Okay, how does it work? Good, easy. Right? We give ourselves a language, we have two
kinds of things we're modeling in this world, we have truth values, and I'll sometimes decorate
those with a quote-unquote type Omicron, and we have individuals which I'm
sometimes going to annotate with a type Iota, and then we have a signature which
is essentially the things you can write down in in this logic. We have the
connectives we all know and love from propositional logic, we have function
constants, we have predicate constants, and we have something called Scholem
constants which we'll need later. The function constants and predicate
constants we think of as being finite and of the Scholem constants we give
ourselves an infinite supply and we have variables as well which we think give
ourselves an infinite supply of as well. Okay? So those are the things we're going
to use and then we make terms. Terms are just made variables are terms and if I
have a k-ary function and k terms then I can apply the function f to k terms and
get another term. And if I have a unary function, one argument function, say the
mother function, and I have a zero-ary function Peter, right? Peter is an
individual without applying it to anything so we're then I can have mother
of Peter and that's a term. And of course by the same rule mother of mother of
mother of mother of Peter. Great great grandmother. Okay? So we can also give
ourselves propositions, another kind of objects. I do that by if I have a k-ary
predicate and apply it to k terms I'm going to get an atomic proposition and
if I have propositions then I can make new propositions by applying connectives
to them true as a composition and I can make a new given a variable X I can make
a new proposition for all XA. Okay? Now that's a very traditional way of
writing this down. What you want to think of this as a computer science is really
think of this as a grammar. Think of this as a grammar where we say well what are
the terms just the context free grammar well they're variables and a k-ary
functions. Yeah let's call this t and this n should be a k. Okay? So terms that
that's really these two productions are just these two rules. And then we have
formulae which is essentially true is a formula P of t1 to tk for a k-ary thing
and what else? A and A1, A2 or not A2 or for all XA. It's a good thing it's so
easy otherwise you wouldn't be able to see what I have read. Okay? That's a
grammar of a formal language and that's exactly what I'm doing here. And in a way
we have propositional logic as a sub language. And instead of propositional
variables which are black box propositions we have these atomic
formulae. And the only thing that's really really interesting and new is
this. There. This is new interesting for all XA where A is any kind of a
proposition which allows us to write down things like for all X even of X if
and only if even of successor of successor of X. Right? Why? Because even as
a unary predicate which I can apply to one term and variables are terms so this
is actually this is actually of type omicron. I have another one of those X is
a variable so S of X is the application of one a unary function to a term is a
term. I can do it again so S of S of X is a term. I can apply the unary predicate
I can apply the unary predicate even to it. I have another proposition here. Two
proposition joined by if and only if gives me a proposition for all X in
front of it. Wow there it is. Okay? That's exactly what we do with this kind of a
grammar. Okay. Syntax done. Oh yes and you can see I did it again. I just said well
I have true and or and for all which is kind of a minimal set of logical
constants but of course I want much more. I want to have or which I abbreviate in
the usual way implies and so on but also I want to have an existential
quantifier there is an XA and I take that to be an abbreviation of it is not
Presenters
Zugänglich über
Offener Zugang
Dauer
00:17:09 Min
Aufnahmedatum
2020-11-27
Hochgeladen am
2020-11-27 13:28:43
Sprache
en-US
Explanation of syntax, semantics and free/bound variables in First-Order Logic.