Classic Style Intellectual Worlds

The photo above is one I recently took at the Vatican. For me, it illustrates some key principles of classic artistic style. This style tends to be a fractal collection of structures at different scales, structures that frame spaces of many different sizes. Each structure doesn’t use up all local space, but instead leaves open holes for other works of art. Most all space is used by art of some sort, but items that help define larger structures are more homogeneous.

For example, the key arches in this picture have some patterns within them, but these vary less in texture, style, color, and theme, so that the arch itself can be more clearly visible. A similar pattern happens for the art in the spaces between the arches: works in larger spaces can be more complex, with more variations in textures, styles, colors, and themes. In contrast, works in smaller spaces are more constrained to fit well into the patterns around them.

Higher status artists are allocated to fill the larger spaces, where artists are allowed more discretion. This is a sense in which status is correlated with creativity. But its not at all that all artists are being as creative as they can, with the highest status artists capable of the most creativity. Instead, artists are only allowed to be more creative when they are higher status. That is, we don’t so much like high status because that indicates an ability to be creative, we instead like to see creativity because that indicates that the creator was high status.

Now consider this as a metaphor for academia and related intellectual worlds. These worlds make many products that fit into many structures on many different scales, including fields, subfields, topics, theories, and methods. Imagine that the world mainly wants all these products to fit into a pleasing aesthetic structure with an overall classic artistic style. When an individual makes a product and proposes to put it in some particular place within this overall structure, that proposal is accepted or rejected largely on the basis of how well it improves the overall artistic composition.

If so, most individuals will be rewarded for making impressive small variations. For example, if lots of people are talking about how AI will soon take all the jobs, and how a UBI could solve that, then aspiring intellectuals are rewarded for talking about modest variations on such topics. It is not that hard to be very critical of such scenarios, or to talk about very different future problems and solutions. And yes that might create more social value in terms of intellectual progress. But talking about those things clashes with the rest of the conversation, and so doesn’t make as aesthetically pleasing a whole. So you’ll instead want to be the most witty, clever, articulate, inspiring, rigorous person who talks about relatively small variations on what others are talking about.

That is, as an aspiring intellectual, you should mainly imagine yourself bidding to make one of the tiny artworks in the picture above. You’ll need to both stand out in some way, but also make something that fits very closely with nearby works. You may be capable of far greatly creativity, which could lead to greater intellectual progress. At least if anyone were to listen to you, and be tempted to build on your work. But if you aren’t one of the most impressive people doing lots of stuff that fits aesthetically with what others are doing nearby, its quite likely that no one will listen to you.

If you succeed on that usual path, you may someday be allowed more creativity, to contribute something bigger. Perhaps even something that produces real intellectual progress. And then future historians may say yay, what a great system that gives the biggest rewards to the most creative, surely it must be designed to maximize intellectual progress. Which if you are paying attention, you’ll know to be bull. Though you’ll probably also know to keep quiet about, as most everyone around you prefers a more flattering view of how their world contributes to intellectual progress.

Why Weakly Enforced Rules?

While some argue that we should change our laws to open our borders, it is more common for pro-immigrant folks to argue for weaker enforcement of anti-immigration laws. They want fewer government agencies to be authorized to help enforcement, fewer resources to go into finding violators, and weaker punishment of violators. Similar things happen regarding prostitution and adultery; many complain about enforcement of such laws, and yet don’t support eliminating them.

The recently celebrated “criminal justice reform” didn’t make fewer things illegal, or substitute more efficient forms of punishment (eg torture, exile) for less efficient prison. It mainly just reduces jail sentence durations. When I probed supporters, they confirmed they didn’t want fewer things illegal or more efficient enforcement.

The policing reforms that many want are not to substitute more cost-effective enforcers such as bounty hunters, or stronger punishments against police misconduct, but to instead just have police do less: pull over fewer drivers, investigate fewer suspects, etc.

When I claim that stronger norm enforcement is a big advantage of legalized blackmail, many people say that’s exactly the problem; they want less enforcement of common norms. For example, Scott Sumner:

Great literature and great films often turn people violating society’s norms into sympathetic characters, especially when they are ground down by “the machine”. I suspect that the almost universal public opposition to legalizing blackmail reflects society’s view (subconscious to be sure) that enforcing these norms (especially for non-criminal activities) requires a “light touch”, and that turning shaming into an highly profitable industry will do more harm than good. It will turn society into a mean, backstabbing culture. The people hurt most will be sensitive good people who made a mistake, not callous gang members who don’t care if others think they are evil.

On the surface, all of these positions seems puzzling to me; if a norm or law isn’t worth enforcing well, why not eliminate it? Some possible explanations:

  1. People like the symbolism of being against things they don’t really want to stop. It is more about wanting to look like the sort of person who doesn’t fully approve of such things.
  2. Having more rules that are only weakly enforced allows the usual systems more ways to arbitrarily punish some folks via selective enforcement. You might like this if you share such system’s tastes re who to arbitrarily punish. Or if you want to signal submission to authorities who want to use such power.
  3. If these things were actually legal and licit, people might sometimes publicly suggest that you are engaging in them. But if they are illicit or illegal, there’s a norm against accusing someone of doing them without substantial evidence. So if you want to discourage others from lightly accusing you of such things, you may want those activities to be officially disapproved, even if you don’t actually want to discourage them.
  4. We mainly want these norms and laws to help us deal with some disliked “criminal class” out there, a class that we don’t actually interact with much. So when we see real cases in our familiar word, they seem like they are not in that class, and thus we don’t want our norms or laws to apply to them. We only want less enforcement for folks in our world.
  5. What else?

Added 26Feb: I clearly didn’t communicate well in this post, as many commenters and this responding post saw me as arguing that all punishment, conditional on being caught and convicted, should either be zero or max extreme (eg death). Yes of course it is often reasonable to use intermediate punishments.

But enforcement also includes a chance of being caught, not just a degree of punishment, and there are issues of the cost-effectiveness of the processes to catch and punish people. There are many who want less punishment if caught, and less chance of catching, for most all offenses, and don’t want more cost effective catching or punishment, for fear that this might lead to more catching or punishing. To me, this seems hard to explain via just thinking that we’ve overestimated the optimal punishment level for some particular offenses.

Added 3Mar: A striking example is how in WWI recruits were supposed to be age 19 or older, but it was easy to lie and get in at younger ages, and most everyone knew of someone who had done this. We tsk tsk about child soldiers elsewhere, but don’t seem much ashamed of our own.

Enforce Common Norms On Elites

In my experience, elites tend to differ in how they adhere to social norms: their behavior is more context-dependent. Ordinary people use relatively simple strategies of being generally nice, tough, silly, serious, etc., strategies that depend on relatively few context variables. That is, they are mostly nice or tough overall. In contrast, elite behavior is far more sensitive to context. Elites are often very nice to some people, and quite mean to others, in ways that can surprise and seem strange to ordinary people.

The obvious explanation is that context-dependence is gives higher payoffs when one has the intelligence, experience, and social training to execute this strategy well. When you can tell which norms will tend to be enforced how when and by whom, then you can adhere strongly to the norms most likely to be enforced, and neglect the others. And skirt right up to the edge of enforcement boundaries. For weakly enforced norms, your power as an elite gives you more ways to threaten retaliation against those who might try to enforce them on you. And for norms that your elite associates are not particularly eager to enforce, you are more likely to be given the benefit of the doubt, and also second and third chances even when you are clearly caught.

One especially important human norm says that we should each do things to promote a general good when doing so is cheap/easy, relative to the gains to others. Applied to our systems, this norm says that we should all do cheap/easy things to make the systems that we share more effective and beneficial to all. This is a weakly enforced norm that elite associates are not particularly eager to enforce.

And so elites do typically neglect this system-improving norm more. Ordinary people look at a broken system, talk a bit out how it might be improved, and even make a few weak moves in such directions. But ordinary people know that elites are in a far better position to make such moves, and they tend to presume that elites are doing what they can. So if nothing is happening, probably nothing can be done. Which often isn’t remotely close to true, given that elites usually see the system-improving norm as one they can safely neglect.

Oh elites tend to be fine with getting out in front of a popular movement for change, if that will help them personally. They’ll even take credit and pretend to have started such a movement, pushing aside the non-elites who actually did. And they are also fine with taking the initiative to propose system changes that are likely to personally benefit themselves and their allies. But otherwise elites give only lip service to the norm that says to make mild efforts to seek good system changes.

This is one of the reasons that I favor making blackmail legal. That is, while one might have laws like libel against making false claims, and laws against privacy invasions such as posting nude picts or stealing your passwords, if you are going to allow people to tell true negative info that they gain through legitimate means, then you should also let them threaten to not tell this info in trade for compensation.

Legalized blackmail of this sort would have only modest effects on ordinary people, who don’t have much money, and who others aren’t that interested in hearing about. But it would have much stronger effects on elites; elites would be found out much more readily when they broke common social norms. They’d be punished for such violations either by the info going public, or by their having to pay blackmail to keep them quiet. Either way, they’d learn to adhere much more strongly to common norms.

Yes, this would cause harm in some areas where popular norms are dysfunctional. Such as norms to never give in to terrorists, or to never consider costs when deciding whether to save lives. Elites would have to push harder to get the public to accept norm changes in such areas, or they’d have to follow dysfunctional norms. But elites would also be pushed to adhere better to the key norm of working to improve systems when that is cheap and easy. Which could be a big win.

Yes trying to improve systems can hurt when proposed improvements are evaluated via naive public impressions on what behavior works well. But efforts to improve via making new small scale trials that are scaled up only when smaller versions work well, that’s much harder to screw up. We need a lot more of that.

Norms aren’t norms if most people don’t support them, via at least not disputing the claim that society is better off when they are enforced. If so, most people must say they expect society to be better off when we find more cost-effective ways to enforced current norms. Such as legalizing blackmail. This doesn’t necessarily result in our choosing to enforce norms more strictly, though this may often be the result. Yes, better norm enforcement can be bad when norms are bad. But in that case it seems better to persuade people to change norms, rather than throwing monkey-wrenches into the gears of norm enforcement.

So let’s hold our elites more accountable to our norms, listen to them when they suggest that we change norms, and especially enforce the norm of working to improve systems. Legalized blackmail could help with getting elites to adhere more closely to common norms.

Aliens Need Not Wait To Be Active

In April 2017, Anders Sandberg, Stuart Armstrong, and Milan Cirkovic released this paper:

If a civilization wants to maximize computation it appears rational to aestivate until the far future in order to exploit the low temperature environment: This can produce a 1030 multiplier of achievable computation. We hence suggest the “aestivation hypothesis”: The reason we are not observing manifestations of alien civilizations is that they are currently (mostly) inactive, patiently waiting for future cosmic eras. This paper analyses the assumptions going into the hypothesis and how physical law and observational evidence constrain the motivations of aliens compatible with the hypothesis. (more)

That is, they say that if you have a resource (like a raised weight, charged battery, or tank of gas), you can get at lot (~1030 times!) more computing steps out of that if you don’t use it  today, but instead wait until the cosmological background temperature is very low. So, they say, there may be lots of aliens out there, all quiet and waiting to be active later.

Their paper was published in JBIS in a few months later, their theory now has its own wikipedia page, and they have attracted at least 15 news articles (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15). Problem is, they get the physics of computation wrong. Or so says physics-of-computation pioneer Charles Bennett, quantum-info physicist Jess Riedel, and myself, in our new paper:

In their article, ‘That is not dead which can eternal lie: the aestivation hypothesis for resolving Fermi’s paradox’, Sandberg et al. try to explain the Fermi paradox (we see no aliens) by claiming that Landauer’s principle implies that a civilization can in principle perform far more (∼1030 times more) irreversible logical operations (e.g., error-correcting bit erasures) if it conserves its resources until the distant future when the cos- mic background temperature is very low. So perhaps aliens are out there, but quietly waiting.

Sandberg et al. implicitly assume, however, that computer-generated entropy can only be disposed of by transferring it to the cosmological background. In fact, while this assumption may apply in the distant future, our universe today contains vast reservoirs and other physical systems in non-maximal entropy states, and computer-generated entropy can be transferred to them at the adiabatic conversion rate of one bit of negentropy to erase one bit of error. This can be done at any time, and is not improved by waiting for a low cosmic background temperature. Thus aliens need not wait to be active. As Sandberg et al. do not provide a concrete model of the effect they assert, we construct one and show where their informal argument goes wrong. (more)

That is, the key resource is negentropy, and if you have some of that you can use it at anytime to correct computing-generated bit errors at the constant ideal rate of one bit of negentropy per one bit of error corrected. There is no advantage in waiting until the distant future to do this.

Now you might try to collect negentropy by running an engine on the temperature difference between some local physical system that you control and the distant cosmological background. And yes, that process may go better if you wait until the background gets colder. (And that process can be very slow.) But the negentropy that you already have around you now, you can use that at anytime without any penalty for early withdrawal.

There’s also (as I discuss in Age of Em) an advantage in running your computers more slowly; the negentropy cost per gate operation is roughly inverse to the time you allow for that operation. So aliens might want to run slow. But even for this purpose they should want to start that activity as soon as possible. Defensive consideration also suggest that they’d need to maintain substantial activity to watch for and be ready to respond to attacks.