, to respect (reasonable) claims of IP, that doesn't imply there should be laws about it."

I suppose I agree with that, but I don't think that the antecedent is true, even generally. So you can't conclude that I think people should respect IP, only that I think it's all right if they choose to.

Thom Blake said...

A search on the word 'desert' (which is sadly ambiguous in a book discussing social morality) turned up page 7 as the source of the thought experiment.

Cyan said...

I love that you take the time to specify disliking hellfire as a condition for acting to avoid eternal hellfire.

Silas Barta said...

@Thom: Thanks for finding that!

@Cyan: Well, you know me ... ;-)

Silas Barta said...

@nfactor13: Re: fashion industry: First, keep in mind that I'm advocating a specific kind of IP, not the specific cases of clear abuse you've heard. I know it sounds like a cop-out, but a) I'm going to elaborate, and b) I think it's just as fair for advocates of property rights to disclaim responsibility for e.g. when farmers were awarded the right to keep aircraft from flying 15,000 feet above their property based on property-rights arguments.

For IP in the fashion industry, I would agree that it's ridiculously unfair for e.g. someone to be barred from making a design they came up with because of, at most, minor inspiration for a "patented" design. And I think most of the repulsion at fashion patents is how you have to call out a part of the designspace that will inevitably collide with genuinely original designs.

What complicates it further is that fashion is highly positional good: it's usually the mere *fact* that a designer was responsible for a specific fashion that makes it valued rather than (relatively) objective criteria. So really it's a non-zero sum situation when someone comes up with a design tailored to the rich, and worsened further (than it otherwise would be) when they have to do that repeatedly as people keep copying the rich. This is why I'm perplexed by IP opponents who view this fashion industry phenomenon as good -- they're essentially saying it's good that people have to buy new clothes more often to signal the same level of trendiness/status. Huh?

With all that said, I do think there is a reasonable set of fashion IP rights that would not infringe on designers who came up with something original, even if it had inspriation from IP-protected ones.

Re: Pareto-optimality, I think I did make a valid comparison. If you imagine a pre-creation-X state, and what happens when someone creates X, and others respect (or enforce respect of) X's exclusivity, you see that they are comparable, and everyone has become better off. Even those who are kept from pirating it, at least now have the *option* to buy rights to it. (Just like how thieves are better off in a society with strong pro-property rights norms, as it leads to things being more worth stealing, which is part of why people hate them so much -- they sap off the very reason for their practice!)

This doesn't require any extra abstraction: even if there are other ways, the fact that people don't choose them when there's IP means that they're a less efficient way of doing things. Think about it like this: yes, there are models in which people give away free goods to each other, but that can exist, whatever the property rights are. With property rights, though, people are better off because they also have the option of the goods that required a different model.

Or, imagine the efficiency gain when a store comes to a down of thievish people vs. a town of paladin-types. In the former, the overall gains are less because the store will have to resort to a "non-standard model" that involves more guards, less ability to browse goods, etc. In the paladin down, he doesn't have to pay for security at all, and everyone reaps the gains -- even though, of course, someone "could" start up shop with a heavily-guarded store.

I suppose I agree with that, but I don't think that the antecedent is true, even generally. So you can't conclude that I think people should respect IP, only that I think it's all right if they choose to.

Could you clarify what regard as good, in what ranking?

(Hope this has been at least somewhat informative.)

nfactor13 said...

Thanks for the comments, and I'll either reply to this again in a couple days or just wait for future blog posts. Life getting in the way of important stuff again. :-)

I appreciate hearing your perspective because even when I disagree, it forces me to think more deeply about why.

Just briefly on the fashion question, I'm not one who cares much about it, but I don't know that it's legitimate to say that people who place a high value on trendiness/novelty would be better off if we slowed down the fashion cycle.

I'll look at it again, though, so thanks.

Jonii said...

Good post, but at first, you gave an impression that you were about to give an argument for IP rights, but the lesson ended up being tangential to that question at best.

Kaj Sotala (Xuenay) said...

This argument seems to be premised on the assumption that copyright is necessary for the production of works. While a very limited form of copyright may indeed be useful for helping create new works, there's little evidence that anything near the current system would be necessary.

I'd give Omega the $5 just for the gratitude of being rescued. I also attempt to pay for my entertainment when I can. I also support a drastic curtailment of intellectual property, including the abolishment of at least software, bio- and pharmacutical patents and probably also the rest of them, a complete legalization of non-commercial copying and shortening commercial copyright to maybe 15 years or so. Trademarks are mostly fine, though not when they're abused in a copyright-like manner.

A desire to reward the makers of art for their work does not equate to a desire to give them harmful monopoly rights. It's better to compensate them without giving them a monopoly.

Silas Barta said...

Kaj_Sotala_(Xuenay): I wasn't arguing for the exact current system (which I agree goes overboard in many respects), and so the theoretical failures of the current system don't refute the points I made.

And the desire to reward creators certainly doesn't justify "harmful monopolies", but it definitely justifies a lot more than "I'll pay you when I can", and pick and choose which classes of innovations I deem deserving.

The "monopoly" on the innovations allows for returns comparable to devoting labor toward physical (rather than intellectual) goods. If you understand why property for the former (physical) produces the appropriate return for the producer, why don't you see it in the case of intellectual works?

I also generally question the usefulness of the term "monopoly" in this context. It's a monopoly on something that only exists *because* of the creator. Sure, they deprive you of the ability to copy, but had they not produced the work *on the basis* of expecting certain IP rights, you wouldn't have that ability either! Really, what are you complaining about?

(Yes, there are cases of coincident discovery, but most IP proponents and I wouldn't deny that such discoverers should have a claim to the IP rights as well. Address the least convenient possible world.)

Kaj Sotala (Xuenay) said...

I also generally question the usefulness of the term "monopoly" in this context. It's a monopoly on something that only exists *because* of the creator.

Something only existing because of the creator doesn't make a monopoly any less a monopoly.

Also, the main issue isn't in any particular creator happening to have a monopoly on his products. The main issue is the automatic protection that is given to any cultural products. If only one photographer had a monopoly on his photos, it'd be no problem. But if clearing the copyrights on countless of works can e.g. stall the re-release of a documentary for 13 years, then there's a problem.

Sure, they deprive you of the ability to copy, but had they not produced the work *on the basis* of expecting certain IP rights, you wouldn't have that ability either!

They could just as well have produced the work on the basis of expecting profit, without IP rights. (At least not IP rights of the modern kind.)

Really, what are you complaining about?

I'm complaining about a special privilege that is harmful for society. I admit that a limited form of it can in some situations be a necessary evil, but I object to causing broad economic damage by making that privilege any broader than it need to be.

There are also the moral-theoretically questionable implications of limiting what I'm allowed to do with my own mind or my own private property (see the previously linked "Why IP is not property"), though I'm a pragmatist and thus don't find theoretical implications as serious as practical consequences.

Silas Barta said...

Something only existing because of the creator doesn't make a monopoly any less a monopoly.

Nor does it become any more relevant to point out the fact of the monoply. You have a monopoly on "Kaj Sotala's mp3 player". So? Does it carry the same significance as other kinds of monopoly's? No -- so from the fact that it's a monopoply, it does not follow that it carries the negatives we normally associate with monopolies.

Also, the main issue isn't in any particular creator happening to have a monopoly on his products. The main issue is the automatic protection that is given to any cultural products.

And I wasn't defending the exact present system, so this is irrelevant. Of course IP should be something you have to opt into so that the IP protections only attach to those who made the creation predicated on having IP rights. You're trying to equate all advocacy of IP with advocacy of the present system, while trying to deny the distinction in other context.

They could just as well have produced the work on the basis of expecting profit, without IP rights. (At least not IP rights of the modern kind.)

Sure, just like someone could produce lumber in a society without property rights in lumber while expecting to make a profit despite not having recognized property rights in the lumber. That doesn't make it any more relevant to the issues at hand.

I'm complaining about a special privilege that is harmful for society. I admit that a limited form of it can in some situations be a necessary evil, but I object to causing broad economic damage by making that privilege any broader than it need to be.

And again, I didn't defend the specific, broad form of it that exists today, just the concept of recognizing limited exclusivity in the right to use an idea for those responsive for the idea's existence (or rather, instantiation). Furthermore, your claim that it's a necessary evil prematurely assumes that IP is necessarily evil.

There are also the moral-theoretically questionable implications of limiting what I'm allowed to do with my own mind or my own private property

And it's likewise questionable to automatically assume that because you own the implements, you necessarily have a right to do anything with them. If you own a radio transmitter, does it follow that you have the right to broadcast at all frequencies? Remember, those are just as much "pattern privileges". And before you make the obvious responses, see my handling of it here.

(see the previously linked "Why IP is not property"),

Yes, I'm intimately familiar with the arguments therein, and if you just click on the intellectual property category, you'll see my refutations of those. Your opposition to IP doesn't look any less automatic and political-minded as Stephan_Kinsella's.

Silas Barta said...

Correction: The first link should go here.

kodos96 said...

I was unfamiliar with the Parfit's Hitchhiker problem prior to your parallel post on LW - it's a very interesting problem, and I found your post on it quite enlightening.

But trying to analogize it to intellectual property is, I think, a losing proposition. There's just not enough similarity. Sure, these kinds of thought experiments can often be useful even if they don't map precisely to the real world case, but the differences here are so large and numerous that I think they cross the point where it's just not a useful analogy anymore. To make the problems equivalent, you would have to make the following modifications to PH:

-There are a million people stranded in the desert, rather than one.
-Omega has to choose whether to rescue all of them or none of them.
-His decision is based on whether some minimum number of them pay the $5.
-Omega has ways of financially benefiting from rescuing people other than the $5 payments.
-Omega receives non-monetary compensation for rescuing people, regardless of the $5 payments (fame, respect, admiration, i.e. status)
-Omega personally enjoys rescuing people. (I'm an amateur musician. I write and record songs, for myself. It's fun).

etc etc etc.

Now I'm not even necessarily saying that these modifications push the decision process in one direction or the other, just that they're big enough differences that PH is simply not a good intuition pump for IP.

And then I have a second objection: even if PH, or thinking along similar lines, did provide a slam-dunk argument for IP, it's not a pro-IP libertarian argument. It's a utilitarian/consequentialist argument. If you want to call yourself a pro-IP libertarian, you have to justify IP on libertarian grounds. That means you have to hold "liberty" (however you define it) as the terminal value to which your political beliefs are instrumental. Now it just so happens that many libertarian policy positions (including anti-IP, IMHO) happen to also be "good for the economy" (i.e. justifiable on utilitarian grounds, holding GDP or whatever as your terminal value)... but that's just a side benefit! Being a libertarian means supporting the liberty-maximizing position, for liberty's sake alone, even if it happens to be "bad for the economy".

kodos96 said...
This comment has been removed by a blog administrator.
kodos96 said...
This comment has been removed by a blog administrator.
kodos96 said...
This comment has been removed by a blog administrator.
Kaj Sotala (Xuenay) said...

And I wasn't defending the exact present system, so this is irrelevant.

I'll ask again: exactly what kind of system are you defending? There's no way for me to productively critique an opinion before I know what the opinion is. For that matter, if you laid out your preferred system, we might even discover that we have no disagreement in practice.

Silas Barta said...

@Kaj_Sotala_(Xuenay): What does that matter? My only point is to justify the existence of *some* kind of IP against the countervailing "but it's not scarce anymore"/"I've already been rescued" argument. And you already agree with this conclusion. It's just that you're trying to equate my position with the exact present system, when it makes no attempt to do so (though I should have perhaps been more clear about which rights you need to recognize to get the "rescue").

This is an unfair attribution, and reveals an attempt to use flaws of one specific system to justify removal of all IP (or at least the IP you don't feel like recognizing), which looks a lot like motivated cognition to me.

@kodos96: Good points, all. A few of my own:

1) Yes, you produce music in the absence of IP, but that's about as relevant as saying that people still produce things the absence of physical property rights -- I wouldn't want to depend on that. Would you?

2) I don't think I'm a libertarian in the sense you describe. I regard myself as a type of libertarian because it closely describes what I believe would be necessary for an ideal society, and does so because it promotes (what I regard as the important aspects of) liberty. The match isn't perfect, though, and so non-liberty desiderata will come into play, and almost certainly motivate most other self-described libertarians.

3) Also, the IP issue for libertarians is partly about what kinds of property rights one can accept whilst still being a libertarian, so non-liberty desiderata will be relevant in that respect as well.

Hope you don't mind if I clip your duplicates. It's a good idea to check if anything got posted if you get an error message (personal experience, applicable to all discussion sites).

Kaj Sotala (Xuenay) said...

What does that matter?

Umm... curiosity? Desire to continue a conversation? I obviously misinterpreted your position at first, so I'd like to know exactly what your position is, so I can avoid such a mistake in the future. And I'm genuinely curious about what exactly your position is, since my priors would have said that somebody wielding the sort of rhetoric you were wielding would have been in favor of rather strong IP.

In my experience, hardly anyone is in favor of removing all IP, so defending the claim that "there must be at least some IP" would be so trivially easy as to not be worth doing. Therefore I assume that anyone setting out a defense of IP is advocating relatively strong IP.

This is an unfair attribution, and reveals an attempt to use flaws of one specific system to justify removal of all IP (or at least the IP you don't feel like recognizing), which looks a lot like motivated cognition to me.

That's hardly fair. When I listed some of the problems in copyright, I explicitly noted that some of these may be irrelevant, depending on exactly what sort of protection you had in mind. Then I invited you to expand on your position.

The reaction most people would have given would've been to say "those don't apply, because my advocated system is instead like this", and then explain their preferred system. That was the response I was hoping to get, and which I then explicitly asked for because I didn't get it. It wasn't an attempt to justify the removal of all IP.

kodos96 said...

Silas: Yes, you produce music in the absence of IP, but that's about as relevant as saying that people still produce things the absence of physical property rights -- I wouldn't want to depend on that. Would you?

This may or may not be applicable to IP in general (i.e. tech patents etc), but in the specific case of art/music, yes, I would want to rely on that - I actually think that the average quality of artistic works would increase in the absence of IP. You show me a musician who would stop making music if he stopped getting paid for it, and I'll show you a musician who makes shitty music.

Kaj Sotala (Xuenay) said...

kodos96: You show me a musician who would stop making music if he stopped getting paid for it, and I'll show you a musician who makes shitty music.

It's not that the musicians are greedy, it's that they need to earn their rent somehow. It's hard to make quality art, at least in any reasonable time, if you also have to work in a day job. Especially if the job is one you don't like, since that will drain your energy disproportionatelly. If you add kids to the mix, it becomes next to impossible.

Silas Barta said...

@Kaj_Sotala_(Xuenay):

In my experience, hardly anyone is in favor of removing all IP,

A ... bu ... I've been dealing with them for the past five years. Click through to the Stephan_Kinsella character above, as well as the "IP is not property" argument you referred to. It's why I've made the arguments in the "intellectual property" category in the first place!

As for my beliefs about what the ideal IP system would be, I agree with you that terms should be shorter, and it should be something you have to specifically assert on introducing the work, and it should be noted in some central registry so people can check the status of unknown works, and independent discovery should be a valid defense. (I also proposed a plausible, different system here.)

I think abolishing pharma patents is insane, as those are pretty much an ideal case: it's near impossible to accidentally create an infringing one yourself (and it's provable if you do).

But that's way beyond the scope of the article, which, again, is to address the "but it's not scarce anymore!" argument, and if you think I'm making that up, go here.

Quote: "The purpose of the market is to make useful things relatively less scarce. With information, that goal is already here."

... you see what I mean?

kodos96 said...

Kaj Sotala: It's not that the musicians are greedy, it's that they need to earn their rent somehow. It's hard to make quality art ... if you also have to work in a day job. Especially if the job is one you don't like

I disagree. Have you ever noticed how for so many bands, their best album is their first one? Or the one they made right before they got famous and quit their day jobs? Having a day job that you hate can be a powerful inspiration for making great art. Once an artist is famous and has groupies following him everywhere, he's no longer a relatable person with life experiences to write about that normal people can identify with.

Silas Barta said...

@kodos96: This may or may not be applicable to IP in general (i.e. tech patents etc), but in the specific case of art/music, yes, I would want to rely on that - I actually think that the average quality of artistic works would increase in the absence of IP.

How does IP prevent them from using non-IP based models?

You show me a musician who would stop making music if he stopped getting paid for it, and I'll show you a musician who makes shitty music.

How about if I just showed you people who have the highest comparative advantage in creating music, but, absent IP, would apply their brains in the advertisement industry, where they could make more money but produce something of lower value?

I can point to lumberjacks who would keep cutting down trees if we abolished property rights in lumber. They'd have to buy the tree farms with money from their "real" jobs as ditch diggers, and lumber would be much harder to find, as you'd have to get it from a friend who happened to have some boreal land, or through the informal economy.

But lumberjacking is all they've known, so they'll keep at it. It doesn't sound very appealing to me -- and I'm not a lumberjack.

Silas Barta said...

(And that's okay.)

kodos96 said...

Silas: How does IP prevent them from using non-IP based models?

It doesn't. But note the bit of rhetorical sleight of hand I used: "average quality of artistic works". If all the uninspired hacks who are just in it for the money dropped out, the only artists remaining would be those with genuine creative passion, and therefore the average quality would increase.

How about if I just showed you people who have the highest comparative advantage in creating music, but, absent IP, would apply their brains in the advertisement industry, where they could make more money but produce something of lower value?

This kind of reasoning may mnake sense for non-artistic IP, like tech, or even for certain forms of art, like movies, which require major capital to produce. But for music, literature, and lots of other forms of art, nothing is stopping anybody from working in advertising and also making art in their garage after work.

I know I'm sounding like a crazy hippy here who thinks that money corrupts everything.... I swear, I'm not! I'm a libertarian! But art is a really special case, in which lots of standard economic analysis breaks down. I can't think of any other kind of good or service where the quality of the product is actually directly determined by its creators motivations, passions... the content of his "soul". Econ concepts like comparative advantage just don't capture that.

Silas Barta said...

@kodos96, I don't see how you're looking at the relevant set. People whose work really sucks because they're in it for money doesn't actually dilute the pool of things you look at, because of its self-limiting nature, right?

kodos96 said...

I'm not sure what you mean by "self-limiting nature"

Silas Barta said...

What I mean is, looking at the set of music you allow yourself to be exposed to, music that really sucks doesn't have much power to stay in that set, by virtue of its sucking.

Ditto for the general public's set, though their taste probably differs from yours.

kodos96 said...

Oh I see what you're saying. In a sense though, the existence of bad art does dilute the pool, in that it makes it more difficult to find good art - requiring lots of effort to be expended on sophisticated machine learning algorithms to filter and recommend good art - effort which could have been spent elsewhere. It probably isn't really worth arguing this point though, since it isn't "my real rejection" of IP.

Bakabon said...

I came to this site after noting and respecting your very rational analysis of Mankiw's recent tax argument at another blog. It so happens that I am also a fan of the IP debate, so I had a question (You've probably heard this argument before. If so, could you point me to a link with your defense?)

I don't understand how being compensated for the labor of producing an intellectual work should be any different than compensation for a non-intellectual one.

Say I am a carpenter who makes chairs. I could make chairs in my free time, or I could make a living out of it. In that case, if someone has a demand for a chair, they pay me a price we both agree on as compensation for the time and labor I spend on its production.

Now I am a song writer. Sometimes I write songs for fun, but it is also my job. A popular singer comes to me and tells me they will pay me to write a song. We agree on a price that reflects the time and labor I put into writing it.

It seems to me that the second case would easily exist without copyright. Some people are good song writers, some are good singers, some are good carpenters. Each gets payed for their labor based on supply and demand of that resource. Why are any monopoly rights then necessary? Why should the song writer be given some extra (often potentially infinite) reward that was not agreed upon for doing the labor in the first place?

You could argue that it would be nice if I wrote more songs. But you could say the same for more chairs...

Thanks for your comments.

Baka

Silas Barta said...

@Baka: Thanks for visiting and for your comments. (For anyone who's wondering, I think Baka is referring to this discussion about Mankiw.

To answer your question: yes, in the absence of IP, it would be possible for someone to commission a song from you, and thus it would be possible to make money this way. But two problems arise:

1) Most importantly, this assumes the deal can be struck *before* the song exists -- but with intellectual works like songs, you don't know what the song is like in advance (or you'd create it yourself), and don't know you like it until you hear it. Like with the Parfit's Hitchhiker problem (and as I spelled out in the analogy), Omega (the writer) only decides to rescue (create music) based on expectations of how it will be treated post-rescue (post-writing) -- a time when the rescue (creation of the work) is irreversible!

Most song's can't be produced by this model -- someone writes, and *then* others decide whether they like it.

2) If, upon creating a chair, someone doesn't pay, you can withhold the chair from them. But you can't do that for an intellectual work now, due to the ease of copying. So in any case where the work wasn't commissioned in advance, you will never be able to do the analog of selling chairs unless there are strong pro-IP norms (whether or not they are codified).

Now, on the issue of incentive structures, and how profitable certain activitities "should" be:

The problem isn't "can this make money?" per se, but rather, "Does the return scale with the social benefit?" There is a net loss if people are, in the aggregate, willing to pay ~$1 million to bring about the existence (initial instantiation) of a particular, good new song, which it only "costs" (in terms of forgone other opportunities for the writer) ~$10,000 to create. (This is indeed similar to the externality problem -- which is not that people get music without paying, but that nobody pays, and nobody gets the music.)

In most cases, they can't arrange this Pareto-efficient move, *even if they could coordinate*, because they don't know if they like the work until it's created. Even if someone is a wealthy enough to commission a work, they will woefully underproduce relative to what people are willing to pay for (and, as Jaron Lanier points out in You Are Not a Gadget, the content of the work will just be skewed toward the tastes of the super-wealthy and government propaganda, despite people willing to pay for other stuff).

I also think the role of profitability in IP is over-emphasized. Physical property rights are good, whether or not anyone intends to use those rights to make a monetary profit. People can similarly enjoy the exclusivity that IP rights bestow, and understand the social benefit thereof, whether or not creators intend to make a monetary profit.

I hope you found this responsive. If interested, see my other intellectual property posts here, or just watch the Mises Blog for whenever they discuss IP, in which I inevitably swoop in and comment.

Silas Barta said...
This comment has been removed by the author.
Silas Barta said...
This comment has been removed by the author.
Silas Barta said...

Sorry, accidentally triple-posted there, that's what I deleted.

One other thing: I wrote this LessWrong article as a non-IP-related follow-up.

Bakabon said...

Thanks for the quick and thorough response.

I still don’t understand why commissioning wouldn’t be possible.

The lack of IP protection does not stop the ability to form contracts.
I could create a song and allow you to see it under the condition (that you agree to in writing) that you not show it to anyone else or use the idea yourself. If you violate this agreement I can take you to court. If you like the song, you can pay me for my labor, and I would allow you to then use the work. Depending on the type of contract agreed to, the entire IP regime could theoretically be drafted if the two parties consented. All this could be done without the government granting artificial monopoly over the idea itself. Isn’t this done already with non-disclosure agreements?

As for the returns to scale argument, I may not be clear on your meaning. It seems to me that the future benefit to society is irrelevant. After all, we can’t predict that benefit anyway. This seems similar to the old Marxist (I believe?) argument on capital. Say I build a machine that can produce widgets. If you buy this machine for your factory, you can use it to produce an infinite number of widgets. Therefore, shouldn’t you pay me an infinite number of dollars? After all, the machine is capable of producing that much future value. In reality, you pay me instead for the labor/cost of making the machine. I don’t see why intellectual works should be any different.

Thanks the other links. I am pretty new to blogging, so I will check them out.

Silas Barta said...

@Baka:

1) Yes, commissioning would be possible -- my point was that it would just miss out on the all the Pareto-improvements that IP exists to facilitate. The method you describe of making sure that the post-creation buyer doesn't just "steal" the work without paying is itself a form of IP (as I tried to arguein my 2) above).

Perhaps you don't regard that as a kind of IP, but in any case, the substance of our positions would be the same; and the people I argue with on this issue believe that their opposition to IP means that such contracts should not be enforceable.

2) The benefit to the future society is not completely irrelevant, because it determines the maximum amount someone should be willing to pay to produce the intellectual work. The reason that infinite benefit never implies infinite price is that competing producers will find a finite cost to production, and competition will ensure that no one has to pay the infinite amount.

Still, since the future benefit functions as an upper bound on what creators should invest in producing such an idea, then "concealing" this benefit (i.e. when people can flout IP laws/norms) will lead to an inefficiently low amount of production of such ideas -- inefficient in the sense that everyone could be better off if more ideas were produced.

Hope that answers your question.

Bakabon said...

Thanks again for the response.

It seems then that we are in agreement on point 1. I don’t know why such contracts drawn up by individuals shouldn’t be enforceable. A contract is a contract. Rules should be fair and consistent, not arbitrary. My position is that such agreements are possible without the need for additional government intervention. A person should be compensated for their time based on a contractual agreement – nothing more, nothing less.

As for point two, I believe that a government mandated IP protection actually reduces the output of more ideas by raising the cost of entry, as well as artificially restricting the supply (of something that can be reproduced for 0 cost). New ideas do not come out of the blue; they arise by integrating and reordering pre-existing ideas. Adding artificial costs and limitations to these ideas only slows down the creative process.

Example: It is possible that I, a non-millionaire/corporation, have a story involving Harry Potter that is more interesting than Rowlings’. It would raise global utility to get this new story out there for children and fans across the world. It is impossible for me to benefit financially from this idea, however, because the cost of entry – buying the rights - are too high. Lacking monetary incentive, the idea remains in my head, never “benefitting” society. In a world without government mandated copy protection (but with contracts like normal), I could try to sell my idea to whatever publisher would take it.

Silas Barta said...

@Baka:

I agree with you that government management of IP rights is really bad, just like its handling of property rights in general. And there are easy improvements that can be made to IP law, like shorter copyright terms, less restrictive fair use and derivative work restrictions, and greater certainty about boundaries.

My "battle" here is mainly with the total IP abolitionists, like Stephan_Kinsella, who posted the first reply here. We can disagree about the details of what IP norms are appropriate, while still agreeing that some level of exclusivity should be granted to the people responsible for the very existence of a creative work.

Some, however, won't even go that far, and seem to make a far more serious error in reasoning about the origin of ideas...

Hashim said...

once we have better narrow AI, and that doesn't seem to be that far off into the foreseeable future, we'll never need good artists etc, so any arguments in support of IP are moot.

Kevin Brown said...

Once I finally got through the "puzzle" (which was pretty painful in and of itself) I was very ready to hear how this relates to "IP". Unfortunately I was sorely disappointed by the first bullet which gives no justification for "certain exclusivity rights". Like most IP advocates the author simply assumes such rights exist and are valid yet this is precisely what Kinsella and company question. IP advocates consistently beg the question about IP "rights" by reseting all of there arguments on the premise that IP "rights" exist and are valid; This is where all IP advocates fail entirely and no amount of utilitarian arguments and emotional pleas of "how will they eat" will change this.

Silas Barta said...

@Kevin_Brown: (just saw your comment) Whoa, where did I assume "such rights exist and are valid". All that my argument relies on is that there is a tradition of granting IP rights to certain creators. (Yes, people respond to incentives, if you can believe it.) How does your comment refute any point I presented?

To make the analogy clearer, would you refuse to pay Omega on the grounds that, "dur, I done already been rescued"? If so, why do you endorse the common anti-IP argument that, "dur, I already know the content of your intellectual work, what do I need *you* for now?"

See, it's this complete disconnect, this inability to engage an actual argument without reverting to your standard comfort zone cliches that makes it impossible to have productive dialogue with anti-IP fanatics.

gwern said...

> IP advocates consistently beg the question about IP "rights" by reseting all of there arguments on the premise that IP "rights" exist and are valid; This is where all IP advocates fail entirely and no amount of utilitarian arguments and emotional pleas of "how will they eat" will change this.

IMO, the real question-begging here is whether IP is isomorphic to the Hitchiker in the first place.

In the Hitchhiker, there are no negative effects or externalities as a result of being rescued aside, rescuing is definitely something that is good, and so on. These rule out a lot of common anti-IP arguments - that IP leads to deadweight, that we invest too much in IP, that the monopolies cause further inefficient behaviors, and so on.

It's an interesting vantage, but it's not much different an argument from 'IP is good for society, but it's hard to enforce. Morally, shouldn't you respect IP?'