SIAI Matching Challenge: Choose Your Own Research Project

Between now and 28 Feb. 2010, SIAI is offering a matching-funds challenge up to $100,000. Intriguingly, donors can choose particular grant proposals to fund -- including, e.g., papers on the following topics:

• "The Coherence of Human Goals"
• "AI Risks Philanthropy: How Many Lives Can We Save per Dollar?"
• "Digital Intelligences and the Evolution of Superorganisms"

In addition, the page explains that donors contributing at least $1K can contact Anna Salamon to discuss the possibility of a new research topic. So, utilitarians: If you're interested in donating and have a project in mind, do contact Anna and see what can be done. SIAI might, for instance, fund an exploration of the types of suffering computations we decide to care about. Or perhaps a paper assimilating research on some aspect of mathematics, physics, computer science, economics, psychology, or cognitive science that is crucially important to know about when trying to reduce large-scale suffering.

Each paper has an associated expected cost figure, like this one for the anthropics article:

Total budget: $5,960, including:

• Conference fees, air travel, motel: $1,400
• Costs for researcher time: $4,560
  

How research costs are estimated:

• Person-months for research and writing: 1.9 (obtained by taking our standard estimate[1] of 1.25 person-months per conference paper and multiplying by 1.5, since this paper requires thinking through and aggregating many different topics).

• Dollars required to support one skilled full time researcher-month[2]: $2,400

Excellent Introduction to SIAI

I quite enjoyed Anna Salamon's talk, "Shaping the Intelligence Explosion," from the Singularity Summit 2009. Unlike many futurist speakers and authors, Salamon presented basic statements about what motivates the Singularity Institute (SIAI) in a fashion free from a lot of the unecessary transhumanist baggage (pet concerns like life extension or multiple-universe hypotheses) that can turn away people from other backgrounds who fundamentally care no less about these issues.

Salamon presented (~1:17 in the video) "Four Key Claims":

1. Intelligence can radically transform the world.
2. An intelligence explosion may be sudden.
3. An uncontrolled intelligence explosion would kill us and destroy practically everything we care about.
4. A controlled intelligence explosion could save us, and protect practically everything else we care about. It is difficult, but worth the attempt.

I'm personally rather skeptical that an intelligence explosion will ever occur -- indeed, I may assign the scenario a very low probability. On the other hand, if one did occur, the magnitude of its impact on our region of the cosmos would be so profound that I think focusing our efforts preparing for such possibilities has high expected value. (Think about why you wear a seat belt the next time you drive to your friend's house down the street.) I liked the way Salamon explained SIAI's core mission as something that almost anyone, even skeptics like me, ought to care about -- not just computer geeks and sci-fi aficionados. (As far as the potential plausibility of intelligence explosion itself, I do think the discussion around 18:00 of whole-brain emulation and the Hansonian takeoff scenario was well done.)

Of course, SIAI is fundamentally an academic organization, and most of its research is highly valuable whether or not an "intelligence explosion" ever occurs. Indeed, I encourage donations to SIAI mainly to fund projects that will help us better understand how to reduce massive amounts of suffering in our multiverse. SIAI explores fundamental questions about physics, Bayesian statistics, anthropics, decision theory, infinitarian consequentialism, consciousness, and cognitive science need to be studied regardless of what happens with AI.

Finally, readers may be interested in this other post on SIAI's matching-grant challenge, in which donors can choose their own research projects to support.

Procrastination: "Being in the Mood"

Feeling Good by David Burns has a nice discussion of why people procrastinate. I particularly enjoyed this piece of advice:

Motivation does not come first, action does! You have to prime the pump. Then you will begin to get motivated, and the fluids will flow spontaneously. [...] Individuals who procrastinate frequently confuse motivation and action. You foolishly wait until you feel in the mood to do something. Since you don't feel like doing it, you automatically put it off. (qtd. in Bonnie Runyan McCullough, Totally Organized: The Bonnie McCullough Way, p. 52)

I proffer some additional notes. They're all pretty obvious, but I find that I benefit from reminding myself of them frequently.

• Decide on goals and tasks; then act. The point of avoiding procrastination is to get important things done. This requires that you know (1) what you consider important (your objective function) and (2) how best to achieve your goals (what your tasks are). Decide those first; then use Burns's anti-procrastination technique to do the highest-value tasks.
  
• Update your to-do list over time. There's no need expending willpower to accomplish unimportant tasks, even if they're included on a to-do list. As you learn more and as priorities change, update the ordering of the to-do list. Drop old tasks that you once found important but now do not.
• Don't "save work for later" unless you're sure you'll get it done later. This is the old "never do tomorrow what you can do today" maxim. There are times when I'm tempted to put off a high-value task for later (like checking emails) because I know it'll be fun. This is sometimes a good idea, but not if it causes a backlog of tasks to build up over time. Then finishing them all becomes a chore. And in general, I've found "there's plenty more where that came from," i.e., I can pretty much always find some high-value task for which I'm in the mood, and I don't need to save particular fun tasks for that purpose.
  
• Avoiding procrastination allows for more time when you don't need to override your mood. In many cases, if there's no hard deadline or stark difference in productivity value between several actions, you ought to do the one that you most want to do, knowing that you'll probably want to do others later on. And if not, you can expend some willpower to get yourself started on them at a later point. Such is the luxury afforded by not being late in finishing time-sensitive tasks: You don't need to constantly expend willpower in forcing yourself to do the next put-out-the-fire item on your list.
  

Reflecting on Your Cognitive Algorithms

This post is largely a personal musing; the substantive content has been discussed elsewhere by many other authors.

One of the things that has most transformed the way I look at the world has been cognitive science, specifically the philosophical understanding that grounds it: Seeing the brain as a collection of cognitive algorithms running on biological hardware. This focus on not just what the brain does but how it might do it is fundamentally transformative.

For as long as I can remember, I had known about the types of psychological facts commonly reported in the news: For instance, that this particular region of the brain controls this particular function, or that certain drugs can treat certain brain disorders by acting in certain ways. And it's basic knowledge to almost everyone on the planet that operations inside the head are somehow important for cognitive function, because when people damage their brains, they lose certain abilities.

While I knew all of this abstractly, I never thought much about what it implied philosophically. I saw myself largely as a homunculus, a black box that performed various behaviors and had various emotions over time. Psychology, then, was like macroeconomics or population biology: What sort of trends do these black boxes tend to exhibit in given circumstances? I didn't think about the fact that my behaviors could be reduced further to particular cognitive-processing steps inside my brain.

Yet it seems pretty clear that such a reduction is possible. Think about computers, for instance. Like a human, a computer exhibits particular behaviors in particular circumstances, and certain types of damage cause certain, predictable malfunctions. Yet I don't think I ever pictured a computer as a distinct inner self that might potentially have free will; there were no ghosts or phantoms inside the machine. Once I had some exposure to computer architecture and software design, I could imagine what kinds of operations might be going on behind, say, my text-editor program. So why I did I picture other people and myself differently? My conceptions reflected how an algorithm feels from inside; I simply stopped at the basic homunculus intuition without breaking it apart.

Picturing yourself as a (really complicated and kludgey) computer program casts life in a new light. Rather than simply doing a particular, habitual action in a particular situation, I like to reflect upon, What sort of cognitive algorithm might be causing this behavior? Of course, I rarely have good answers -- studying that is what cognitive science is for -- but the fact that there is an answer soluble in principle gives a new angle on my own psychology. It's perhaps like the Buddhist notion of looking at yourself from the outside, distanced from the in-the-trenches raw experience of an emotion. And, optimistically, such a perspective might suggest ways to improve your psychology, perhaps by adopting new cognitive rituals. That is, of course, what self-help books have done for ages; the computer analogy (e.g., "brain hacks" or "mind hacks," as they're sometimes called) is just one more metaphor for describing the same thing.

Related is the realization that thought isn't a magical, instantaneous operation but, rather, requires physical work. Planning, envisioning scenarios, calculating the results of possible actions, acquiring information, debating different hypotheses about the way the world works, proving theorems, and so on are not -- as, say, logicians or economists often imagine them -- immediate and obvious; they involve computational effort that requires moving atoms around in the real world. For instance, the fact that you considered an option and then disregarded it is not a "wasted effort," because there's no other way to figure out the right answer than actually to do the calculation. Similarly, you're not at fault for failing to know something or for temporarily holding a misconception; the process of acquiring correct (or at least "less wrong") beliefs about the world requires substantive computation and physical interaction with other people. Changing your opinions when you discover you're in error isn't something to be embarrassed about -- it's an intrinsic step in the algorithm of acquiring better opinions itself.

Pain-free Animals?

The current Vegan Outreach newsletter contains a link to a New Scientist piece (as well as an unfortunate editorial) based on a fascinating article: "Knocking Out Pain in Livestock: Can Technology Succeed Where Morality has Stalled?" by Adam Shriver. The moral urgency of such a proposal seems to me obvious, so I was most interested in the discussion of its scientific plausibility.

Shriver presents two example proposals for what might be done. First, we might

create knockouts of other mammals (cows and pigs for starters) lacking the AC1 and AC8 enzymes. Interfering with the cAMP cycle in the brain reduces the affective dimension of chronic or persistent pain, rather than pain full stop, but this would still be an improvement over current circumstances. If we could eliminate the sensitization that occurs as a result of painful or traumatic experiences, the animals would still be better off than they are now.

Secondly,

Zhou-Feng Chen and colleagues searched the Allen Brain Atlas to find genes that were highly expressed in the ACC but not other areas of the brain [29]. One strong candidate was the peptide P311. The researchers created knockout mice lacking the expression of P311 and found that heat and mechanical sensitivity were normal in the animals. However, they then performed a conditioned place aversion test on the animals and found that the knockouts no longer demonstrated the conditioned place aversion caused by formalin injections, in stark contrast to control rats. Thus, at first glance, it appears that knocking out P311 in mice strongly diminishes the affective dimension of pain while keeping acute responses intact.

Furthermore, P311 is likely to play a similar role in all mammals (Chen, personal communication), so one presumably could engineer other mammals that have a reduced affective dimension of pain while maintaining the sensory dimension of pain.

Since I'm even more interested in wild-animal suffering than farm-animal suffering, in view of the vast difference in numbers of animals involved, my immediate question was whether similar techniques might one day be applicable there. Doing so is a lot trickier, because evolution produced the badness of pain for a reason. Shriver mentions this concern:

Since it seems likely that the affective dimension of pain played some role in determining the evolutionary fitness of organisms, we might question whether knockout livestock could really survive up through the point where they are normally slaughtered. However, it appears that the experimental rats were able to survive without complication at least in their cages (Chen, personal communication). This would be a good model for sows or veal calves who spend most of their lives confined in small pens where they can’t do much of anything that would injure or otherwise harm themselves.

Producing genetically fit wildlife without pain might require not just knocking out pain but replacing the "pain" - "pleasure" axis with a "less pleasure" - "more pleasure" axis, which could be much more difficult.

I mentioned that Shriver's proposal seems obviously valuable from my perspective, but unfortunately this isn't necessarily the case among the general public. As the New Scientist article notes:

[Alan] Goldberg also contends that public attitudes may make pain-free livestock a non-starter. He and colleague Renee Gardner conducted an online survey on the use of pain-free animals in research and found little public support, even among researchers who experiment on animals (Alternatives to Animal Testing and Experimentation, vol 14, p 145).

This underscores the importance of public outreach to change hearts and minds about wild-animal suffering and how it could be prevented.

Ham and Eggonomics, Part 2

Below are some quotes from and comments on Ch. 8 of the Ham and Eggonomics book introduced in Part 1.

The authors begin with a discussion of the fallacy that "my individual purchases don't matter." They give a nice illustration:

Suppose that we take 5,280 [one foot] rulers and placed them in a straight line, end to end. This line of rulers would then be one mile long. It would appear as one long line, and if you could view the entire mile of rulers from above, you would not be able to see one single ruler. If you removed one ruler, the line would grow shorter; there is no doubt as to that. Viewed from above, removing one ruler would not appear to have any effect on the line—but again, it does.

Of course, the analogy to food purchases isn't quite accurate. In practice, as discussed here, an individual's purchasing choice is extremely unlikely to change the number of animals raised, because food is produced and sold in bulk units. However, in the event that a consumer does have an effect, that effect will be huge. Thus, in ignorance of whether a particular purchase is the one that "breaks the camel's back," the expected values of each purchase do add in the same way as the rulers. (To the extent that this reality may be de-motivating for potential vegetarians, perhaps it's better not to mention it too much?)

Pages 3-4 contain a nice discussion of the relevance of elasticities to the question of how an individual's purchases affect the quantity supplied by the market. The authors argue that the supply curves for beef, and to a lesser extent milk, are likely inelastic, while those for pork, and probably also chicken and eggs, are probably relatively elastic. More elastic supply means a bigger change in production when consumer behavior changes. Thus, for instance, abstaining from eating 1 kilogram of chicken has a bigger expected impact on the kilograms of chicken produced than abstaining from 1 kilogram of beef has on the kilograms of beef produced, other things being equal. As far as demand elasticity, the studies that the authors have done suggest a slightly bigger kilogram-for-kilogram impact of abstaining from chicken, pork, veal, and milk relative to beef or eggs. The total results--combining information about supply elasticities and demand elasticities--are shown in Figure 8.2 (see this document for the Chapter 8 figures), which I've reproduced below in sorted order:

If [someone] gives up Total Consumption of ... the Product Falls By ...
One Pound of Milk 0.56 lbs
One Pound of Beef 0.68 lbs
One Pound of Veal 0.69 lbs
One Pound of Pork 0.74 lbs
One Pound of Chicken 0.76 lbs
One Egg 0.91 egg.

Of course, this is not the end of the story. Other (often more important) factors to consider when deciding on dietary purchases include the quality of the lives of animals of different types, and the number of animals required to produce a given quantity of meat, counting both the animals themselves and their parents. Pages 5-6 explain Bailey's own views on the quality of life of various farm animals on a scale of -10 to 10 (see Figure 8.4).

In particular, Bailey thinks some farm animals have lives worth living. Looking at only the non-breeder animals, these are cows (+6), AWA-certified pork (+4), broiler chickens (+3), and cage-free hens (+2). On broiler chickens, Bailey says they "have a life worth living, but because of their leg problems and confined environment, do not fare as well as beef cattle" (p. 5). I'm more skeptical that broilers on average enjoy their lives, but even if they do, I would still be wary of giving them positive welfare because the painfulness of slaughter has to be considered. I personally wouldn't want to live even a mildly pleasant life for only 45 days if it meant that I would afterwards endure slaughter. This is probably true even if I were given electrical stunning and definitely true if I were one of the birds for which the stunning was not effective.

I assume Bailey has included the painfulness of death in his numbers, but it would be good to make this explicit. Otherwise, many readers will just imagine what a broiler chicken looks like during a typical moment of its life, multiply that by ~42 days of life, and conclude that the total experience is positive. An explicit mention of the relevance of lengths of lives would be helpful as well; indeed, this consideration makes Bailey's numbers seem a little odd. How can a beef cow, which lives for 402 days (see the "Beef" section here) have only twice the total happiness (+6 instead of +3) of a broiler chicken that lives 42 days? If Bailey's calculations do involve multiplication of his welfare numbers by lifespan, I missed that part of the text. In any event, doing straight multiplication in that way would still be misleading because it ignores the painfulness of death at the end of a life, unless stress during transport and slaughter has been implicitly incorporated into the per-day average.

Pages 6-7 discuss the impacts of farming on wild animals. In my view, this is the most important part of the calculation, especially if we give more than vanishingly small probability to insect sentience. Of course, if insects' short lives aren't worth living, then it's not clear that pesticide use in crop production represents a net harm (though whether it is or not, the chemicals could still potentially be made less painful).

As they did in Chapter 6, the book authors comment on the possibility that bigger wild animals also suffer enormously:

Animal rights groups tend to romanticize the life of animals in the wild, but anyone who has watched wildlife documentaries can attest to the cruelty of nature. We ask you, the reader, would you rather be a Wildebeest in Africa who must constantly roam for food, always in pursuit by lions and crocodiles, or would you rather be a cow in the U.S., or a hog in the U.S.? (pp. 6-7)

The authors continue with a discussion of the consideration, How many animals does it take to produce a given quantity of meat? This is the primary variable of interest in my own calculations of suffering per kilogram of meat, but the book authors do a more thorough job, by including numbers of parent animals that need to be raised, as well as the efficiency of production under various conditions. An example of the latter is that cage-free hens produce fewer eggs per week than caged hens.

In fact, this last point is rather important to the question of whether to purchase cage-free eggs (or, at least, whether to encourage others to do so). As the authors explain (p. 11), if you believe that both caged and cage-free hens suffer and that cage-free hens suffer at least ~2/3 times as much as caged hens, then for efficiency reasons, caged-hen eggs entail less total suffering than cage-free-hen eggs. However, Bailey's personal opinion (see Figure 8.4) is that cage-free hens have net happy lives, in which case cage-free eggs would clearly be preferable.

The chapter continues with interesting discussions of public attitudes toward factory farming as assessed by a nationwide telephone survey, as well as how one could compute willingness-to-pay for animal welfare by different consumers. The Appendix describes the mechanics of how elasticities can be used when assessing changes in the quantity of a good supplied, as well as the detailed calculations of how many animals need to be raised to produce a given amount of meat.

Toward the end of the main chapter, the authors make a disturbing comment, though perhaps not one that comes as a surprise:

There is evidence to believe that many Americans simply do not care very much about the well-being of farm animals. In our conversations with 300 individuals from three cities in the U.S., one-third told us that they would rather not know how farm animals are raised. They simply want to continue consuming their delicious, safe, and inexpensive food without worrying about whether the animals that provide that food suffer. (p. 17)

I hope to get a chance to read more of the book at some point; it contains lots of high-quality and thoughtful discussion. Thanks to the authors for writing it!

Ham and Eggonomics, Part 1

Bailey Norwood is a professor of agricultural economics at Oklahoma State University. Much of his recent research on food preferences and farm-animal care is of interest to the animal-welfare community. For instance, one of his papers examines economic impacts of switching to vegetarian diets, giving consideration to the supply-and-demand interrelationships of the chicken, pork, beef, and corn markets.

One of Norwood's recent projects is a website called Ham and Eggonomics, which includes an associated blog. In the page's own words:

This website is intended to be an information source for consumers about the food they eat, with specific emphasis on the welfare of farm animals. Finding good, useful information is difficult. Scientists are objective but often to timid to make general statements. Food industries have their obvious bias. Even authors who write books about food generally distort and sensationalize the truth to scare consumers.

Ham and Eggonomics states the facts plainly and is not scared to tackle an issue head-on. The website only desires to provide you with the information you require for your food choices, and has no desire to tilt your political leanings or recruit you into a specific cause.

The page features the a book with the same name as the site. The book is in draft form, and "comments are welcome," so I'll make a few remarks on Chapters 1 and 6 below. In Part 2, I comment on Chapter 8.

Chapter 1 lays out the slant from which the authors approach animal welfare: Namely, the economics of revealed preferences, rather than philosophical prescription.

Whereas ethicists and philosophers will argue about the morality of treating animals in certain ways, the economic approaches assumes that each person is entitled to their own view of animal welfare. (Ch. 1, p. 11)

The economic approach is to instead treat peoples’ views on animal welfare as preferences. That is, instead of arguing with you about what is right or wrong, or trying to persuade you to adopt a particular system of moral beliefs, we simply ask you, and your fellow citizens, what you think is right and wrong and make that the goal of society. For example, if instead of animal welfare we were dealing with the contentious issue of abortion, we would not try to persuade you to be for or against abortion. Instead, we would poll the public, and if a majority of Americans was against abortion, we might would conclude that public policy should outlaw abortion. That is just an example, an example to illustrate that we seek to determine how animals should be raised, based on the moral and ethical beliefs of you and your fellow citizens. (Ch. 1, p. 3)

Needless to say, I don't agree with this approach, because it conflates prescription about what should be done with description of the desires held by an arbitrary group of agents (adult Americans, in rough proportion to their voting / purchasing power). I demur at the suggestion that "animals matter because people care about them" (Ch. 1, p. 5). But my objection here is no different from my disagreement with the standard economic approach in general, and I'm glad the authors make their bias explicit. In addition, I appreciate the authors' aim to make reasoned decisions based on academic research into the costs and benefits of particular actions, without being afraid to upset standard ideological positions. (We merely differ on what to measure when assessing costs and benefits.)

The tone of the book is very "mainstream":

In this book, we offer no grand moral philosophy that makes your food choices straightforward. We do not believe ourselves so smart and knowledgeable that we can dictate to you the choices you can make. We can, however, help you understand the consequences of your choices, and help you better form your views on farm animal welfare. (Ch. 6, p. 20)

This may ultimately prove beneficial, because the message is not off-putting to majority of Americans in the way that much animal-rights literature is.

I largely agree with the authors that "the study of animal welfare is, for all practical purposes, the study of farm animal welfare" (Ch. 1, p. 4) because farm animals vastly outnumber those used in labs or for fur. Indeed, I would go further and say that the study of animal welfare is, for all practical purposes, the study of chicken welfare, because the number of pigs and cows slaughtered is tiny by comparison.

I do wish this point were recognized more by animal-welfare organizations, which -- even when they recognize the importance of farm animals -- often continue to focus resources on, say, welfare standards for veal calves and sows. Even the authors of Ham and Eggonomics fall into this, by giving roughly equal time to welfare questions about the raising of cows, pigs, and chickens. Really, the only animals people should think about when discussing land-based factory farms are broiler chickens and laying hens.

The above statement ignores fish and crustaceans, which may also suffer in huge numbers. And of course, it ignores wild animals. The book authors are not silent on this point, however:

It would also be foolish romanticism to believe that the lives of all domesticated animals are inferior to their ancestors or to modern animals that live in the wild. Domesticated animals live without the fear of predators and have ample supply of food and water; one need only turn their television to the Discovery channel to see that wild animals live in constant fear of being mauled and are no strangers to hunger and starvation. (Ch. 1, p. 9)

Some animal rights groups are even opposed to the owning of animals as pets. They only wish animals to exist in their wild state, which as any avid watcher of nature shows can attest, has its own distinct forms of cruelty. Wild animals, though adapted to their environment, face many obstacles to receiving adequate nutrition, and most face constant pursuit as prey. In our opinion, wild animals do not have a high level of well-being. (Ch. 6, p. 19)

The authors go on to cite my piece on wild-animal suffering on p. 19. (The blog mentions it as well.) They suggest an interesting conclusion, one that I probably agree with:

It is possible, and to us likely, that the only manner in which animals can enter this world and for their species as a whole to experience more happiness than suffering is for them to be under the stewardship of humans. Whether they are pigs destined for slaughter or dogs destined for doting, nature’s hand has no choice but to be cruel, while humans have the choice to become compassionate stewards. If an animal species is to exist, between [sic.: better] for them to exist under the care of humans, generally speaking. (Ch. 6, p. 19)

This is perhaps unsurprising: Evolution does not optimize an objective function of cumulative happiness minus pain. (Of course, neither do almost all breeders of farmed animals.) Human engineers with the specific goal of reducing suffering and promoting wellbeing ought to be able to do better, unless they are too incompetent to succeed at the task. In practice, incompetence may very well prevail with respect to wild animals at the moment (evolution, for all its "blind, pitiless indifference," may do a better job at currently evolved optima), but this needn't be the case indefinitely, as human technology and intelligence progress.

Vegans who (Sadly) Support Wilderness

The forum of the Vegane Gesellschaft Österreich or Vegan Society Austria features an interesting discussion of wild-animal suffering from September-October 2008. Here is the link to the first page in German. Below I've copied some comments that I thought were worth highlighting, converted to English using Google Translate.

The topic was raised by "mars mensch":

I'm vegan because I do not want the animals to suffer. I think veganism and animal rights, the idea is to spread a very effective way the suffering of farm animals (and perhaps humans) to be reduced.
But what about wild animals? I can hardly anything on the Internet can find. There are most probably suffering in wilderness. Wild animals suffer from all kinds - hunger, thirst, cold, disease, parasites, injuries, .. depending on the type and probably also fear, grief, stress, sexual frustration ..
Somehow, nobody seems to be wondering about it. I do not think of people suffering caused is worse than "natural", as some seem to perceive. Suffering is suffering.
So the question is: What can be done against the suffering in the wilderness do I have to honestly have no clue.

here are two interesting links on the topic, but also without any real answers:
http://utilitarian-essays.com/wild-animals.pdf
http://www.felicifia.com/showDiary.do?diaryId=161

Some respondents suggested that wild animals may not need help:

I think animals are in the wilderness pretty well. You can provide for themselves or a family, have an enormous amount of space to move and they die at some point. The fact that they are not always made great must simply accept, as people in affluent societies, which have sufficient food and shelter often beyond problems. (Heavy Heavy Low Low, p. 1)

but the few animals are still wild eg: on an island, does not need our help, because the suffering or death keeps everything in balance. (The Andi, p. 1)

I like the reply that mars mensch gave (p. 1):

I think some here have a rather romantic idea of the transformed nature or seem to think that nature is good and man is the source of the evil in the world. This is also anthropozentisch and is not true. Nature is cruel.
Animals are not perfectly adapted to the environment. Many survive the winter because they do not starve or freeze. Dry periods and droughts regularly delete almost complete populations. Diseases and parasites are widespread. There are predators, etc., etc.
We encourage vegans to get us to that calves are taken away from their mothers. But also in the freedom of pups lose their parents and vice versa.
I really do not know what one against the suffering of wild animals can do, but we should in any case, start thinking about it. Perhaps the best thing you can do at the moment, to disseminate that anything should do.

Some further posters expressed the view that humans should leave wildlife alone:

And I personally will not if I now beginning with a game to feel compassion, precisely because it is free in nature can live and also the natural circumstances is exposed!
For me, this is just nature and wilderness: The life enjoy, with all its vagaries and dangers [...] I've also someday read a quote from someone: A world without suffering is a dead world! So somehow related to the natural sorrow which is not vermeitbar has this quote already in my opinion. (leather, p. 2)

when Veganism is not yet a matter of principle, suffering-free world to create (yes this is absolutely not in the range of our options), but only a question of our behavior to be corrected, so that we are sentient beings do not expect further suffering and cruel treated, just because we are stronger than they are. in their nature but we have already sufficient interfered - I personally think that the wildlife rather want us to leave, instead of us even in their affairs. (spaghetti, p. 2)

I appreciated this sentiment from "Mercy" (p. 1):

Whenever I am on television reports about natural disasters can see, such as earthquakes, floods, forest fires, then I think to the animals as victims! Of course, I also find the fate of the people who are affected gaaanz awful, no question, but then I always think that anyone who cares for the animals?

Famine and disease are "natural" in humans, yet most people seem to care about preventing them. Why is there a difference when these occur in animals, apart from the difficulty of the task?

mars mensch offered a nice concluding thought (p. 4):

Human rights are sometimes divided into several generations:

"Rights of the first" generation "refer to the classic civil and political freedom and participation rights, as established by the French Revolution were formulated. They are inter alia in the UN civil pact, or even in the European Convention on Human Rights established. The pact includes a civil general prohibition of discrimination as well as basic defensive and rights (right to life, prohibition of torture and inhuman or degrading treatment or punishment, prohibition of slavery and forced labor), the vital for the protection of human dignity are more then Civil freedom and political participation rights (personal freedom and security, thought, religion, expression, assembly, freedom of association, etc.) and justice-related rights (equality before the law, presumption of innocence, fair trial, etc.). The national and international protection for civil and political rights are still the most developed.

Rights of the second "generation" include Economic, Social and Cultural Rights (in short: WSK-human rights or social rights), which since the 19th Century in the wake of the industrial revolution emerged. The central point of reference of these rights is the UN social pact, which includes rights and labor, social security, nutrition, housing, water, health and education anchored. Long time these rights were not "real" human rights, but rather as a political target views, which - in contrast to civil and political rights - legally not sufficiently identifiable and judicially reviewable hardly had. Since the 1990s, however, the content and the violation of human rights, social events significantly concretised. Social human rights are now widely deemed to be politically and are by their nature as enforceable (tangible justiciability). "Http://www.fes.de/handbuchmenschenrechte/03-menschenrechte-einstieg.html

Maybe it will be in the animal rights as well give something. Only defensive rights against the people, or the right not to be imprisoned and tortured, as response to factory farming, etc. And then to also claim rights to help those whose suffering is not caused by humans. If the species boundaries in ethics exceeded only once, it's just logical.

It's nice to get a sense of how people in the animal-rights movement respond to the idea of wild-animal suffering. Though some are uninterested, there are probably more sympathetic souls among this community than the general population. Raising the topic in forums like this does seem to be a good way of making people think about the issue.

Caring about Animal Suffering

What are some examples of experiences that lead people to give serious concern to the suffering of animals? For Peter Singer, as he describes in Animal Liberation, it was logical argument by vegetarian friends that persuaded him to think about the issue. For Josh Balk, it was watching a video containing scenes of animals being killed. For Jon Camp, it was a college ethics course. Some people discovered animal cruelty by being stuck in traffic behind trucks bringing animals to slaughter. Many others have been affected by Vegan Outreach leaflets. In my own case, I simply needed to realize that animals could feel pain, a point that somehow escaped me until I came across an online excerpt from Animal Liberation (pp. 10-12, 14-15).

I ask this question because it seems to me that serious attention to animal suffering is one area that is perhaps most lacking from the mainstream intellectual sphere of moral concern. Everyone in the modern western intelligentsia seems to care about racism, homophobia, poverty, disease, and other conventional issues, yet many still fail to think twice about eating a turkey sandwich for lunch. I worry about how animals might fare in a post-human technological society -- not because I think many people would favor deliberate harm to animals, but simply because decisions with vast consequences for animal suffering might be made without giving animals a second thought.

The worst source of animal suffering is undoubtedly predation, disease, and death in the wild. A serious solution to this problem appears far off. Therefore, it may be that animal-welfare supporters can best address the wild-animal problem by promoting concern for animal suffering generally, both to ensure that animal welfare does have a place within the scope of concern of future humans and to hasten the development of technologies that might address the issue.

It's important, though, that utilitarians promote the right kind of concern for animals. Many in the animal-rights movement feel that animals inherently deserve to live in nature without disturbance from man; it would be a terrible tragedy to promote "animal liberation" in such a way as to increase the number of people demanding preservation of animals in their natural state because this is "the way things are supposed to be." Similarly, it's important that concern for animal suffering be linked with a mindset of cost-effectiveness analysis, lest well meaning people spend their lives opposing a few circuses and zoos while billions of farmed chickens suffer, to say nothing of the orders of magnitude more wild animals.

What are some ways to promote this kind of utilitarian concern for animals? I think Vegan Outreach brochures, as well as ads by The Humane League, are impressive and have strong economies of scale. They're very good for opening people up to concern for animals, although they aren't sufficient for the subset of the population -- animal activists and other philosophically minded thinkers -- who already agree that animal suffering matters and who are ready for the next step: Applying that concern to wild animals.

Some have suggested that in-vitro meat may reduce speciesism, on the grounds that many people resist giving moral consideration to animals precisely because of the undesirable consequences of doing so on their food choices. Perhaps the possibility of producing meat without factory farming would free people up to care more about animals than before, just as the advent of industrial technologies to replace slave labor may have facilitated moral opposition to slavery. That said, this approach seems less direct and less obviously helpful than veg outreach. Maybe if fewer people confront the dilemma of eating factory-farmed animals, there will be less of a "hook" for people to latch on to concern for animal suffering at all. (In this model, ethical vegetarianism is like a gateway drug to caring about suffering in nature.)

I think videos can be effective, and I find watching them to be one of the most effective ways to trigger the release of chemicals in my brain that motivate me to want to do something to prevent animal suffering. Perhaps one could make a video that includes images of brutality in nature. An important challenge would be finding a way to avoid leaving people feeling merely depressed, or leading them to conclude that concern for animals is hopeless because the suffering of wild animals is overwhelming and intractable. Indeed, arguments about "how far do you go with concern for animals?" are often used as reductios against vegetarianism. It may be that talking about the possibility of insect suffering will entirely put off many people who are just beginning to consider the idea of extending some moral concern to chickens. Different audiences are at different stages. Committed animal-welfare supporters definitely need to hear about insects, but for the majority, we should probably stick to chickens for now. :)

Formalizing "Shorter Programs"

In a separate discussion, my friend Carl said:

With respect to Occam assigning exponentially diminishing probability to special miracles, I tend to think of this in terms of the broad set of hypotheses to be considered: if my probabilities are to sum up to 1 I can't coherently assign all my 'the world is a lie' probability mass to whatever hypothesis has been brought to my attention in the last five minutes. The code of a short program can be contained in astronomically many ways within a larger program. An indifference principle gets you going from there.

By "short program," Carl is presumably referring to something like the minimum description length (MDL) principle for explaining our observations. I'm curious to know how exactly he's envisioning its application, though.

For purposes of illustration, consider this fictional scenario. My friend Joe calls me in the evening with a worried tone in his voice. He says, "I've got something to tell you. I was just brushing my teeth, when I heard a voice. It said, 'Joe, I have an important message. You need to write a special number on your toothpaste tube, or else your toothpaste will cease to work properly. That number is 2835023981. Do as I say and all will be fine.' Then the voice disappeared."

Now, there are lots of hypotheses we can imagine here. In particular, I'll consider 10^10 + 1 of them:

(0) My friend imagined the whole thing. Writing a number on the toothpaste tube will accomplish nothing.
(1) In fact, my friend needs to write a number on the tube, but that number is not the one he was told, but rather 0000000001.
(2) My friend needs to write not the number he was told, but 0000000002.
...
(2835023981) My friend needs to write the number he was told, 2835023981.
...
(10^10) My friend needs to write not the number he was told, but 1000000000.

Obviously, there are lots more hypotheses to consider -- e.g., that my friend needs to write a number bigger than 10^10, that it needs to include decimals, that he needs to write it on his forehead instead of his toothpaste tube, that he needs to eat green cheese instead of writing a number, and so on. But just these 10^10 + 1 hypotheses give a sense of the literally exponential number of potential complicated scenarios.

How does MDL evaluate each of these hypotheses? One suggestion I can imagine is as follows.

(0) This hypothesis just involves ordinary physics -- things like Maxwell's equations, or perhaps rules of string theory -- plus maybe some physical constants. Given those initial conditions, it would be possible in principle to compute the entire history of the universe, including the evolution of humans, the birth of my friend, and my reception of his phone call. (If the "data" to be explained here are my personal observations, then perhaps the program would also have to specify who I am. It could then compute the pattern of perceptual inputs I receive throughout my lifetime, including the auditory waves from the speaker of my phone with my friend's voice.)
(1) This hypothesis involves mostly ordinary physics, including all of the same information as before. However, it includes an extra specification that, contrary to ordinary physical law, my friend should start getting cavities unless he writes 0000000001 on his toothpaste.
(2) Ditto as above, except with 0000000002.
...
(10^10) Ditto as above, except with 1000000000.

I think this illustrates what Carl meant about a shorter program being contained within astronomically many longer programs.

However, there's a problem here. It may be that computing program (0) would allow one successfully to determine my pattern of observations, including my friend's delusion of hearing a voice and the specific sequence of neuron firings that caused him to pick the number 2835023981. But I don't have the computing power or time to test whether that's the case. For all I know, the laws of physics could predict that my friend would imagine he had to write the number -17.6 on his mirror instead. For practical purposes, the level of abstraction here is too fine-grained to be useful for ordinary humans. It's like trying to predict the stock market by modeling quark-level interactions in traders' brains.

So we move to a higher-level model, perhaps psychological. For example,

(0) The human brain is prone to certain kinds of imagined experiences. In order to explain all sorts of psychological phenomena throughout history, this hypothesis has a probability distribution over types of malfunctions that tend to produce weird sensations. Joe's experience corresponds to malfunction #611 combined with #28, plus a specific association with toothpaste and the number 2835023981.

Even this explanation assumes a more sophisticated model of psychology than we currently possess, but I think it gets at the idea of trying to explain the observation using fewer bits than just restating the entire account of what happened. In contrast, hypothesis (2835023981) still has to model most of human psychology, but it also includes the stipulation that "There is indeed an exception to ordinary laws of dental hygiene that will give Joe cavities unless he writes 2835023981 on his toothpaste, and moreover, this information will be communicated to Joe by a pattern of sound waves in his bathroom."

Of course, the other hypotheses seem even worse in description length. For instance, hypothesis (5928342301) has to model most of human psychology and then specify, "There is an exception to ordinary laws of dental hygiene that will give Joe cavities unless he writes 5928342301 on his toothpaste. Moreover, a pattern of sound waves in Joe's bathroom will give him a false message, telling him that the number is actually 2835023981." Here, we're encoding two ten-digit numbers instead of one, plus some extra linguistic information.

Carl, is this roughly the kind of reasoning that you had in mind? What should we do about the fact that, in practice, I don't have a good enough theory about the distribution of human mental abnormalities to say that Joe's experience corresponds to malfunctions #611 and #28? My actual description of his experience -- for instance, an email message I might send to you, written to be as short as possible but still understandable -- would require almost as many extra bits as hypothesis (5928342301) does, no?

Finally, here's a slightly tangential question that I'm also curious about. Basic Solomonoff induction, were it computable, would give us a prior distribution over finite or infinite binary strings. How would we transform our experiences of the world, like Joe's phone call, into binary strings in order to apply these prior probabilities? Or would we apply Solomonoff induction in a way that doesn't require predicting digits of binary strings?

Normal Beliefs: An Insanity Defense?

I am a primate running patchwork cognitive algorithms on relatively fragile wetware. We know that brain devices fail at relatively high rates. 19% of the US population has a mental illness of some sort, with a small fraction of these cases involving serious insanity or delusion. In addition, some people simply lack certain normal abilities, such as the 7% of males who are colorblind.

I and many of my associates have extraordinarily strange beliefs. Many of these are weird facts -- e.g., that an exact copy of me exists within a radius of 10^(10^29) meters. But others are logical conclusions (e.g., that libertarian free will is incoherent) and methodological notions (e.g., that Occam's razor makes the parallelism solution to the mind-body problem astronomically improbable). These latter kinds of beliefs theoretically involve certainty or near certainty.

But given my understanding of the frailty of human beliefs in general -- to say nothing of the tempting possibility that correct knowledge is out of the question, or that all of these statements are entirely meaningless -- should I assign nonzero probability to the possibility that I'm wrong about these conceptual matters?

One answer is to say "no": We all start with assumptions, and I'm making the assumptions that I'm making. This is my attitude toward things like Bayes' theorem and Occam's razor. In the same way that my impulse to prevent suffering is ultimately something that I want to do, "just because," so my faith in math and Bayesian epistemology is simply something the collection of atoms in my brain has chosen to have, and that's that. (I wonder: Is there any sense in which it would be possible to assign probability less than 1 to the Bayesian framework itself? Prima facie, this would be simply incoherent.)

But what about other, less foundational conclusions, like the incoherence of libertarian free will? It's not obvious to me that the negation of this conclusion would contradict my epistemological framework, since my position on the issue may stem from lack of imagination (I can't conceive of anything other than determinism or random behavior) rather than clear logical contradiction. On this point itself I'm uncertain -- maybe libertarian free will is logically impossible. But I'm not smart enough to be sure. And even if I felt sure, I very well might be mistaken, or even -- as suggested in the first paragraph -- completely insane.

Can probability be used to capture uncertainties of this type? In practice, the answer is clearly yes. I've done enough math homework problems to know that my probability of making an algebra mistake is not only nonzero but fairly high. And it's not incoherent to reason about errors of this type. For instance, if I do a utility calculation involving a complex algebraic formula, I may be uncertain as to whether I've made a sign error, in which case the answer would be negated. It's perfectly reasonable for me to assign, say, 90% probability to having done the computation correctly and 10% to having made the sign error and then multiply these by their corresponding utility-values-if-correct-computation. There's no mystery here: I'm just assigning probabilities over the conceptually unproblematic hypotheses "Brian got the right answer" vs. "Brian made a sign error."

In practice, of course, it's rarely useful to apply this sort of reasoning, because the number of wrong math answers is, needless to say, infinite. (Still, it might be useful to study the distribution of correct and incorrect answers that occur in practice. This reminds me of the suggestion by a friend that mathematicians might study the rates at which conjectures of certain types turn out to be true, in order to better estimate probabilities of theorems they can't actually prove. Indeed, statistical techniques have been used within the domain of automated theorem proving.) When someone objects to a rationalist's conclusion about such and such on the grounds that "Your cognitive algorithm might be flawed," the rationalist can usually reply, "Well, maybe, sure. But what am I going to do about it? Which element of the huge space of alternatives am I going to pick instead?"

Perhaps one answer to that question could be "Beliefs that fellow humans, running their own cognitive algorithms, have arrived at." After all, those people are primates trying to make sense of their environment just like you are, and it doesn't seem inconceivable that not only are you wrong but they're actually right. This would seem to suggest some degree of philosophical majoritarianism. Obviously we need to weight different people's beliefs according to the probability that their cognitive algorithms are sound, but we should keep in mind the fact that those weights are themselves circular.

How concerned should we be that, say, people who believe in parallelism of mind and body are actually correct?

Save Notes to Wikipedia

When I'm reading a good book or important article, I sometimes want to jot down notes about a key idea or save a link to the piece for future reference. Bookmarking sites like Delicious are a convenient way to do this for quick website references, but often I want to save a particular piece of information in a way that I'll naturally return to it later on, which may not be very likely for notes in a Delicious bookmark. Plus, if I find the information useful or important, I want to share it widely.

In such cases, I sometimes use Wikipedia as a "personal notebook" of sorts. When I found some interesting studies on crustacean pain, I added a description and footnotes to the article on lobsters. When I noticed that the Wikipedia article on Givewell lacked the most important information from a utilitarian standpoint -- what organizations Givewell has recommended -- I added a section summarizing Givewell's top charities, with some cost-effectiveness statistics. And when I read an interesting perspective on qualia by Gary Drescher, I added it to the page on that topic.

In addition to having social value, this reference-saving system is convenient for me, because the Wikipedia page on a topic is often the first place I go to look something up. The main downside is that writing a note in a format readable by others and integrated into an existing article takes a fair amount longer than jotting down some quick notes.

One aspect of contributing to Wikipedia that I found slightly difficult was learning to use footnotes. The article on citing sources was useful here, as were tools like this one from Google Scholar for automatically generating footnotes in proper bibliographic format.

Worms in the Rain

Update (2013): I no longer squish dying worms because I fear doing so may cause more harm than if they died on their own. I do, however, try to move them off the pavement onto the grass so that other pedestrians won't inadvertently half-crush them.


--------------

It rained last night, and this morning, I was not surprised to see a number of earthworms strewn across a stone walking path. While a few of the worms appeared healthy enough to return safely to the soil, most were clearly either dead or sufficiently incapacitated that they would die within a few hours.

The sight is truly painful. When I'm not in a hurry, I feel obligated to stop and completely squish those worms that appear to be suffering but not yet dead. I make sure to slide my foot along the pavement so that the guts of the worm are stretched out, ensuring a quick and certain death. This is, I think, what I would most want someone to do to me if I were going to be transformed into a worm in such a situation. (It's the opposite approach, I assume, from what a Buddhist or Jain would do, which seems to me unfortunate given their supposed concern for all creatures. Sometimes reverence for life ends up causing more suffering.)

The task of squishing worms feels overwhelming, because there is a practically unlimited supply of nearby places I could go where I could find dying worms on pavement. It would probably not be optimal for me to spend my life seeking out worms so that I could put them out of their misery. But this is not because the expected reduction in suffering due to stepping on worms is small: Indeed, if we give worms, say, a 1/3 chance of being able to feel pain, and if only 1/3 of the worms I step on are not yet dead or unconscious, then in the five minutes it takes me to step on 25 of them, I will have averted roughly three expected experiences of slow, lingering death. Rather, the reason it probably is not cost-effective to spend my life on this task is that the stakes in other domains are so much higher: The amounts of suffering in nature as a whole are vastly higher, "beyond all decent contemplation."

In theory, it would be optimal for me to bypass the worms on the sidewalk so that I could have an extra few minutes to do more abstract work that has far higher expected returns. The reason I don't do so, apart from feelings of direct pity, is somewhat selfish: Thinking about how much I ought to be getting done in the five minutes I save is stressful and overwhelming. It's hard to go through life with the mindset that every five minutes you waste on trivialities amounts to (far) more than three expected worms suffering while they die helplessly. And yet this is true. The best excuse I can give is that humans are not built to handle emotional burdens on this scale.

By the way, if readers have suggestions on the worms-in-the-rain situation, I would be glad to hear them. As far as I can tell, the worms come out to escape drowning, though they may also use the moisture as a chance to mate. In either case, though, it's clear that many of the worms on the pavement are in no position to return back to the soil, as is demonstrated by their shriveled-up remains the following day. It's not obvious that the rain or pavement themselves are to blame, because, as Charles Darwin suggested, it may be that many of these worms "were already sick, and that their deaths were merely hastened by the ground being flooded." If that is the case, then worms in the rain represent merely a glimpse of the vast amounts of sickness and death that occur all the time among animals in the wild.

Note: Professor Jeffrey Lockwood wrote a nice reply to this blog post, which appears as the third comment below.

Carter Center and PSI

Givewell has a new review of the Carter Center, focusing on its health programs, which comprise 80% of total spending. The summary chart includes a section "What does $100 do?" with some impressive figures, including one of the following depending on the program:

• "Averts 12-25 cases of guinea worm," or
• "Averts 10-50 years of serious debilitation (blindness, low vision, or irritating skin disease)," or
• "Averts 1-30 years of blindness and another 1-30 years of low vision (surgeries); little or unknown (other components)," or
  
• "Averts 15-85 total years of lymphedema (swollen limbs) and 25-165 total years of hydrocele (swollen scrotum)," or
• "Enables ~29 additional years of school attendance by treated children," or
• "Averts 10-47 malaria episodes (1 in ~320 is fatal)."

These are some excellent concrete scenarios to imagine when you're wondering, say, whether to spend $100 on a luxury or whether making an extra $100 is really that important.

Givewell's previous recommendation of Population Services International still stands. The 2007-2008 report concluded:

We estimate that it costs PSI $650-$1000 to prevent a case of HIV/AIDS and $500-$2500 to prevent a death from malaria; across the organization, we estimate that it costs PSI about $650-$1000 to save a life. These estimates do not include other benefits of PSI's activities, such as preventing unwanted pregnancies and reducing non-fatal malaria infections.

PSI is arguably a better choice than the Carter Center for international health, inasmuch as it devotes its entire budget to the task, rather than just 80%, but a specific examination of treatments would be in order. Either one seems like an excellent choice.

Horror movies

I really dislike horror movies, often feeling rather sick when I watch them; they're sort of like a miniature form of torture. (While I don't have mirror-touch synesthesia, I do probably have stronger mirror neuron firing than most other movie viewers?) But I think I generally feel more revulsion against pain afterwards than before. Just the few horror movies I have seen are probably somewhat responsible for my general disposition of wanting to prevent suffering.

It's interesting that a stereotypical horror movie watcher is perhaps the opposite of this: The usual image of such a person is a young person, probably male, and probably somewhat macho. It's also likely that watching horror movies repeatedly leads to desensitization. So a strategy of aiming to make people care more about suffering by promoting extensive consumption of horror movies would probably backfire.

Watching disgusting films is a unique experience that has made me a slightly different person from who I otherwise would be. This isn't surprising, perhaps, because apart from being a victim of violence or torture oneself, there are no other ways to encounter cruelty so graphically and forcefully. I wonder: What other human experiences are similarly powerful, to the extent that I should seek them out in order to broaden my perspective on the world?

About this blog

This blog contains some short ideas or notes that I want to record. However, my main site has generally more substantive content. I also post occasionally on the Felicifia forum.