I created a pamphlet to promote concern for wild animals. I might try distributing it Vegan-Outreach-style at my local mall, pehraps along with Peter Singer's excellent "Do Animals Feel Pain?" excerpt.

The pamphlet is still a draft, so let me know if you have comments. What questions have I not addressed? Do I say anything that's confusing or needlessly offputting? (I deliberately omitted my favorite Dawkins quote about the "blind, pitiless indifference" of nature in case its inclusion would offend religiously minded folks who might still be receptive to the message. I assumed it was safer to include Stephen Jay Gould's "Nonmoral Nature" as a reference, but perhaps even that could be seen as offensive by some?) By the way, the paper-printout version of the pamphlet doesn't quite fold correctly -- I'll need to adjust the column widths and positions a bit.

Hi Alan,

I think that this is a good start for a pamphlet. I could imagine it having some meaningful effect, especially with polishing.

The formatting could probably use some improvement. I will send it to my father to see what he thinks. I presume that the references and contact information will appear at the end of the pamphlet when it's folded up? The page that starts "...and while there may be little humans can do now to address the problem, we should remember that it matters." is awkward - there's too much hyphenation.

I would make the references to insects less salient. It may be too big of a leap for your audience. Indeed, I remember your sister suggesting that you're crazy, me disputing her claim, and her justifying her assertion by pointing to your concern for insect suffering. I think that your claim that the problem matters because there are a lot of animals is strong enough without the reference to insects. On the other hand, I think would be okay to list "do insects feel pain?" as an open question.

Perhaps instead of "When assessing that question, we should bear in mind that most species give birth to hundreds or thousands of offspring at a time, almost all of which die shortly afterwards. Moreover, many adult insects live just a few weeks before dying of dehydration, disease, or entanglement in a spider’s web." (the last line of which doesn't really add anything) you could say "When assessing that question, we should bear in mind that most species give birth to a large number of offspring most of which die soon thereafter. This is especially true of insects, and the question of whether insects feel pain is an open one."

The sentence "Rather than policing nature, we should focus on further research into the welfare of animals in the wild — do they suffer more than they’re happy, for instance — and on developing technologies that, over the long run, might some day allow our descendants to aid wild animals without causing disaster." seems awkward to me. Maybe I would omit the clause "do they suffer more than they're happy" entirely - you already address this point when you say "humans should think carefully about whether, on balance, wilderness contains more suffering or happiness.

Thanks for the feedback, Jason! I've replied offline to talk about detailed and uninteresting points, but I hope to update the online version of the pamphlet once our discussion is done.

I'm ambivalent about relaxing the insect reference, because part of me says, "If people aren't caring about wild insects (and other small, highly numerous animals), then I don't really care that much about the issue in general." I don't want readers to come away from the pamphlet thinking that I'm advocating the sorts of "wild-animal welfare" projects that people conventionally think of, like flipper bands for penguins or "extracting bile from gallbladder of bred bears." It's not that those projects aren't good, but they seem no more cost-effective -- probably far less so -- than conventional efforts to improve the welfare of farm animals. The whole point of raising the wild-animal issue is to ensure that people care about the vast numbers of small animals with which humans have essentially no direct interaction.

As I said in my email to you, I think that when trying to solve accomplish a large scale task it's important to be wary of artificial self-imposed constraints. The best thing to do is to state the goal clearly with a minimum amount of baggage, consider the set of all possible ways to achieve it, and pick the most promising out of them. In particular, you should consider the question of whether it's genuinely necessary to get people concerned about wild insects right now for you to maximize the chance of people behaving ethically toward sentient beings in the long run.

You might successfully distinguish your point of view from that of advocating typical "wild-animal welfare" projects by emphasizing that our present capacity to engage in wild-animal welfare projects is many orders of magnitude too small to make a dent in the larger situation and that what's really needed is research and technological development with awareness of the larger situation in mind.

Jason, do you think this passage adequately distinguishes my position from that of ordinary "wild-animal welfare" projects?

There may be a few instances, especially where humans already interfere with ecosystems, in which intervention could reduce suffering without much cost. But the vast majority of wilderness suffering is on the part of small animals, most in the oceans, that are hard to reach and help, even if we knew what “helping” them would look like.

Also, in your email, you wrote:

But I feel like in the present sociological climate, concern for insect welfare is destined to be confined to a very small group of people. When you say "lots of people take the welfare of insects seriously" you mean "lots" in the same sense that a mathematician might say "lots of people are in awe of Deligne's proof of the Weil conjectures," right? Which is to say, something like 0.01% of the population of the developed world.

I think you'd be surprised just how much ordinary people take this issue seriously. Indeed, I almost suspect that rates of concern for insects are lower among academically inclined animal supporters than among everyday animal lovers. Try searching "Do you think insects feel pain?" (with or without quotes) and see how many forum and survey sites come up in which ordinary people -- many non-vegans -- debate this issue.

Hi Alan. Great initiative! A few comments and suggestions:

1. Perhaps a couple of lines on the moral significance of suffering at the beginning of the first section could increase the effectiveness of your appeal. As it stands, the pamphlet simply makes a factual claim, without being explicit about its ethical relevance. A possible revision might be:

Suffering is horrible--it should not exist. Yet most of the suffering in the world is experienced by wild animals. There are 6.7...

2. The reference to your essay on 'How many wild animals there are?' should, I think, be moved to the 'further reading' section, in order to avoid cluttering the main text unnecessarily. Alternatively, you may use a Chicago-style in-text citation.

3. The sentence that begins

But the vast majority of wilderness suffering is on the part of small animals

should, I think, be changed to

But the vast majority of suffering in the wild is experienced by small animals

4. The sentence

Rather than policing nature, we should focus on further research into the welfare of animals in the wild — do they suffer more than they’re happy, for instance — and on developing technologies that, over the long run, might some day allow our descendants to aid wild animals without causing disaster.

may perhaps be broken down into smaller sentences, and enriched with more examples. A possible version:

Rather than policing nature, we should focus on further research into the welfare of animals in the wild. This research would address questions such as 'Where does sentience begin in the animal kingdom?' 'Do animals experience more suffering than happiness overall?' 'Is brain size significantly correlated with intensity of experience?' In addition, we should focus on developing technologies that, over the long run, might some day allow our descendants to aid wild animals without causing disaster. [Perhaps add one or two examples.]

Much appreciated, Pablo! I took 1, 3, and a modified version of 4. In particular, on 4, I rewrote the paragraph as follows:

Rather than policing nature, we should focus on research into the welfare of animals in the wild — studying such questions as “Where does sentience begin in the animal kingdom?” and “Do animals experience more suffering than happiness overall?” Doing so would inform policies — like habitat preservation or, more speculatively, spreading life into space — which affect the large-scale numbers and types of wild animals that exist. In addition, we ought to encourage the development of technologies that may one day allow our far-future descendants to relieve the suffering of animals in the wild in ways that don’t cause ecological catastrophe.

I'll update the PDF on my site later on....

"The death of one man is a tragedy, the death of millions is a statistic".
Alas Stalin alas right; and I fear the same is true for non-human animals too.
However, I wonder if Alan's excellent pamphlet may be too pessimistic about what we can do this century?

IMO we should aim systematically to "police" Nature. Suffering in the living world can be first reduced and then abolished. The technologies to do so are either available now - notably depot-contraception (immunocontraceptives, etc) to regulate fertility, remore-controlled neuroimplants to tweak violently aggressive behaviour, GPS tracking devices to monitor health-status etc - or are a recognizable extension of existing technologies (e.g. habit-forming catnip-laced lab-cultured mincemeat distributed to obligate carnivores in our wildlife parks, etc). No exotic physics or speculative metaphysics (e.g. a technological Singularity) need be invoked to deliver humane ecosystems in our wildlife parks. The worst examples of non-human suffering are typically also the most easily accessible to biomedical intervention because they involve large, slow-breeding vertebrates (e.g. whales, elephants) rather than small insects or marine molluscs. This generalisation is not intended to trivialize the plight of lesser beings. Rather, it's a plea for hard-headed utilitarian triage ( http://en.wikipedia.org/wiki/Triage ) Thus managing the lowest trophic levels of the existing food-chain won't be technically feasible for several decades; large vertebrates could be helped now. Comprehensive marine ecosystem management based on compassionate principles will presumably require Drexlerian nanobots - although I think the ethical / ideological challenges to building a cruelty-free world dwarf the implementation details.

Here is a concrete example of "policing" Nature - or rather, building a compassionate cross-species welfare state. I intend to write this up as a detailed (and costed) feasibility study - rather than the broad-brush treatment of http://www.abolitionist.com/reprogramming/index.html

ELEPHANTS
African population c 500,000. Annual cost of a comprehensive welfare package c. $2 billion [?]
The most common cause of wild elephant morbidity and mortality is malnutrition / starvation. Elephants typically go through six sets of molars in a lifetime. As his or her sixth molars wear down, an elephant slowly starves to death. Elderly elephants eventually collapse from hunger and exhaustion. They are then commonly eaten alive by lions or scavengers. The suffering involved is horrible beyonds words. However, there is no technical reason why mature elephants can't be fitted with dentures that can last indefinitely, or at least until their replacement decades hence. [This dental procedure was recently performed on a Thai zoo elephant http://news.bbc.co.uk/cbbcnews/hi/anima ... 375305.stm ] No wild or captive elephant should ever have to die of hunger or thirst. Consequently, elephant fertility needs to be managed via depot contraceptive implants in relation to the ecological resources of our wildlife reserves. Clearly, any current estimates of the financial cost of providing elephant orthodontics, obstetrics, immunocontraception and so forth must use back-of-an-envelope speculation. But the figures aren't completely fanciful - and costs would fall.

Case studies and then policy initiatives as on elephants above could in time be repeated all the way "down" the phylogenetic tree - though the complex ecological modelling needed should be done by ecologists and population biologists, not philosophers.

Thanks, Dave! We do disagree on this matter a bit.

For one thing, I'm not entirely sure I agree with weighting our concern strictly in proportion to brain size. Tiny insects with just a few brain cells display some amazing behaviors that I think may need to count as more than their weight in neurons. That said, if we do weight by brain size alone, your elephant population of 500,000 is equivalent to perhaps 2.5 million humans, assuming a ratio of 5:1.

Of course, there may be instrumental reasons to give much more weight to humans: For instance, bringing people out of destitution and giving them an education may make them more likely to have the time and resources to care about concerns like wild-animal suffering. I should add that not all educated humans do care about wild-animal suffering, unfortunately: Take Ned Hettinger and some other ecological philosophers as examples.

I'm inclined to agree with Alan here: the aggregate suffering undergone by invertebrates very likely dwarfs the pain that "higher" beings collectively experience, even granting the assumption that intensity of experience varies with brain size.

It is interesting to note that the superior cost-effectiveness of the most promising interventions from a utilitarian perspective -- such as reducing the risk of human extinction and reducing the suffering of wild animals-- derives from the sheer number of the beings involved, as opposed to the intensity of the suffering experienced.

I think it needs to give examples of what we might eventually be able to do to reduce suffering, such as genetically modifying animals to replace happiness and sadness with degrees of happiness.

Unfortunately, I don't think there are any ways that don't appear extremely radical.

Also, I think the picture of the dead squirrel would work better if it wasn't lying on cement.

It annoys me that it says "billion billion" like I don't know what a quintillion is, but I guess most people don't. I have a habit of assuming people know a lot of things I do.

"Respecting nature means accepting the cruelty it contains." I think that's correct. I just don't see any reason to respect nature. I consider the artificial world we live in proof that nobody does.

Perhaps you should comment on the double standard that people think animals should live in nature, yet do all they can to avoid it themselves when it suits them. For instance, vaccinations and antibiotics to get rid of natural diseases. That could be a bit problematic, though, as people tend to think living in nature is an ideal, even if they avoid it themselves.

Very good initiative!

Some comments:

Alan:
1A. I Agree, prima facie, with Jason's point. It seems more promising to first try to convince people to care about wild animals at all, then extend the circle of care to smaller and smaller animals through several steps.

1B. To support the point: I've met several people who thought the idea of spending resources on insects was a reductio of utilitarianism, or even madness.

2. You have an objection against this point that academics may care less about insects compared to ordinary people.

It is an interesting data-point that many ordinary people ask whether insects feel pain. However, there is a big difference between asking if something feels pain and caring about it enough to spend resources. In fact, the point that many people wonder if insects feel pain may speak against including insects in the pamphlet since hardly anyone doubts that mammals and other bigger animals feel pain.

3. I'll try handing out the pamphlet my department for more (academic) feedback, but perhaps it's better to wait for the next version.


Pablo:
4. Even if the suffering of elephants doesn't matter that much, it could be a good start to get people on the slippery slope towards caring about insects (or at least, the hoards of smaller mammals, birds and reptiles).

DanieLC:

Perhaps you should comment on the double standard that people think animals should live in nature, yet do all they can to avoid it themselves when it suits them.

The perhaps most interesting double standard here is that almost everyone (including animal rights/welfare people) would be prepared to (and think it morally necessary to) defend an unknown human child against predators, but not that we should protect any other animals from harm in the same way.

Jason, do you think this passage adequately distinguishes my position from that of ordinary "wild-animal welfare" projects?

Quite possibly, but I'm not sure, this is an empirical question.

No distribution of discrete pinpricks - whether millions, billions or googols of them - will ever amount to a state of agony. To believe otherwise is to fall victim to the fallacy of composition. So I think we should prioritize getting rid of the worst forms of suffering - even though we can eventually abolish pinpricks. Our concern shouldn't be weighted strictly in proportion to brain size, but rather to intensity of experience as determined by neural structure. For example, single nucleotide polymorphisms (SNPS) of a single gene SCN9A determine whether one never has any sensation of phenomenal pain at all, weak pain sensation, normal pain sensation, or extreme forms of agony.
Thus see e.g.:
http://www.opioids.com/pain/scn9a.pdf
Getting rid of the nastier variants of the SCN9A gene in humans, and systematically working our way though other species' genomes, is a perfectly feasible engineering challenge. IMO it's the ethical/ideological obstacles that are truly daunting: I don't know if they can be overcome.

No distribution of discrete pinpricks - whether millions, billions or googols of them - will ever amount to a state of agony. To believe otherwise is to fall victim to the fallacy of composition.

If what you mean by the first sentence is that the phenomenology of experiencing any given number of pinpricks in succession differs from the phenomenology of having an experience of agony, then I quite agree. The relevant question, however, is not about phenomenology, but about morality: is there some number of pinpricks such that their value equals the value of an experience of agony? I don't fall victim to any "fallacy of composition" by virtue of answering this question in the affirmative, any more than you do by claiming that two equally intense experiences of agony are worse than one slightly more intense experience. Indeed, your view seems to commit you to the answer I myself give to that question. By constructing a series of experiences such that each is only slightly less intense than the preceding one and making successive pairwise comparisons between neighbouring experiences in the series, it's easy to show that you are committed to the very claim that you deny in the passage quoted above. Do you think you can avoid this implication?

(Edited for clarity.)

No distribution of discrete pinpricks - whether millions, billions or googols of them - will ever amount to a state of agony.

That's not totally impossible, but it leads to some paradoxes. For example: suppose you give someone a million pinpricks and one happy event half as good as necessary to counter it out. His life isn't worth living. Now make a large number of people that have lives exactly like that. Now the total amount of pain is about infinity pinpricks, and the total amount of pleasure is about infinity of whatever that larger unit is. Life is now worth living.

In other words, it's possible to have a world where each individual's life isn't worth living, but together it is (or vice versa). It's similarly possible to have each day not worth living, but your total life worth living.

If you're wondering how this concept of utility is possible, here it goes:
A combination of events is represented by function, f(x). One event is the function f(x)=sqrt(a)/k*(H(x)-sqrt(abs(a))), where a is the utility of that event, and k is how many times the utility you'd get if it happened infinity times. Adding two functions together results in the function representing the combination of functions.
The utility, U(f(x))=2*k*integral(0,infinity)x*(1-e^(-f(x)))dx

To focus our discussion, it may be helpful to consider the following principle:

For any unpleasant experience of a given duration and intensity, there is an unpleasant experience of the same duration and slightly less intensity such that sufficiently many instances of the latter, experienced either simultaneously by different creatures or sequentially by the same creature, are on the whole worse than a single instance of the former.

Dave, do you accept this principle? If not, why not?

That's not a necessary consequence of what he said. It only has to be possible for some decrease in intensity. He never said they can't add in some form.

Although I'm not sure what he meant by the fallacy of composition. I don't think it means what he thinks it means.

Thanks to all for the continued comments. I've updated my site to include a new PDF (see the "pamphlet" hyperlink). Also, because some of you might want to modify the pamphlet for your own purposes (e.g., to make revisions that you find preferable), I've included links to the source files -- a .pub file for those who have Microsoft Publisher, as well as the plain text and pictures for those who don't.

I'll reply to other comments in a bit.

DanielLC:

You make some good points on the pamphlet. I've opted to err on the side of keeping the pamphlet too simple rather than too verbose, allowing readers to see my essays for further details. (Same goes for the evidence on insect suffering, etc.) However, do feel welcome to modify the text as you see fit! And yes, I deliberately avoided "quintillion" (or -- what's easiest to read -- 10^18) for the sake of lay readers.

I agree the squirrel picture (and perhaps the others) aren't necessarily optimal. I chose them because they were the best public-domain photos I could find without expending too much effort. If you identify better ones, please let me know.

I've tried to comment on the double-standard by the comment about "the 'natural' suffering of humans due to malaria, cancer, or starvation," though you're right that I haven't explicitly mentioned to readers their own vaccines and air-conditioned houses. I'm not sure if such points would sound unnecessarily accusatory; at least (almost) everyone seems to agree that malaria is morally bad.

Jesper:

I know a number of people who try to avoid harming insects in their direct physical interactions with them no less than they aim to avoid harming, say, a cat. (And no, they're not Jains -- just ordinary people who care about "all creatures" and such.) Of course, they don't extend this logic terribly far: For instance, they probably don't try to drive less in order to avoid killing insects in their windshield:

Nearly 40,000 volunteers affixed "splatometers" -- cardboard counting grids -- to their front license plates, stopping every 20 to 80 miles to count the number of smashed insects. A total of 324,814 bugs died for science, an average of one every five miles.

(See also this page. And no, I don't expect taking up this issue is cost-effective. For one thing, I expect splattering on a windshield is a relatively quick death by comparison with most ways in which insects perish. Secondly, I'm sure humans affect insect populations vastly more in other ways.)

Do let us know what feedback you get from your department! Of course, feel welcome to modify the pamphlet before distributing it.

Earth's most successful animal species in terms of biomass are probably copepods and Antarctic krill:
http://en.wikipedia.org/wiki/Copepod
http://en.wikipedia.org/wiki/Antarctic_krill
At the risk of trivializing their experiences, I'm going to call their worst aversive sensations "pinpricks", though this needs to be shown, not merely asserted as obvious. If there is a common currency for units of pain - let's call them "dolors" - then the aggregate of dolors undergone by copepods and Antarctic krill may exceed the aggregate of dolors suffered by vertebrates by several orders of magnitude. So does proportionately greater moral urgency lie in reducing aggregate pinpricks of copepods and Antarctic krill than reducing the suffering of large-brained vertebrates? Clearly, in one sense the question is academic. We lack the computational and nanorobotic technologies to micro-manage marine ecosystems at this trophic level. But later this century and beyond, it's conceivable that the technologies will exist. As utilitarians we should presumably promote their development, at least if we want to sustain Nature - an open question IMO.

A counterargument is that "dolors" aren't identical and interchangeable in this way; moral calculus can't be done in such a simplistic fashion. Thus phenomenal agony-so-bad-one-wants-to-die isn't just quantitatively more painful than a pinprick. Such agony is qualitatively different, too, and consequently it licenses a different moral response than anything based on aggregate "dolor counting". A metaphor that captures the distinction I've in mind is some kind of phase change. H2O at 99 degrees C is essentially different from H2O at 101 degrees C. The two phases of water share a common numerical scale of measurement (i.e. temperature) but a qualitatively different phenomenon occurs as temperature rises. Merely adding more units of water at 99 degrees C doesn't turn it into water at 101 degrees C - and because the phase change is a qualitatively discontinuous phenomenon, a principled difference in our response may be appropriate too. However, I'm not sure how to turn this loose analogy into a rigorous argument. So sorry Pablo, I haven't adequately answered your question. And intuitively (for what that's worth), we do in some contexts want to trade off large aggregate numbers of pains / pleasures against fewer but slightly sharper intensities.

What do other people think here? And practically, what species members should utilitarians prioritize now?
IMO the phenomenally most intense forms of suffering are typically the most accessible i.e. they occur in large-brained vertebrates. These are the nastiest kinds of suffering I think we should tackle first.

Again, excellent pamphlet Alan. We should make more!

A counterargument is that "dolors" aren't identical and interchangeable in this way; moral calculus can't be done in such a simplistic fashion. Thus phenomenal agony-so-bad-one-wants-to-die isn't just quantitatively more painful than a pinprick. Such agony is qualitatively different, too, and consequently it licenses a different moral response than anything based on aggregate "dolor counting". A metaphor that captures the distinction I've in mind is some kind of phase change. H2O at 99 degrees C is essentially different from H2O at 101 degrees C. The two phases of water share a common numerical scale of measurement (i.e. temperature) but a qualitatively different phenomenon occurs as temperature rises. Merely adding more units of water at 99 degrees C doesn't turn it into water at 101 degrees C - and because the phase change is a qualitatively discontinuous phenomenon, a principled difference in our response may be appropriate too.

Although your analogy is to some extent illuminating, I don't think it can be used to support a plausible moral position. As far as I can see, the analogy is supposed to work as follows. Physical temperature corresponds to phenomenal unpleasantness; the physical properties of gases correspond to the phenomenal properties of pinpricks; and the physical properties of liquids correspond to the phenomenal properties of agonies. The claim is then that, just as a phase transition occurs when temperature drops below a certain critical point, so a "moral phase transition" occurs when unpleasantness attains a certain critical degree of intensity. I would like to make a few comments about this analogy.

1. The analogy assumes that there is a common phenomenal denominator underlying both instances of agony and instances of pinprick: the common denominator is the property of phenomenal unpleasantness. This property corresponds, as I noted, to the property of temperature, which both gases and liquids possess, albeit to different degrees. This seems to be in tension with some of the things you say in this and earlier posts, where it is suggested that these different experiences lack any such common phenomenal property.

2. Insofar as your position attaches normative significance to moral phase transitions, it ceases to be hedonistic, for it no longer regards unpleasantness as the sole determinant of value. It is the new, different phenomenal properties that an experience acquires at the point of "moral condensation" that explain the moral difference between agonies and pinpricks. The position that one can reconstruct from this analogy is instead a form of value pluralism, which either regards unplesantness as one among many normatively relevant phenomenal properties, or abandons the view that unpleasantness is intrinsically bad altogether, embracing instead the view that these other phenomenal properties are the sole determinants of value.

3. Regardless of the form it takes, a position of this sort seems to me fundamentally objectionable on the grounds that it fails to do justice to the phenomenology of moral experience. Phase transitions are easily observable in the external world, but are nowhere to be found in our inner mental life. There simply are no phenomenal experiences with a degree of unpleasantness corresponding to a temperature of 99 (or 101) degrees C [for water at normal conditions of pressure]. Slightly increasing the intensity of an unpleasant experience, however mild or intense this experience may be before the increase, does not produce an experience with radically new phenomenal properties.

4. A position of this sort, moreover, would have extremely implausible implications. Let us grant that such radically new phenomenal properties emerged at certain critical points. How would someone who attached normative significance to these properties express the change in value involved in this phenomenal discontinuity? As far as I can see, there is only one way: by holding that the instantiation of the new emergent properties has infinite value. Otherwise, there would be a finite number of experiences of pinprick whose value would exceed that of an experience of agony. But if you do attach infinite value to experiences that are past the critical point, then you are committed to the absurd view that no number of unpleasant experiences above this critical point will ever be as bad as one single unpleasant experience below the critical point, however minuscule the difference in intensity between these two experiences might be.

Phase transitions are easily observable in the external world, but are nowhere to be found in our inner mental life. [...] Slightly increasing the intensity of an unpleasant experience, however mild or intense this experience may be before the increase, does not produce an experience with radically new phenomenal properties.

I don't think that this is clear. I personally have a subjective sense qualitative changes in my experience that have occurred at various times.

I think I would choose 100 years of extremely high quality of life over millions of years of mildly pleasant life. This preference may be (a) borne of scope insensitivity, (b) it may be that what I conceptualize as "mildly pleasant life" is actually not worth living, or (c) it may be that the positive experiences that humans have differ in intensity by many orders of magnitude.

I bet you'll jump on (a) as the probable explanation for my choice. But at present there's no objective criterion to determine whether or not this is the case. You can say "You're being insensitive to scope!" and I can say "No, no, you don't understand, the two positive experiences really are that different!" and we have no way of reconciling the situation.

Explanation (b) is possible, but again there's a difficulty of making a meaningful assessment of whether it's so.

That leaves the possibility of (c), which seems worthy of consideration.

Now, given that I'm willing to forgo millions of years of mildly pleasant life, I feel like I'd be willing to forgo billions or trillions of years of mildly pleasant life - it just doesn't feel that different, and if . This probably seems like an indicator of scope insensitivity and it may be. But if it is then that means that there's some number X such that 100 years of extremely high quality life is worth exactly X years of mildly pleasant life. This means that if I should choose 100 years of extremely high quality life in over X - 0.01 years of mildly pleasant life but not over X years of mildly pleasant life.

This seems counterintuitive to me. I guess where my intuition objects is that right now I seem to be be living a mildly pleasant life and my preference is as stated above. This means that at every moment of an eternal mildly pleasant life, I would be willing to trade it all for 100 years of extreme happiness. Similarly, I think that in my most blissful moments I would choose lengthening them over an eternal mundane existence.

Similarly, on the negative side, if I was experiencing an eternal life with some mild pain (e.g. a slightly achy back) that was not accompanied by any positive experiences and was given the option of being tortured for 100 years in exchange for the right to die, I think I would prefer to keep living with my mild pain rather than be tortured. If I were in the middle of being tortured for 100 years, I would certainly choose to end it in exchange for an eternity of mild back pain.

So I'm led to suspect that (regardless of how one chooses the numbers), assigning each affective state a real number and making choices based on maximizing the integral of the affective state over time could in principle lead to shifting an organism from outcome A to outcome B when the organism prefers outcome A to outcome B whether he/she/it is in a state in outcome A or a state in outcome B and this seems to be a reductio ad absurdum. (Note that dead people don't have preferences - it seems like all conscious versions of me involved in either outcome A or outcome B prefer outcome A.)

Maybe with more reflection or greater cognitive resources I would come to a different assessment of the situation and agree that we can associate a real number to each affective state and have everything work out well. But the idea that it's obvious that this is so seems to me glib at best.

As far as I can see, there is only one way: by holding that the instantiation of the new emergent properties has infinite value

Quite possibly the extended real numbers are not a good number system to use for ordering the moral significance of outcomes. Maybe Z x R with the lexicographic ordering would be more suitable. I think that sometimes the fitting of intuition to mathematical models is misguided and that one should instead create a mathematical model that fits one's intuition better.

But if you do attach infinite value to experiences that are past the critical point, then you are committed to the absurd view that no number of unpleasant experiences above this critical point will ever be as bad as one single unpleasant experience below the critical point, however minuscule the difference in intensity between these two experiences might be.

It sounds like here you're falling back on assuming that there are not state changes rather than granting that there can be state changes.

Maybe you're thinking that there seems to be something fishy going on because the idea that the addition of a single neurotransmitter could induce a qualitatively new affective state seems weird. But this seems possible to me: chemical reactions have an activation energy, etc.

This seems counterintuitive to me. I guess where my intuition objects is that right now I seem to be be living a mildly pleasant life and my preference is as stated above. This means that at every moment of an eternal mildly pleasant life, I would be willing to trade it all for 100 years of extreme happiness. Similarly, I think that in my most blissful moments I would choose lengthening them over an eternal mundane existence.

I do not dispute these claims. What I dispute is that the preferences or value judgements of these people can fit into a consistent preference function or theory of value. The same person who is willing to exchange a drab eternity for a century of ecstasy is willing to exchange a century of ecstasy for sufficiently many slightly less ecstatic centuries, these many slightly less ecstatic centuries for sufficiently many even less ecstatic centuries, and so on, in a way that commits her to valuing a very long but drab existence over a century of ecstasy. (Indeed, the drab existence need not be eternal, so the conclusion is even stronger.)

The objection I'm raising could be restated as follows. Consider these three claims:

1. For any unpleasant experience of a given duration and intensity, there is an unpleasant experience of the same duration and slightly less intensity such that sufficiently many instances of the latter are on the whole worse for the person experiencing them than a single instance of the former.

2. There are some unpleasant experiences so intense that no number of mild unpleasant experiences experienced in succession can ever be as bad for the person experiencing them as experiencing just one of those intense experiences.

3. Claims (1) and (2) are mutually exclusive.

The objection is that, although most people are unwilling to reject either (1) or (2),they are yet unable to show that (3) is false.

You need one more premise to get a contradiction:

If a given set of experiences is good, having that set of experiences multiple times is also good.

The example I gave earlier woks for premises 1 and 2, but fails for the one I just mentioned.

2. Insofar as your position attaches normative significance to moral phase transitions, it ceases to be hedonistic

Wouldn't a view like the one Jason suggested, postulating a utility function which maximizes several dimensions of real numbers lexicographically ordered still be hedonistic? (I hope I got that right, Jason)
This seems similar to what Dave is suggesting: each state change would correspond to a whole new dimension, which may be only taken into account after the previous one is maximized.

Phase transitions are easily observable in the external world, but are nowhere to be found in our inner mental life. [...] Slightly increasing the intensity of an unpleasant experience, however mild or intense this experience may be before the increase, does not produce an experience with radically new phenomenal properties.

I don't know if it is what you mean by "radically new phenomenal properties", but there seems to be, at least, a point where pain is such that the only thing that is important is the removal of pain, even before the preservation of life itself. This appears to be a nice candidate for a phase transition.

I agree with your point about the three claims; this position seems to amount to a rejection of the first one: *

1. For any unpleasant experience of a given duration and intensity, there is an unpleasant experience of the same duration and slightly less intensity such that sufficiently many instances of the latter are on the whole worse for the person experiencing them than a single instance of the former.

Is this position untenable?

*edited to remove wrong and pointless remark which was here between brackets.