[FRIAM] Newborn Heart Rate

uǝlƃ ☤>$ gepropella at gmail.com
Mon Oct 4 11:10:44 EDT 2021


So, here again, we seem to be dancing around the hegemony [ξ] of consistency. EricS brings in "coherence", which I like better. But I think it's the same concept. Monism, "not being self-contradictory", objective Bayesian priors, coherence, the ontological status of actual infinities, integrated personality, value alignment, partition/predicate crispness, XOR choices, etc. all target the same fundamental bogey: 

   inconsistency

And that's fine. But it seems, to my biased eye, that we usually leave "completeness" to take care of itself ... as part of the negative space in the picture. The best definition I've seen of completeness is from a presentation by Greg Restall (paraphrasing): "If X models A completely, then we can derive A from X." I like this because it smells like reachability, "can we get there from here". When we harp too much on not being inconsistent, we end up in some sort of word game ... like some wak logicbro trying to pwn the libs. But when we talk about completeness, we talk about what is *sayable* in our language ... It's less about what we can't say and more about what we can say.

That makes consistency the spastic little sibling of completeness. Yes, mom told me I have to take it along with me on the bike ride. But everyone hates it because it never shuts up and always says stupid stuff.

[ξ] I wanted to use a new phrase, "linguistic salience bias", in place of "hegemony". But my epistemic status for the use of that phrase is 50%. Hegemony has a nice political tone, too. I kinda like dominance or tyranny. Maybe I should have gone with "gravity well" to indicate that consistency is a kind of least common denominator ... the type of thing people like grammar nazis and logicbros focus on. But I'd rather highlight the more accurate state of affairs, which is that those who study expressibility are underclass citizens compared to those who study correctness. Sure, when the expressors finally "make it" (such that nobody can deny their impact --- think Tom Waits, not Elon Musk), we all gather round and use them as an excuse to party. But we never go back and knead the tortuous pipeline of consistency they *survived* to get there.

On 10/3/21 9:41 AM, Marcus Daniels wrote:
> A compiler for a programming language with an advanced type system can essentially reject loose talk, but also give powerful tools for
> automated reasoning about consistency.   Getting past this merciless editor gives one confidence, or even a certification, that one is not being self-contradictory.

On 10/3/21 2:43 AM, David Eric Smith wrote:
> ... and when they got comfortable that they had a constructive language whose propositions would carry some weight and not break into inconsistencies, they stopped protesting against taking limits.  So one could dig back into all that laborious history, which
> ... Then we can go round and round about the axiom of choice and so forth, versus Voevodsky and univalent foundations, or Brouwer and intuitionism.  There were a few turns of that wheel of samsara here a few months ago, but I think people ran out of things to comment on and drifted away.
> 
> ... and still be coherent.  
> 
> ... there is no “objective Bayesianism”.  ... then chooses however one will.  The point is not to ask God to save you from making a choice.  The point is to acknowledge and embrace that you will make a choice, and then accept that all the consequences of it are yours as well.



-- 
"Better to be slapped with the truth than kissed with a lie."
☤>$ uǝlƃ



More information about the Friam mailing list