Why is it more rational to believe that other people are sentient rather than have no belief one way or the other?

[Sicnoo0]

I'm not talking about certainty; disproving solipsism is impossible, after all.
I'm talking about likelihood.

Any given organism is either sentient or not.

Why should I have any inclination to believe strongly one way or the other as to whether a given organism is sentient?
Why shouldn't I just remain entirely agnostic as to whether anyone is sentient or not?

If I'm 50/50 on whether or not others are sentient, then I will still treat others with kindness as though they are sentient, because there is a chance that they might in fact be sentient.

So it's not like being 50/50 on the matter would turn me into a horrible person as far as the consequences of my beliefs are concerned.

And it's not like any of that matters in the first place; a belief having bad consequences does not imply the belief is in disagreement with reality.

[FredVegrox]

It is better to act on the apparent sentience, instead of assuming the evidence of it is deceptive.

[Red]

Well, how can I know if you're sentient? Would you appreciate it if I treated things as if you weren't?

[brimstoneSalad]

I'm not talking about certainty; disproving solipsism is impossible, after all.
I'm talking about likelihood.

An unknowable probability should not be treated as 50/50 simply because you have only imagined two scenarios.
This is the same issue Pascal's wager has. States do not equal to probability, and when you assert there are only two model-able states, you are inevitably wrong. For instance, the third state of the devil running the show in Pasal's wager itself.

It is a way to model decisions and their consequences when you have to in approach to a particular state challenge, but that doesn't mean genuine agnosticism about the states is warranted.

Any given organism is either sentient or not.

How about a hypothetical state of anti-sentience?
Also, it's not a dichotomy. Sentience is a gradation, and can even apply differently to different dimensions of experience and comprehension of ourselves and our place in reality.

The bottom line is that given an unknown like this, you treat the mind as a black box, not something for you to make up models on and assign arbitrary percentages to. You must assume the state that the black box gives evidence for following Occam's razor, which in most animals is varying degrees of sentience.

Why should I have any inclination to believe strongly one way or the other as to whether a given organism is sentient?

Because, any uncountable number of ad hoc hypotheses deserve zero consideration on a decision theory table. Theoretically, they all cancel each other out anyway.

The legitimacy of scientific evidence for sentience is not just one of two viable options, it's the only option if you are to take the approach that the world is knowable and that knowledge can be meaningful: and that IS a valid dichotomy you can run through a decision theory table.

And it's not like any of that matters in the first place; a belief having bad consequences does not imply the belief is in disagreement with reality.

Well, what you're talking about here is not whether it's in agreement with reality, but whether any aspect of knowledge of reality is meaningful at all. Questioning sentience opens a much bigger can of worms.