Schwitzgebel's "no relevant difference" like NTT (but valid?)

Vegan message board for support on vegan related issues and questions.
Topics include philosophy, activism, effective altruism, plant-based nutrition, and diet advice/discussion whether high carb, low carb (eco atkins/vegan keto) or anything in between.
Meat eater vs. Vegan debate welcome, but please keep it within debate topics.
User avatar
ModVegan
Full Member
Posts: 123
Joined: Sun Dec 18, 2016 12:01 pm
Diet: Vegan
Location: Calgary, Alberta
Contact:

Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by ModVegan »

Firstly, I apologize if this has already been covered in other posts. I've been trying to keep up with the discussion here, but I'll admit I haven't really thought through everything.

I was listening to an interview with Eric Schwizgebel this morning (a philosopher interested in the rights of AI), and he proposes a moral test fairly similar to #namethetrait, but which makes a good deal more sense, in my opinion.

Schwitzgebel argues from a standpoint he calls "no relevant difference" - in other words, in order to justify treating two entities morally differently, there has to be a relevant difference between those two entities.

It's more ambiguous than name the trait, but I also think it passes the rock test, and it avoids hangups like Isaac's P1 (humans have moral value). If you're familiar with Nick Bostrom's work, he also makes a similar argument re: sims - (if we imagine we're sims, and if we know we have rights whether we're sims or not, then it follows logically that being a sim does not, ipso facto, represent sufficient grounds for not having rights). You can see Schwitzgebel's paper here: http://www.faculty.ucr.edu/~eschwitz/SchwitzPapers/AIRights-150915.pdf

This is how Schwitzgebel constructs his argument:
Premise 1. If Entity A deserves some particular degree of moral consideration and Entity B does not deserve that same degree of moral consideration, there must be some relevant difference between the two entities that grounds this difference in moral status.
Premise 2. There are possible AIs (In our case, animals) who do not differ in any such relevant respects from human beings.
Conclusion. Therefore, there are possible AIs (Animals) who deserve a degree of moral consideration similar to that of human beings.
inator
Full Member
Posts: 222
Joined: Sat Apr 04, 2015 3:50 pm
Diet: Vegan

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by inator »

ModVegan wrote: Thu Dec 07, 2017 11:06 am This is how Schwitzgebel constructs his argument:
Premise 1. If Entity A deserves some particular degree of moral consideration and Entity B does not deserve that same degree of moral consideration, there must be some relevant difference between the two entities that grounds this difference in moral status.
Premise 2. There are possible AIs (In our case, animals) who do not differ in any such relevant respects from human beings.
Conclusion. Therefore, there are possible AIs (Animals) who deserve a degree of moral consideration similar to that of human beings.
That's a valid argument.
Subjectivists may reject the truth value of Premise 1, though, so I think he'd also have to justify why there "must be some relevant difference" (as opposed to the difference simply being arbitrary).
They could also reject Premise 2 by arguing that there is in fact a relevant difference between the two (relevant to subjectivists), namely that group A is "us" and group B is "not us". It depends on how he defines "relevant difference", and he'll have to justify it.
I haven't read the article yet, maybe he did that already. But his argument is definitely an improvement simply by being valid.
brimstoneSalad wrote: P5 "Any attempted defense of personal interests is an indication of self-value."

A problem here may be that some actions may appear to be an attempted defense of personal interests and actually just be an automatic response. Like plants or potentially some insects.
This doesn't contradict P5, because you'll argue that those instances are not 'true' attempts to defend personal interests given no existing interests. But to "plants have feelings" people it may seem like you're making ad-hoc justifications. A footnote is probably not necessary in this case, but I'm just putting it out there.
User avatar
brimstoneSalad
neither stone nor salad
Posts: 10273
Joined: Wed May 28, 2014 9:20 am
Diet: Vegan

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by brimstoneSalad »

ModVegan wrote: Thu Dec 07, 2017 11:06 am This is how Schwitzgebel constructs his argument:
Premise 1. If Entity A deserves some particular degree of moral consideration and Entity B does not deserve that same degree of moral consideration, there must be some relevant difference between the two entities that grounds this difference in moral status.
Premise 2. There are possible AIs (In our case, animals) who do not differ in any such relevant respects from human beings.
Conclusion. Therefore, there are possible AIs (Animals) who deserve a degree of moral consideration similar to that of human beings.
That sounds like a much better formulation. I think it also does well to avoid potentially insulting aspects of Singer's Marginal Cases.

Mind if I split this off into a new thread?
User avatar
ModVegan
Full Member
Posts: 123
Joined: Sun Dec 18, 2016 12:01 pm
Diet: Vegan
Location: Calgary, Alberta
Contact:

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by ModVegan »

brimstoneSalad wrote: Thu Dec 07, 2017 2:35 pm
ModVegan wrote: Thu Dec 07, 2017 11:06 am This is how Schwitzgebel constructs his argument:
Premise 1. If Entity A deserves some particular degree of moral consideration and Entity B does not deserve that same degree of moral consideration, there must be some relevant difference between the two entities that grounds this difference in moral status.
Premise 2. There are possible AIs (In our case, animals) who do not differ in any such relevant respects from human beings.
Conclusion. Therefore, there are possible AIs (Animals) who deserve a degree of moral consideration similar to that of human beings.
That sounds like a much better formulation. I think it also does well to avoid potentially insulting aspects of Singer's Marginal Cases.

Mind if I split this off into a new thread?
Sounds good to me!
User avatar
brimstoneSalad
neither stone nor salad
Posts: 10273
Joined: Wed May 28, 2014 9:20 am
Diet: Vegan

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by brimstoneSalad »

I wonder if we should mention this argument in the Wiki, since it seems to get at what Isaac wants to do with the argument, without the problems.
inator wrote: Thu Dec 07, 2017 2:30 pm
brimstoneSalad wrote: P5 "Any attempted defense of personal interests is an indication of self-value."

A problem here may be that some actions may appear to be an attempted defense of personal interests and actually just be an automatic response. Like plants or potentially some insects.
This doesn't contradict P5, because you'll argue that those instances are not 'true' attempts to defend personal interests given no existing interests. But to "plants have feelings" people it may seem like you're making ad-hoc justifications. A footnote is probably not necessary in this case, but I'm just putting it out there.
As a side effect, ended up splitting this too. Yep, I've encountered that, getting at intention seems to resolve that pretty easily.
Of course, even if we identify sentience or emotion, there's no way to avoid the "plants have feelings" arguments, it just calls for an empirical response i.e. "no they don't".
User avatar
DrSinger
Full Member
Posts: 134
Joined: Wed Oct 11, 2017 4:34 am
Diet: Vegan

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by DrSinger »

I think the current correction is good, and reasonably compelling (assuming it's valid). Informally it would run in basically the same way NTT currently is, where they have to name a trait that if applied to themselves would make themselves valueless, something that isn't encapsulated in this morally relevant difference argument.

The main issue I can see would be defending its first premise that humans only have moral value if and only if they have a certain trait e.g. sentience

http://philosophicalvegan.com/wiki/index.php?title=NameTheTrait#Correction
User avatar
ModVegan
Full Member
Posts: 123
Joined: Sun Dec 18, 2016 12:01 pm
Diet: Vegan
Location: Calgary, Alberta
Contact:

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by ModVegan »

brimstoneSalad wrote: Fri Dec 08, 2017 12:24 am I wonder if we should mention this argument in the Wiki, since it seems to get at what Isaac wants to do with the argument, without the problems.
inator wrote: Thu Dec 07, 2017 2:30 pm
brimstoneSalad wrote: P5 "Any attempted defense of personal interests is an indication of self-value."

A problem here may be that some actions may appear to be an attempted defense of personal interests and actually just be an automatic response. Like plants or potentially some insects.
This doesn't contradict P5, because you'll argue that those instances are not 'true' attempts to defend personal interests given no existing interests. But to "plants have feelings" people it may seem like you're making ad-hoc justifications. A footnote is probably not necessary in this case, but I'm just putting it out there.
As a side effect, ended up splitting this too. Yep, I've encountered that, getting at intention seems to resolve that pretty easily.
Of course, even if we identify sentience or emotion, there's no way to avoid the "plants have feelings" arguments, it just calls for an empirical response i.e. "no they don't".
Isaac is not a fan of relevant difference. I received a message from him that I don't think he'd mind me sharing, since it clarifies why he prefers NTT to the relevant difference argument:
"The huge disadvantage here is that with the relevance argument unless someone buys your subjective definition of relevance they can name whatever trait they want and deem it “relevant”, whereas with NTT all traits fall into the category of “produces contradiction” or “doesn’t produce contradiction”, so there is no need to squabble over how to define relevance."

In other words, he prefers depending on his interlocutor's acceptance of the logic of his premise.

Personally, I prefer placing the onus on the interlocutor to justify why a given difference is, in fact, relevant.
User avatar
brimstoneSalad
neither stone nor salad
Posts: 10273
Joined: Wed May 28, 2014 9:20 am
Diet: Vegan

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by brimstoneSalad »

Thanks for sharing. If Isaac is serious about defending his views, he should start an account here. :lol: He's not going to get banned.

Isaac, I know you're reading this:
This is basically a free speech forum, you can disagree with us as much as you want and cuss to your heart's content (most of us are mature enough not to do that, but we don't ban people for bad manners); your replies will remain visible, and they'll be inline in the threads so people can see your responses.
If you're confident enough to defend your views, you can do it here in a debate where nobody can shout over anybody else. :D
You're always welcome.
Isaac Brown wrote: The huge disadvantage here is that with the relevance argument unless someone buys your subjective definition of relevance they can name whatever trait they want and deem it “relevant”, whereas with NTT all traits fall into the category of “produces contradiction” or “doesn’t produce contradiction”, so there is no need to squabble over how to define relevance."
As you have explained the traits under consideration, including traits that are logically and metaphysically impossible to change, your P2 seems to contradict itself making the entire argument fail due to the principle of explosion.

I'll spare you the "Chinese", you can read about it here:
https://en.wikipedia.org/wiki/Principle_of_explosion

If you really maintain that ANY trait, including traits that are logically impossible to change based on the structure of P2, can be changed, a conclusion of "Therefore Santa Claus is real" would be just as valid, because ANYTHING follows from a contradiction.

So if P2 is inherently contradictory, it is not true that there are two categories "produces a contradiction" and "doesn't produce a contradiction"; they are one in the same, and mutually incoherent.

If you were to fix P2 by changing the wording or interpretation, you're either left with an obvious non sequitur or a circular argument that begs the question; neither are useful or convincing, and neither do what you want them to (the former leaves people to squabble over arbitrary exemptions, the latter is only an assertion).

ModVegan wrote: Fri Dec 08, 2017 4:33 pmIn other words, he prefers depending on his interlocutor's acceptance of the logic of his premise.
And in that case, he's really relying on confusion of what his premise means; if he explained it clearly I suspect nobody could accept it unless they were already committed to this.
A confusion driven argument can work, it has worked for theistic arguments in the past, but they don't have a great half-life, particularly on the internet where this kind of stuff gets debunked.
And unfortunately for veganism, with something like NameTheTrait, people take a debunking of the argument to be convincing evidence of the opposite position (unless it comes from a vegan, but it probably won't for most people).

ModVegan wrote: Fri Dec 08, 2017 4:33 pmPersonally, I prefer placing the onus on the interlocutor to justify why a given difference is, in fact, relevant.
I agree, I think this is the best and most eloquent framing of this kind of argument I've seen. It doesn't disparage the disabled, and it's also not absolutist in its conclusion, it doesn't say anything with certainty, but just opens up the possibility which may make it easier to keep people thinking about it rather than shutting it down.

I'd love to see how this works in practice.
Margaret Hayek
Junior Member
Posts: 93
Joined: Wed Nov 15, 2017 12:45 am
Diet: Vegan

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by Margaret Hayek »

ModVegan wrote: Thu Dec 07, 2017 11:06 am This is how Schwitzgebel constructs his argument:
Premise 1. If Entity A deserves some particular degree of moral consideration and Entity B does not deserve that same degree of moral consideration, there must be some relevant difference between the two entities that grounds this difference in moral status.
Premise 2. There are possible AIs (In our case, animals) who do not differ in any such relevant respects from human beings.
Conclusion. Therefore, there are possible AIs (Animals) who deserve a degree of moral consideration similar to that of human beings.
I think that something like this might be very good, but trying to use this in its present form to support the moral importance of non-human animals faces a big complication that stems from the fact that that we aren't concerned with possible non-human animals; we're concerned with actual ones - and indeed all sentient actual ones. Put in terms of actual non-human animals the argument looks like this:

P1. If Entity A deserves some particular degree of moral consideration and Entity B does not deserve that same degree of moral consideration, there must be some relevant difference between the two entities that grounds this difference in moral status.
P2. There are actual non-human animals who do not differ in any such relevant respects from human beings.
Conclusion. Therefore, there are actual non-human animals who deserve a degree of moral consideration similar to that of human beings.

The problems with this formulation are:

(1) Depending on the humans and non-human animals in question - and what you mean by 'relevant' and 'degree of moral consideration', P2 seems obviously false. E.g. your typical human audience member may have a life expectancy of 60 additional years, while your average mouse in the wild may have one of a few months. This is relevant to whom to save from death. Peter Singer (and others, like Jeff McMahan, David DeGrazia, Alastair Norcross, etc.) has argued very convincingly that it is NOT however relevant to whose equally sized interests (e.g. in avoiding suffering of a given intensity of which both humans and mice are capable) are more important. But this formulation doesn't get at how to make that argument. In contrast, offensive though it may be, Singer's does, since he makes the comparison to intellectually disabled humans who may have similarly short life expectancies, and clarifies that he's talking about equally sized interests, not other sorts of things that can be relevant (like how much certain things - like death - may harm different individuals)

(2) The conclusion doesn't really go far enough; we don't want this to just support an argument for e.g. giving moral consideration to the smartest non-humans like chimps and dolphins, but to all sentient non-human animals. Again, even though it might be insulting to some (but honestly if pitched right not to me, and I'm someone who has had profoundly intellectually disabled humans in my life), Singer's argument from less able humans fixes this by comparing any given sentient non-human animal to a sentient human whose capacities are comparable to hers.
User avatar
ModVegan
Full Member
Posts: 123
Joined: Sun Dec 18, 2016 12:01 pm
Diet: Vegan
Location: Calgary, Alberta
Contact:

Re: Schwitzgebel's "no relevant difference" like NTT (but valid?)

Post by ModVegan »

Margaret Hayek wrote: Sat Dec 09, 2017 12:35 pm I think that something like this might be very good, but trying to use this in its present form to support the moral importance of non-human animals faces a big complication that stems from the fact that that we aren't concerned with possible non-human animals; we're concerned with actual ones - and indeed all sentient actual ones.
That's a very fair point, Margaret, and the reason I prefer this argument (as I shared with Isaac) is that I'm more interested in exploring the case for AI rights than I am for animals, at this point in time (with animals I think the best arguments we currently have come from economic efficiency and sustainability, not moral logic). I like Schwitzgebel's argument (found elsewhere in his work) that we ought to take other beings "at their word" (in other words, if an animal exhibits signs of pain, it's best to assume they actually feel pain until proven otherwise. Similarly, if animals appear to be "subjects of a life" it's best to take that at face value until we have more information that might disprove this.
Margaret Hayek wrote: Sat Dec 09, 2017 12:35 pm (2) The conclusion doesn't really go far enough; we don't want this to just support an argument for e.g. giving moral consideration to the smartest non-humans like chimps and dolphins, but to all sentient non-human animals..Singer's argument from less able humans fixes this by comparing any given sentient non-human animal to a sentient human whose capacities are comparable to hers.
Perhaps, although relevant difference would ask (much like NTT) what is the relevant difference between a mentally disabled human and pig that allows us to keep one in the equivalent of a lifelong concentration camp, while cherishing the other.

I will definitely grant that relevant difference cannot be used to make extraordinary claims. But it's a good placeholder until we learn more about the true similarities between animal and human sentience.
Post Reply