Great comments on new #namethetrait video

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.
Post Reply
User avatar
brimstoneSalad
neither stone nor salad
Posts: 10273
Joined: Wed May 28, 2014 9:20 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by brimstoneSalad »

Margaret Hayek wrote: Sun Dec 10, 2017 2:19 pmThis seems to go against the original English presentation of NTT, which says "there are no traits absent in animals which if absent in humans would cause us to deem ourselves valueless," so the entities losing the traits were clearly supposed still to be us after losing them.
It also conflicts with how Isaac presents the argument, asking "if your DNA were changed so you were no longer human" or "if your consciousness were transmitted into a cow" and all other manner of science fictional scenarios.

I have yet to see him invoke a time machine scenario that "If somebody went back in time and made it so you were born a cow instead of a human".
Margaret Hayek
Junior Member
Posts: 93
Joined: Wed Nov 15, 2017 12:45 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by Margaret Hayek »

DrSinger wrote: Fri Dec 08, 2017 5:18 am Margaret can you check over this

http://philosophicalvegan.com/wiki/index.php/NameTheTrait#Part_1_Counterexample

I am not sure if I have done the instantiation thing correctly, it is something I saw nightcell do
Unfortunately I don't think that those counterexamples quite work, because they don't have the right sort of existential import (sorry I didn't note this earlier). You can check this with your theorem prover, to see the conditional "IF (i), (ii), and (iii), THEN P1, P2, and Not C" is valid, and it appears it isn't:

Conjunction of P1 and P2 and not C:

(\forallx ( Hx \to Mx ) \land \forallx ( Ax \to \neg \existst ( Tt \land \neg Pxt \land ( \forally (Cy \to ( \neg Pyt \to \neg My ) ) ) ) )\land\neg(Ax \to Mx))


Conjunction of (i), (ii), and (iii):
(\forallx ( Hx \to Mx ) \land \forallt\forallx( Ax \land Tt \land ( Pxt \lor \negPxt) \to \negMx )\land\forallt\forallx( Cx \land Tt \land ( Pxt \lor \negPxt) \to Mx ))

IF the latter then the former:
(\forallx ( Hx \to Mx ) \land \forallt\forallx( Ax \land Tt \land ( Pxt \lor \negPxt) \to \negMx )\land\forallt\forallx( Cx \land Tt \land ( Pxt \lor \negPxt) \to Mx ))\to(\forallx ( Hx \to Mx ) \land \forallx ( Ax \to \neg \existst ( Tt \land \neg Pxt \land ( \forally (Cy \to ( \neg Pyt \to \neg My ) ) ) ) )\land\neg(Ax \to Mx))

As you can see, the countermodel had a completely empty domain.


I think that a really simple (but not completely irrelevant) countermodel to NTT part 1 would be one where:

There is exactly one sentient human; the set of sentient humans humans = {h}
There is exactly one sentient non-human animal; the set of sentient non-human animals = {a}
There is exactly one counterpart of a sentient human; the set of counterparts of sentient humans = {c}
There is exactly one (relevant) trait; the set of traits = {t}
The human and the counterpart have moral value and the non-human animal doesn't (nor does the trait!); the set of things with moral value = {h, c}
The human has the trait and the counterpart and the non-human (and the trait!) don't; the set of things that have t = {h}

You could if you like play around with generalizations of this, and check to see if the conditional (IF conjunction of features of counterexample, THEN P1, P2, & negation of C) is valid.

In order to actually specify a reasonably realistic counterexample like the above (which has existential import but doesn't require that the sets of the relevant things are ridiculously small), I think that you may have to say something like this (remember it's OK that counterparts aren't humans, even on the non-alternative interpretation, because humans are humans as they actually are and counterparts are humans as they could be under counterfactual circumstances):

(i) ∃x(Hx) ∧ ∀x(Hx⇒(¬Ax∧¬Tx∧¬Cx∧Mx))
(ii) ∃x(Ax) ∧ ∀x(Ax⇒(¬Hx∧¬Tx∧¬Cx∧¬Mx∧(∃y(Cy∧∀t(¬Pxt⇒¬Pyt))))
(iii) ∀x(Cx⇒(¬Ax∧¬Tx∧¬Hx∧Mx)
(iv) Tx⇒(¬Ax∧¬Hx∧¬Cx∧¬Mx)
User avatar
DrSinger
Full Member
Posts: 134
Joined: Wed Oct 11, 2017 4:34 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by DrSinger »

Thanks, it works!

Counterexample

((\existsx(Hx)\land\forallx(Hx\to(\negAx\land\negTx\land\negCx\landMx)))\land(\existsx(Ax)\land\forallx(Ax\to(\negHx\land\negTx\land\negCx\land\negMx\land(\existsy(Cy\land\forallt(\negPxt\to\negPyt))))))\land(\forallx(Cx\to(\negAx\land\negTx\land\negHx\landMx)))\land(Tx\to(\negAx\land\negHx\land\negCx\land\negMx)))

NTT negated conclusion

(\forallx ( Hx \to Mx ) \land \forallx ( Ax \to \neg \existst ( Tt \land \neg Pxt \land ( \forally (Cy \to ( \neg Pyt \to \neg My ) ) ) ) ) \to\neg\forallx (Ax \to Mx))

Counter example -> NTT negated conclusion

((\existsx(Hx)\land\forallx(Hx\to(\negAx\land\negTx\land\negCx\landMx)))\land(\existsx(Ax)\land\forallx(Ax\to(\negHx\land\negTx\land\negCx\land\negMx\land(\existsy(Cy\land\forallt(\negPxt\to\negPyt))))))\land(\forallx(Cx\to(\negAx\land\negTx\land\negHx\landMx)))\land(Tx\to(\negAx\land\negHx\land\negCx\land\negMx)))\to(\forallx ( Hx \to Mx ) \land \forallx ( Ax \to \neg \existst ( Tt \land \neg Pxt \land ( \forally (Cy \to ( \neg Pyt \to \neg My ) ) ) ) ) \to\neg\forallx (Ax \to Mx))
Gregor Samsa
Junior Member
Posts: 62
Joined: Tue Oct 03, 2017 10:16 pm
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by Gregor Samsa »

http://philosophicalvegan.com/wiki/index.php/NameTheTrait#Correction´

Is there a reason why it needs to be both a sufficient and necessary condition? There could be several sufficient traits / sets of traits for moral value; as vegans we're only concerned that animals have at least one of them (sentience). No?

P1) There exists a trait or a set of traits T, such that, if an individual is a human then the individual has R if they possess T

(P2) If humans have R if they possess T, then all beings have R if they possess T.

(P3) If an individual is a sentient nonhuman animal, then there is no trait absent in the individual, which if absent in a human, would cause the human to not have R.

Therefore (C) Sentient nonhuman animals have R

Might be missing something. Please explain :)

(perhaps you'd need to rephrase p3 to something like: "If an individual is a sentient nonhuman animal, then there is no trait or set of traits absent in that individual, which if absent in a human, would cause the human to not have T and therefore not have R". But I'm not sure if that's necessary).
Last edited by Gregor Samsa on Mon Dec 11, 2017 3:51 pm, edited 1 time in total.
Daz
Newbie
Posts: 47
Joined: Sat Dec 02, 2017 7:04 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by Daz »

brimstoneSalad wrote: Mon Dec 04, 2017 10:26 pm
Can you answer my questions, so I can understand where your confusion lies?
I suspect it lies in a misunderstanding of what "validity" means in logic.

P1. Bob is using an arbitrary excuse
P2. Some people who use arbitrary excuses are contradicting themselves

So what is the appropriate conclusion?

1. C. Therefore Bob is contradicting himself
2. C. Therefore Bob may or may not be contradicting himself (we don't have enough information to determine this)
Well IF Bob's ONLY excuse is an arbitrary one and he cannot name any other excuses, then YES Bob is contradicting himself if he states that it is both OK and not OK to use this excuse for justification in the same situation at the same time. How do we determine if he has any other excuses you ask? Well, we ask him to name any others he may have (NTT). Or ask him if his arbitrary excuse is his only one... If he does have other excuses we can examine these one by one to see if they produce contradictions... That way we have all the information we need to determine this. Your P2 above is misleading and not relevant for this argument, for it would more appropriately read:
P2. People who rely SOLELY on arbitrary excuses are contradicting themselves if they both accept and reject this as an excuse at the same time.

brimstoneSalad wrote: Mon Dec 04, 2017 2:44 pm P1. Bob is a transvestite
P2. Some transvestites are gay
C. Therefore Bob is gay.

We know that argument is invalid. What if Bob is actually gay? Does that make it valid in your view?
No of course not.



brimstoneSalad wrote: Mon Dec 04, 2017 2:44 pmI have linked you to this before, please read it so you understand the definition of validity:
http://www.iep.utm.edu/val-snd/
I understand the definition of validity.
brimstoneSalad wrote: Mon Dec 04, 2017 2:44 pm
If they believe sometimes arbitrary justifications are OK and sometimes they are not, they are employing a double standard, they are not contradicting themselves.

A double standard is not a logical contradiction. Do you understand and agree that these are different things?

Please answer that question so I can see where the misunderstanding here lies.
I understand that a double standard isn't always a logical contradiction however sometimes they are depending on the details.
If someone believes arbitrary justifications are OK for a specific treatment of another person (say Bob) however not ok as a justification for that same treatment of themselves or their family/friends etc (say Lucy) then for that specific treatment/circumstance they are contradicting themselves. It's ok for me to enslave Bob and make him work for me against his will because he has brown eyes, however it is not ok for you to enslave Lucy because she has brown eyes (because she is my sister and I don't care about Bob). In here lies a contradiction if the only justification is the arbitrary one of eye colour. If it is ok as justification and it isn't ok as justification at the same time, this is a contradiction. Do you understand this?





brimstoneSalad wrote: Mon Dec 04, 2017 2:44 pm
Nothing in #NameTheTrait says the trait has to be rationally justified; it can be an arbitrary one.
Yes and if they wouldn't accept that as an excuse for treating themselves or their loved ones in the same way then they are contradicting themselves by deploying it as an excuse for the treatment of others. It is an ok justification and at the same time not an ok justification. Contradiction.

P1 - it is OK to enslave Bob because he has brown eyes
P2 - it is NOT OK to enslave Lucy because she has brown eyes

P1 and P2 contradict each other if there are no other factors distinguishing Bob and Lucy apart from eye colour and eye colour is used as the excuse for enslavement. It cannot be both OK to enslave based on brown eyes and not OK to enslave based on brown eyes at the same time. If they have more excuses each one can be examined individually to see if they produce contradictions also.
User avatar
DrSinger
Full Member
Posts: 134
Joined: Wed Oct 11, 2017 4:34 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by DrSinger »

Daz wrote:P1 - it is OK to enslave Bob because he has brown eyes
P2 - it is NOT OK to enslave Lucy because she has brown eyes
This isn't a contradiction because the statements are not direct negations of each other, you need 'p and not p' or 'p∧¬p' for there to be a contradiction. A contradiction would be

P1 - it is OK to enslave Bob because he has brown eyes
P2 - it is NOT OK to enslave Bob because he has brown eyes
User avatar
DrSinger
Full Member
Posts: 134
Joined: Wed Oct 11, 2017 4:34 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by DrSinger »

Gregor Samsa wrote: Mon Dec 11, 2017 3:06 pm http://philosophicalvegan.com/wiki/index.php/NameTheTrait#Correction´

Is there a reason why it needs to be both a sufficient and necessary condition? There could be several sufficient traits / sets of traits for moral value; as vegans we're only concerned that animals have at least one of them (sentience). No?

P1) There exists a trait or a set of traits T, such that, if an individual is a human then the individual has R if they possess T

(P2) If humans have R if they possess T, then all beings have R if they possess T.

(P3) If an individual is a sentient nonhuman animal, then there is no trait absent in the individual, which if absent in a human, would cause the human to not have R.

Therefore (C) Sentient nonhuman animals have R

Might be missing something. Please explain :)

(perhaps you'd need to rephrase p3 to something like: "If an individual is a sentient nonhuman animal, then there is no trait or set of traits absent in that individual, which if absent in a human, would cause the human to not have T and therefore not have R". But I'm not sure if that's necessary).
The reason having the trait has to be a sufficient and necessary condition is so that in P3 if a human loses the trait, then they are guaranteed to lose R, otherwise P3 wouldnt do what it's supposed to do. It does make the first premise slightly more difficult to defend I guess.

You can test it out in the proof generator, if you change the biconditionals to a conditional, it takes a while but it comes out invalid


https://www.umsu.de/logik/trees/

\existst ( Tt \land \forallx (Hx \to (Pxt\to Rx ) ) ) \land \forallt ( Tt \land ( \forallx (Hx \to ( Pxt \to Rx)) \to \forallx (Pxt\toRx))) \land \forallx (Ax \to \neg \existst (Tt \land \negPxt \land (\forally (Hy \to ( \negPyt \to \negRy))))) \to \forallx ( Ax \to Rx )
Margaret Hayek
Junior Member
Posts: 93
Joined: Wed Nov 15, 2017 12:45 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by Margaret Hayek »

DrSinger wrote: Mon Dec 11, 2017 5:14 am Thanks, it works!

Counterexample

((\existsx(Hx)\land\forallx(Hx\to(\negAx\land\negTx\land\negCx\landMx)))\land(\existsx(Ax)\land\forallx(Ax\to(\negHx\land\negTx\land\negCx\land\negMx\land(\existsy(Cy\land\forallt(\negPxt\to\negPyt))))))\land(\forallx(Cx\to(\negAx\land\negTx\land\negHx\landMx)))\land(Tx\to(\negAx\land\negHx\land\negCx\land\negMx)))

NTT negated conclusion

(\forallx ( Hx \to Mx ) \land \forallx ( Ax \to \neg \existst ( Tt \land \neg Pxt \land ( \forally (Cy \to ( \neg Pyt \to \neg My ) ) ) ) ) \to\neg\forallx (Ax \to Mx))

Counter example -> NTT negated conclusion

((\existsx(Hx)\land\forallx(Hx\to(\negAx\land\negTx\land\negCx\landMx)))\land(\existsx(Ax)\land\forallx(Ax\to(\negHx\land\negTx\land\negCx\land\negMx\land(\existsy(Cy\land\forallt(\negPxt\to\negPyt))))))\land(\forallx(Cx\to(\negAx\land\negTx\land\negHx\landMx)))\land(Tx\to(\negAx\land\negHx\land\negCx\land\negMx)))\to(\forallx ( Hx \to Mx ) \land \forallx ( Ax \to \neg \existst ( Tt \land \neg Pxt \land ( \forally (Cy \to ( \neg Pyt \to \neg My ) ) ) ) ) \to\neg\forallx (Ax \to Mx))
Great; I'm very glad to hear that.

Would you like to add a corresponding counterexample for part 2 of NTT?

Also, you might want to provide an intuitive explanation of why the counterexample makes the premises true and the conclusion false (instead of just pointing people to the theorem prover). It shouldn't be hard to do.

I'd also recommend including, if you can, a natural deduction proof of the validity of your corrected version / the validity of things that you say are valid. I know that I steered you in the direction of truth trees, but when the trees are long (as they are in the case of your corrected version) they can fail to provide clear insight into why the argument is valid - and they certainly fail to do this quickly.

In order to understand the basics of natural deduction you would probably only need to read the Teller chapters on

(1) fundamentals of natural deduction for sentence / propositional logic (volume I chapter 5): http://tellerprimer.ucdavis.edu/pdf/1ch5.pdf/view, and
(2) fundamentals of natural deduction for predicate / FO logic (volume II chapter 5): http://tellerprimer.ucdavis.edu/pdf/2ch5.pdf/view.

You might also be able to find an online natural deduction theorem prover to help you / check things. Unless I'm mistaken such a theorem prover was one of the early impressive accomplishments in AI (as I recall it wasn't just a prover that used the truth tree method). Of course, you'll want to cite the theorem prover if you include a proof that it generates (as I recall the early theorem proving AI co-authored some papers!).

Finally when you're linking to resources on logic (e.g. on how to prove things), why not just link to the relevant chapters from the Teller book (or some similar resource)? As I see it there's no need to reinvent the wheel here and have the phil vegan wiki have its own pages on things like this that are very general and not specifically relevant only to veganism in particular.
Margaret Hayek
Junior Member
Posts: 93
Joined: Wed Nov 15, 2017 12:45 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by Margaret Hayek »

DrSinger wrote: Mon Dec 11, 2017 8:05 pm
Gregor Samsa wrote: Mon Dec 11, 2017 3:06 pm http://philosophicalvegan.com/wiki/index.php/NameTheTrait#Correction´

Is there a reason why it needs to be both a sufficient and necessary condition? There could be several sufficient traits / sets of traits for moral value; as vegans we're only concerned that animals have at least one of them (sentience). No?

P1) There exists a trait or a set of traits T, such that, if an individual is a human then the individual has R if they possess T

(P2) If humans have R if they possess T, then all beings have R if they possess T.

(P3) If an individual is a sentient nonhuman animal, then there is no trait absent in the individual, which if absent in a human, would cause the human to not have R.

Therefore (C) Sentient nonhuman animals have R

Might be missing something. Please explain :)

(perhaps you'd need to rephrase p3 to something like: "If an individual is a sentient nonhuman animal, then there is no trait or set of traits absent in that individual, which if absent in a human, would cause the human to not have T and therefore not have R". But I'm not sure if that's necessary).
The reason having the trait has to be a sufficient and necessary condition is so that in P3 if a human loses the trait, then they are guaranteed to lose R, otherwise P3 wouldnt do what it's supposed to do. It does make the first premise slightly more difficult to defend I guess.

You can test it out in the proof generator, if you change the biconditionals to a conditional, it takes a while but it comes out invalid


https://www.umsu.de/logik/trees/

\existst ( Tt \land \forallx (Hx \to (Pxt\to Rx ) ) ) \land \forallt ( Tt \land ( \forallx (Hx \to ( Pxt \to Rx)) \to \forallx (Pxt\toRx))) \land \forallx (Ax \to \neg \existst (Tt \land \negPxt \land (\forally (Hy \to ( \negPyt \to \negRy))))) \to \forallx ( Ax \to Rx )
I wonder if there is a way to tweak P3 so that in conjunction with P1 and P2 (or minor tweaks thereof) it entails that if humans have a sufficient but not necessary condition for the relevant moral consideration / treatment, then non-human animals have the sufficient but not necessary condition for the relevant moral consideration / treatment too (ideally without straying too much from the original text of NTT's P2s). The reason is that, as I think you're both observing, it's always better for things to be logically weaker than logically stronger if they can entail the same conclusion - it would be best not to have to rule out other grounds for the morally relevant treatment besides the ones under discussion.
User avatar
DrSinger
Full Member
Posts: 134
Joined: Wed Oct 11, 2017 4:34 am
Diet: Vegan

Re: Great comments on new #namethetrait video

Post by DrSinger »

I removed the notion of the counterpart from the article (up to the alternate version). I think it is confusing and is not actually a steelman, since it removes any link between the first and second premises, so if the argument was valid it would have to be question begging by default. Based on subsequent comments by AY it's obvious that 'us/ourselves' is supposed to be the thing referred to in P1 i.e. humans, which is how I had always interpreted it, and everyone else seems to. This way the argument can also be more easily compared to the correction as well
Last edited by DrSinger on Tue Dec 12, 2017 4:43 am, edited 1 time in total.
Post Reply