Resolves YES if on January 1, 2100, experts agree that at least one insect existing on April 7, 2023, is sentient. Resolves NO if experts agree that no insects existing on April 7, 2023, are sentient. Resolves N/A if there's no expert consensus or there’s significant disagreement about the relevant experts and their conclusions. This could happen either because of normal uncertainty reasons or because the world looks very different in such a way that causes there to be no consensus (e.g. because everyone is dead).
If sentience turns out to be a confused concept, the market resolves in favor of whatever the expert consensus considers accurate in the spirit of the question. If the expert consensus is that there's really no reasonable answer because the question is too confused, the market resolves N/A.
For this market, sentience refers to the capacity to have subjective experiences with positive or negative valenced states, akin to pleasure or pain. If an insect has subjective experiences but no positive or negative valence attached to them, it is considered conscious but not sentient.
@JonathanRay A common view is that beings can be more or less sentient, but that some things still have no sentience at all. If insects fall into the latter category, this market resolves NO.
If you think everything is sentient to some degree or that the term "sentience" is confused in some other way, this part of the market description applies: "If sentience turns out to be a confused concept, the market resolves in favor of whatever the expert consensus considers accurate in the spirit of the question. If the expert consensus is that there's really no reasonable answer because the question is too confused, the market resolves N/A."
@WinstonOswaldDrummond Which experts will be used in determination of consensus? Am I an expert? What if the experts don't care about Manifold and don't want to spend the time fixing your wording, but they also don't conclude that "there's no reasonable answer" because they have precise answers for every narrowing of your wording? If you or your estate sends out a poll, how do you account for the experts that don't respond to the poll? What if it's well-settled what sentience is and whether it applies, but NO bettors decide to hire some paid actors to pretend to be experts to argue for an NA resolution? Do the insects have to be a species existing as of market creation, or can they be new species bred to be sentient? What's the definition of insect - noting that such definition needs to be robust in the presence of possible gene editing advances over the next 100 years? Are you going to use a separate panel of biologist experts to determine whether a specific insect counts as an insect and is eligible for judging of sentience? Is a panel of sentience experts going to judge a specific insect as sentient, or is argument against the class of organisms enough?
Also: I think anyone who makes a market expiring 2040 or later should be obligated to take out a whole life insurance policy so that when obligation to resolve their markets flows to their estate, it is guaranteed to have resources to fund paying all these experts to render a decision.
@Mira There are usually edge cases with questions involving emergent categories like insects and experts. But I think it's sufficient to define things precisely enough that the edge cases become very unlikely to make a difference to the market resolution.
> Which experts will be used in determination of consensus?
I'm imagining this resolving in a future where it's agreed upon who the experts are or any reasonable class of experts come to the same conclusion. If that's not the case, this clause applies and the market resolves N/A: "Resolves N/A if there's no expert consensus or there’s significant disagreement about the relevant experts and their conclusions." But groups of people I would personally consider experts might be biologists and philosophers of mind.
> What if the experts don't care about Manifold and don't want to spend the time fixing your wording,
People already get surveyed on insect sentience. I would just base the resolution on that. If it comes down to needing experts to comment on something specific to this market and they don't, it would resolve N/A based on this: "Resolves N/A if there's no expert consensus".
> What if it's well-settled what sentience is and whether it applies, but NO bettors decide to hire some paid actors to pretend to be experts to argue for an NA resolution?
I guess I think market manipulation is always a risk. We'd just have to try to make sure we're polling actual experts on their actual views. And since I'd likely base it off non-manifold-related surveys I don't think this should be an issue.
> Do the insects have to be a species existing as of market creation, or can they be new species bred to be sentient?
They have to be existing as of market creation. That's covered here: "Resolves YES if on January 1, 2100, experts agree that at least one insect existing on April 7, 2023, is sentient."
> What's the definition of insect - noting that such definition needs to be robust in the presence of possible gene editing advances over the next 100 years?
I think which currently existing biological species are classified as insects is pretty agreed upon. Future genetically modified species aren't considered for market resolution.
> Is a panel of sentience experts going to judge a specific insect as sentient, or is argument against the class of organisms enough?
Not sure I totally understand what you mean here. But the market resolves based on whether experts think "at least one" insect is sentient.
“Can a fruit fly’s brain have anything in common with a human’s? Out first instinct is to laugh at the idea – we are, after all, comparing our evolved human brain to that of a tiny insect. However, the truth is that no matter their size, animal brains function in similar ways, further evidence of which has recently been provided in a study supported by the EU-funded HBP SGA3 project.
Published in the journal ‘Nature’, the study shows us that fruit flies (Drosophila melanogaster) have more advanced cognitive abilities than scientists had previously thought. The researchers discovered that fruit flies are capable of attention, working memory and conscious awareness – abilities we usually only associate with mammals. By combining virtual-reality behaviour with neurogenetic manipulations and in vivo brain imaging, they observed the formation, distractibility and eventual fading of a memory trace in fruit flies’ brains.”
https://cordis.europa.eu/article/id/435869-fruit-flies-tiny-but-amazingly-smart
@NicoDelon And so I don’t expect a NO resolution (rather NA) but it’s way overpriced if the question is about consensus rather than the actual facts.
@NicoDelon But I think, for betting purposes, the question is more like "Conditional on consensus being reached by 2100, do you expect the consensus to be that any insects are sentient?" Because if you just think consensus won't be reached, there's no benefit to betting NO (since it would resolve N/A) and there's also no downside to betting YES, meaning that's not a reason to think the market is currently overpriced. (I.e., if you were to convince all the YES traders that consensus likely won't be reached, that wouldn't give them a reason to sell their YES shares.)
@WinstonOswaldDrummond Also happy to hear suggestions for better market resolutions in cases like this. I want to make prediction markets about big questions that we probably won’t have an answer to for a long time, but I’m not sure how to best make the markets rigorously defined while still making sure people are betting on the thing I actually care about.
Maybe, instead of specifying a year like 2100, I could’ve just said I’ll resolve it whenever consensus is reached? But you could still think that consensus will never be reached, consensus will be wrong, or picking out who the relevant experts are and what the consensus is would be too subjective.
Another option is I could make the market like a stock that never resolves.
@Gigacasting Most people would say they don't experience anything. (They can be given positive and negative rewards/feedback, but that doesn't necessarily entail conscious experiences, let alone valenced experiences.)