WHAT THE ŁUKASIEWICZ AXIOMS MEAN
Daniele Mundici · Journal of Symbolic Logic · 2020
Abstract Let $\to $ be a continuous $\protect \operatorname {\mathrm {[0,1]}}$ -valued function defined on the unit square $\protect \operatorname {\mathrm {[0,1]}}^2$ , having the following properties: (i) $x\to (y\to z)= y\to (x\to z)$ and (ii) $x\to y=1 $ iff $x\leq y$ . Let $ eg x=x\to 0$ . Then the algebra $W=(\protect \operatorname {\mathrm {[0,1]}},1, eg ,\to )$ satisfies the time-honored Łukasiewicz axioms of his infinite-valued calculus. Let $x\to _{\text {\tiny \L }}y=\min (1,1-x+y)$ and $ eg _{\text {\tiny \L }}x=x\to _{\text {\tiny \L }} 0 =1-x.$ Then there is precisely one isomorphism $\phi $ of W onto the standard Wajsberg algebra $W_{\text {\tiny \L }}= (\protect \operatorname {\mathrm {[0,1]}},1, eg _{\text {\tiny \L }},\to _{\text {\tiny \L }})$ . Thus $x\to y= \phi ^{-1}(\min (1,1-\phi (x)+\phi (y)))$ .