Great comments on new #namethetrait video

[DrSinger]

You actually found a flaw here in P2. Indeed if you consider the set of human to be included in the set of animals ( which it is ), the trait t can't satisfy both propositions you mentioned. If we take a human X, then A(X) => t ∉ T(X) will impose the constraint that the Human X do not have the trait while at the same time looking for the same trait if we instantiate y = X. So we must either :
Consider human to be non animals ( no inclusion )
Adding a proposition/predicate that will ensure the trait is taken out of the set A(x)\H(x) ( interestingly enough I found in my notes back then that I did something similar.
Thanks for spotting that I will add it to the wiki when I have time, unless you want to take a crack at it

Do you think changing the definitions would suffice, so that the sets are distinct? I think that would be fair. AY has said before he's obviously considering humans as distinct from animals in his argument.

Also, do you agree with me that the flaw is present in the original argument as well (that 'human cannot be absent in a human', provided possessing 'human' makes one human)? Because in that case we would be justified from excluding it from the set of traits.

Adding a proposition/predicate that will ensure the trait is taken out of the set A(x)\H(x) ( interestingly enough I found in my notes back then that I did something similar.

This may be an idea. I haven't had the chance to look into fully but I will very soon.

Curious to know what you mean here.

My meaning was some sort of time permitting logic so we could say something like 'H(x) then -H(x)', something like that.


I think we could probably show in the case that we allow 'being part of the set of humans' to be a trait, then P2 becomes vacuously true. And if we don't allow it to be a trait, we can simply point to the moral system 'humans have moral value, nonhumans do not', to show that C does not follow from P1&P2

EDIT: Attempted to show this on the wiki, would love to hear any thoughts on it

[mkm]

I think we could probably show in the case that we allow 'being part of the set of humans' to be a trait, then P2 becomes vacuously true. And if we don't allow it to be a trait, we can simply point to the moral system 'humans have moral value, nonhumans do not', to show that C does not follow from P1&P2

EDIT: Attempted to show this on the wiki, would love to hear any thoughts on it

I think you have shown even more (or maybe it can be reformulated more clearly)- that if there is a member x s.t. A(x) and H(x) then P2 is true, proof basicly intact, since if t is satisfying P2 then we have not(T(x)) by (A(x) then not(T(x))) and T(x) by (H(x) then T(x)), which is a contradiction.
I'm not sure about the second part. Do you want to propose a model in which sets A={x : A(x)} and H={x : H(x)} are disjoint, but P2 is still true?

I would suggest a minor change in formulation of P2. Right now there is ... T(q) = T(y) \ { t } ... and it's not clear that there is such q that T(q) = T(y) \ { t } . If not, again P2 is vacously true. It's not unimagenable, that most traits are associated with each other and we can't remove just one. What do you thinkof changing it to: T(q) is a subset of T(y)\{t} ?

PS: Excuse me my notation, I am yet to learn using math symbols on the forum...

[DrSinger]

I think you have shown even more (or maybe it can be reformulated more clearly)- that if there is a member x s.t. A(x) and H(x) then P2 is true, proof basicly intact, since if t is satisfying P2 then we have not(T(x)) by (A(x) then not(T(x))) and T(x) by (H(x) then T(x)), which is a contradiction.

The section could definitely be explained better, I'm still trying to piece it all together. I'm sure it could be shown more formally, and if you could do this that would be good.

I'm not sure about the second part. Do you want to propose a model in which sets A={x : A(x)} and H={x : H(x)} are disjoint, but P2 is still true?

My intention was to try and show that if we force them to be disjoint (I think this is implicit in the argument in the first place), then the argument becomes an obvious non sequitur, since there's nothing to give animals moral value or give them the trait 'human'. Do you think it's not possible for P2 to be true with A and H disjoint?

I would suggest a minor change in formulation of P2. Right now there is ... T(q) = T(y) \ { t } ... and it's not clear that there is such q that T(q) = T(y) \ { t } . If not, again P2 is vacously true. It's not unimagenable, that most traits are associated with each other and we can't remove just one. What do you thinkof changing it to: T(q) is a subset of T(y)\{t} ?

I didn't come up with the formulation but i think 'T(q) = T(y) \ { t }' is a more accurate translation of the argument, AY talks about it as if you can remove traits independently.

PS: Excuse me my notation, I am yet to learn using math symbols on the forum...

I just copy and paste them

[Nightcell001]

Do you think changing the definitions would suffice, so that the sets are distinct? I think that would be fair. AY has said before he's obviously considering humans as distinct from animals in his argument.

Formally no. We need to explicitly define it. It's not hard though.

Also, do you agree with me that the flaw is present in the original argument as well

Yes totally. But AY argument is informal to begin with so ...

I would suggest a minor change in formulation of P2. Right now there is ... T(q) = T(y) \ { t } ... and it's not clear that there is such q that T(q) = T(y) \ { t } . If not, again P2 is vacously true. It's not unimagenable, that most traits are associated with each other and we can't remove just one. What do you thinkof changing it to: T(q) is a subset of T(y)\{t} ?

Indeed, the argument is Made to handle the case if not such q exists. It is however designed ti translate AY's NTT. If you are talking about subsets we are going toward a different argument ( which is totally fine by the way ).

∄x: (A(x) ∧ H(x)) or equivalently ∄x: (A(x) ∧ h ∈ T(x)).
(there is no x that is both human and animal)

The formulation is fine, but it doesn't translate to reality. Human are classify as animals, I'm not sure we can differentiate between the two. Basically thinking about it in terms of Venn diagram, human is a subset of animals, so we can't pick a trait absent in human animals and present in humans because a trait absent in animals is absent in human as well. If we consider non human animals the argument works fine.

I think replacing the beginning of P2 by :
∃t: t∈T ∧ (∃x: t∈T(x) ∧ H(x)) ∧ (∀x: A(x) ∧ ¬H(x) => t∉T(x))
Should suffice. I need to spend more time on it and I got to run now...

[brimstoneSalad]

Also, do you agree with me that the flaw is present in the original argument as well

Yes totally. But AY argument is informal to begin with so ...

Informal, or an ambiguous and logically invalid attempt at formal argument?

Isaac seems to believe his argument is formally valid, and he put it into a form P1 P2 C etc. which is to create the appearance of a formal argument. I would say that is a significant part of the problem.

I think the flaw is present in the original given a literal reading, and in the way Isaac thinks it works/what answers he has humored in debate.

I'm very interested in seeing what you work out! Following along, sorry I can't contribute much.

[Nightcell001]

Informal, or an ambiguous and logically invalid attempt at formal argument?

Isaac seems to believe his argument is formally valid, and he put it into a form P1 P2 C etc. which is to create the appearance of a formal argument. I would say that is a significant part of the problem.

That's true. The layout of his argument makes us think it is formal.

I'm very interested in seeing what you work out! Following along, sorry I can't contribute much.

Found out that it is much easier than what I wrote earlier.
Replace :

P2:⇔ ¬ ( ∃t: ( ∀x: A(x) ⇒ t ∉ T(x) ) ∧ ( ∀y: H(y) ⇒ ( t ∈ T(y) ∧ ( ∀q: ( T(q) = T(y) \ { t } ) ⇒ ¬ M(q) ) )

by :

P2:⇔ ¬ ( ∃t: ( ∀x: (A(x) ∧ ¬H(x)) ⇒ t ∉ T(x) ) ∧ ( ∀y: H(y) ⇒ ( t ∈ T(y) ∧ ( ∀q: ( T(q) = T(y) \ { t } ) ⇒ ¬ M(q) ) )

We just making it clear that the trait should be absent in non human animals.

[DrSinger]

Found out that it is much easier than what I wrote earlier.
Replace :

P2:⇔ ¬ ( ∃t: ( ∀x: A(x) ⇒ t ∉ T(x) ) ∧ ( ∀y: H(y) ⇒ ( t ∈ T(y) ∧ ( ∀q: ( T(q) = T(y) \ { t } ) ⇒ ¬ M(q) ) )

by :

P2:⇔ ¬ ( ∃t: ( ∀x: (A(x) ∧ ¬H(x)) ⇒ t ∉ T(x) ) ∧ ( ∀y: H(y) ⇒ ( t ∈ T(y) ∧ ( ∀q: ( T(q) = T(y) \ { t } ) ⇒ ¬ M(q) ) )

We just making it clear that the trait should be absent in non human animals.

I think this is a good solution. This would also mean the trait 'human' cannot be present in all animals. In which case neither 'moral value' nor 'human' is present in all animals, making the argument a non sequitur. Do you think we can show this?

I changed the translation back to how it was originally as it better reflects the literal interpretation of NTT. I've added the 'fix' to the section 'Separating humans and nonhuman animals'

PV I also added a meme based on AY's latest vid. Feel free to remove it if you want

Is it possible to increase the max file size on the wiki? I'm having trouble getting Sisyphus Redeemed clips below 2mb with the method I'm using (I can get them below 3mb)

EDIT: I was thinking if we allowed moral value to be a trait then the only way it could satisfy being absent in nonhuman animals and present in humans is if nonhuman animals do not have moral value

[DrSinger]

What do we think of this as an analogy to name the trait. I think it's more obvious than the name the verse analogy and people are likely to agree with P1 & P2 and reject the conclusion

P1 - It is not wrong to kill yourself
P2 - There is no trait absent in others which if absent yourself would cause it to be wrong to kill yourself
C - Therefore without establishing the absence of such a trait in others, we contradict ourselves by deeming it wrong to kill others

'have moral value' becomes 'not wrong to kill'
'humans' become 'yourself'
'animals' become 'others'

[PhilRisk]

First I have to say, that I have been busy and therefore did not had the opportunity to participate in debate.

Now some points I want to say something on:

It's not unimagenable, that most traits are associated with each other and we can't remove just one.

I had stated this in an earlier comment

The problematic aspect is, that losing this single trait and not any other trait. For mosts traits that is simply not possible, because traits have nomological relations. E. g. if you are losing the trait having human genome, you lose the trait to synthesize specific proteins. You would be losing more than just one trait and therefore could not give a counterexample to P2.

This means it is harder to give a counterexample to P2, but does not have to be problematic concerning the validity of the argument.

Leaving this aside I thought about a model fulfilling P1 and P2 and not the conclusion:
I made a case analysis based on P2:
P2 is true, when there is a nonhuman animal with all possible traits. such a being could not be an animal, therefore I don't consider this option as a possible model. But if this would be possible, the conclusion would not follow, because a world with this 'animal' having all traits and all other nonhuman animals having no moral value, would fulfill the conditions and be a counterexample.
P2 is true, if for every trait t, there is a human who don't have t or there is some object, which is identical to a human except t and has moral value.
Imagine a world where there exist three object: One is a human, one is an nonhuman animal without moral value and the other is a strange being which has the same traits as the human except being human and has moral value. t is "being human".
P1 and P2 are correct and not all but one and only one nonhuman animal has moral value.
(Hopefully I did not got confused by all the negations and quantifications)

This would mean that the argument might really establish that there need to something of moral value that is not a human being. I did not come up to a counterexample for this. But it does not leads to the conclusion that all animals have moral value, it could be anything nonhuman.

[DrSinger]

I think we should treat traits as if we can remove them independently, as a steel-man that we can probably prove to be a non sequitur anyway

Maybe you could try 'plugging in' that case into P2 and see if it works out the way you think it does. I think we should be able to give an example where C doesn't follow from P1&P2