XPost: sci.logic, sci.math   
   From: janburse@fastmail.fm   
      
   Hi,   
      
    > Hm. You mention "Horn clause", which is like a closure   
    > or completion, it's basically a stroke, and then for inference's   
    > sake it seems "I've written Horn clause, done, nyah". Yet, courtesy   
      
   A Horn Clause, named after Alfred Horn (February 17,   
   1918 – April 16, 2001), a American mathematician,   
   looks like this:   
      
   H :- B ,   
      
   With :- the left pointing conditional.   
   The example I gave has the same form:   
      
   ''(x) :- t_a(x,y)   
      
   Respectively fully quantified:   
      
   ∀x∀y(''(x) ← t_a(x,y))   
      
   If you take the so called Clark Completion, Keith Leonard   
   Clark (born 29 March 1943) a British computer scientist.   
   the conditional is replaced by a biconditional:   
      
   ∀x(''(x) ↔ ∃y t_a(x,y))   
      
   To form the Clark Completion one has to go   
   to FOL with equality, and do some movements,   
   like move some forall quantifiers inside,   
      
   then then change the polarity and become exists quantifiers.   
      
   Bye   
      
   See also here:   
      
   [Clark, 1978] Keith Clark. Negation as failure. In Herve   
   Gallaire and Jack Minker, editors, Logic and Data Bases,   
   pages 293 322. Plenum Press, New York, 1978.   
      
   Ross Finlayson schrieb:   
   > On 11/08/2025 12:27 PM, Mild Shock wrote:   
   >> Hi,   
   >>   
   >> Lets say we have an ost term t_A for   
   >> some sets of pairs such that:   
   >>   
   >>    t_A(x,y) = tt <=> A(x,y)   
   >>   
   >> Question is what is the term t_B for:   
   >>   
   >>    B(x) <=> ∃y A(x,y)   
   >>   
   >> In the Arrow Functions to Horn Clause   
   >> translation. The existential quantifier   
   >> is a feature of the Clark Completion.   
   >>   
   >> In terms of Cabezas notion:   
   >>   
   >>    t_B = { ''(x) :- t_a(x,y) }   
   >>   
   >> Bye   
   >>   
   >> P.S.: Why does it remind me of the   
   >> K Combinator? Well we have:   
   >>   
   >> ∃y t_B(K(x,y)) = ∃y t_A(x,y)   
   >>   
   >> Not sure whether this is useful.   
   >> Although the above is true because the   
   >> combinator K is defined as Kxy = x,   
   >>   
   >> it can be quite misleading, since   
   >> this here does not necessarely hold:   
   >>   
   >> /* Not necessarely */   
   >> { y | t_B(K(x,y)) } = { y | t_A(x,y) }   
   >>   
   >> So if Feferman had the empty set, he could   
   >> also check for inhabitation, and bootstrap   
   >> existential quantifier via parameterized bags:   
   >>   
   >> t_B(x) = ( { y | t_A(x,y) } =/= {} )   
   >>   
   >> But we don't like bags here..   
   >>   
   >> Mild Shock schrieb:   
   >>> Hi,   
   >>>   
   >>> Now this is an interesting find. It seems   
   >>> not only the Verse Calculus by Peyton Jones   
   >>> hit a wall with existential quantifier ∃.   
   >>>   
   >>> Especially the type free case. Its like in   
   >>> Rossy Boys Russell thing, people are not   
   >>> anymore trained to think about "individuals",   
   >>>   
   >>> the are more bothered by "bags", because this   
   >>> is what the Antinomies of the formal revolution   
   >>> tought us. But the formal revolution has also   
   >>>   
   >>> some nice easter eggs, like Fefermans OST   
   >>> ("Operational Set Theory"), an early form of   
   >>> Predicte Abstraction. With each formula A is   
   >>>   
   >>> associated a term t_A such that:   
   >>>   
   >>>      ∀x[A(x) <=> t_A(x) = tt]   
   >>>   
   >>> https://math.stanford.edu/~feferman/papers/OperationalST-I.pdf   
   >>>   
   >>> The nice thing about the t_A, its a term,   
   >>> possibly a open or closed term, depending   
   >>> on whether there are parameters, and thats   
   >>>   
   >>> what I am now doing for Arrow Functions, when   
   >>> the Prolog systems compiles 0rReference(P1,..,Pk),   
   >>> its basically a term, an individual, that   
   >>>   
   >>> later gets called by call/n, which makes the   
   >>> translation for individual to proposition.   
   >>>   
   >>> Bye   
   >>>   
   >>> P.S.: But somehow Feferman shyed away from   
   >>> definition the unbounded existential quantifier   
   >>> as a projection, there is a easy geometric   
   >>>   
   >>> intution, and every SQL database can do it.   
   >>> Instead he falls back to some Hilber Epsilon   
   >>> analogue such as:   
   >>>   
   >>> Given A(x) = ∃yB(x, y) and t_B for B(x, y);   
   >>> then we can take t_A = λx.t_Bx(C(λyt_Bxy)),   
   >>> using the general choice operator C.   
   >>>   
   >>> https://math.stanford.edu/~feferman/papers/OperationalST-I.pdf   
   >>>   
   >>> Funny!   
   >>>   
   >>> Mild Shock schrieb:   
   >>>>   
   >>>> In halls of Cambridge, where catnip sways,   
   >>>> Sat pioneers lost in existential haze.   
   >>>> “Here lies a term!” they cried, “both bound and free,   
   >>>> A bag of possibilities, as far as we see.”   
   >>>>   
   >>>> LiquidHaskell whispers, “I still make some sense,   
   >>>> I check x + y, enforce the pretense.   
   >>>> But only 1% — the rest, pure ado,   
   >>>> Existentials and predicates, I haven’t a clue.”   
   >>>>   
   >>>> Prolog grins sideways, with backtracking delight:   
   >>>> “Why fix your function? Let each path take flight!   
   >>>> X and Y and Z — all three may roam,   
   >>>> I’ll find a solution, or many, for home.”   
   >>>>   
   >>>> Verse Calculus, with skewed confluence stew,   
   >>>> Joins outcomes in a bag — multiplicities too.   
   >>>> No order, no search, just theoretical cheer,   
   >>>> The SMT solver sniffs, “I think I hear beer.”   
   >>>>   
   >>>> Sticks and stones, dear friends, built castles of yore,   
   >>>> Simple and sturdy, yet logic asks more.   
   >>>> Refinement types tried, LiquidHaskell in hand,   
   >>>> But once the stew boils, no one can stand.   
   >>>>   
   >>>> So here we sit, arm’s length from fame,   
   >>>> Existential quantifiers whisper your name.   
   >>>> A mockery? Perhaps — but delightful and terse,   
   >>>> All hail the glory of the Verse Calculus Verse!   
   >>>>   
   >>>> Franz Sneijders schrieb:   
   >>>>> Mild Shock wrote:   
   >>>>>   
   >>>>>> We use “∃” to bring a fresh logical variable into scope, because   
   we   
   >>>>>> really mean “there exists an x such that ···.”   
   >>>>>   
   >>>>> idiot, there is no any x over there. And it doesn't need to be a   
   >>>>> variable,   
   >>>>> a constant suffices.   
   >>>>>   
   >>>>   
   >>>   
   >>   
   >   
   > Hm. You mention "Horn clause", which is like a closure   
   > or completion, it's basically a stroke, and then for inference's   
   > sake it seems "I've written Horn clause, done, nyah". Yet, courtesy   
   > the riddle of induction (or the fallacy of induction to be stronger)   
   > another can write another different Horn clause, since "done"   
   > was yet "not yet untrue" while "done, done, done", is a bit more   
   > "not ultimately untrue".   
   >   
   > Then, about quantifier disambiguation, it's usually enough framed   
   > about the universal quantifier, while the existential quantifier   
   > deserves its own disambiguation.   
   >   
   > exists (> 0)   
   > exists-unique (exactly one)   
   > exists-distinct (more than one)   
   > not-anywhere-not-exists (now it's the universal quantifier)   
   >   
   >   
   > Then, the universal quantifier has these sorts of example   
   > with common sorts of considerations about them being   
   > the same and about them being different.   
   >   
   > for-any   
   > for-each   
   > for-every   
   > for-all   
   >   
      
   [continued in next message]   
      
   --- SoupGate-DOS v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)