%! TEX root = MT.tex % vim: tw=50 % 23/01/2025 11AM % Contact: c.kestner@imperial.ac.uk \newpage \subsubsection*{Introduction} \begin{example*} $(\Zbb, +)$ is \emph{not} the same as $(\Zbb, +, \cdot)$. $(\Zbb, +)$ is \emph{decidable} (there exists an algorithm to decide whether a given sentence is true or false in the model). However, $(\Zbb, +, \cdot)$ is not decidable (G\"odel's completeness theorem). \end{example*} \begin{example*} \phantom{} \begin{enumerate}[(1)] \item $(\Cbb, +, 0, 1)$ \item $(V, +, f_r)$ (where $V$ is an $\Rbb$-vector space, and $f_r$ is the map $r : v \mapsto rv$ for any $r \in \Rbb$). \end{enumerate} These structures are both \emph{strongly minimal}. \end{example*} \begin{definition*}[Strongly minimal] \glsadjdefn{sm}{strongly minimal}{theory}% A theory is \emph{strongly minimal} if all formulas in one variable are either finite or co-finite. \end{definition*} For the $\Cbb$ example: formulas in one variable are polynomial equations or inequations, so solution set is always either finite or cofinite (recall Fundamental Theorem of Algebra). For the vector space example: the formulas in one variable are of the form $a \mathbf{x} = \mathbf{b}$ or $a \mathbf{x} \neq \mathbf{b}$. Cheats: \begin{itemize} \item Boolean combinations and quantifiers: \[ \exists y axy^2 + y + x^2 + 3 = 0 .\] Need \emph{quantifier elimination} (boolean combinations are easy to deal with). \item Elementary extensions (chapter 1). \end{itemize} Interestingly: strongly minimal structures all carry notion of dimensions. For example: \begin{itemize} \item In $(\Cbb, +, \cdot, 0, 1)$ this is transcendence degree. \item In $(V, +, f_r)_{r \in \Rbb}$, this is linear dimension. \end{itemize} If interested in further reading: see \begin{center} \url{https://forkinganddividing.com/} \end{center} \setcounter{section}{-1} \newpage \section{Review of First Order Logic} \subsection{Languages} \[ \ub{\mathcal{L}}_{\text{language}} = \ub{\mathcal{F}}_{\text{function symbols}} \cup \ub{\mathcal{R}}_{\text{relation symbols}} \cup \ub{\mathcal{C}}_{\text{constant symbols}} .\] \begin{example*} \phantom{} \begin{itemize} \item $\mathcal{L}_{\text{group}} = \{*, e\}$, with example sentences $x \cdot x = e$, $\exists y, x \cdot y = e$. \item $\mathcal{L}_{\text{rings}} = \{+, \times, 0, 1\}$: $x^2 + x + 1 = 0$. \item $\mathcal{L}_{\text{o}} = \{<\}$: $\forall x \forall y ((x < y \wedge y < x) \to x = y)$. \end{itemize} \end{example*} Convention: all languages include $=$. \subsection{Structures} \begin{definition*} Given a language $L$, an $L$-structure is a triple \[ \mathcal{M} = \langle M, \hat{\mathcal{F}}, \hat{\mathcal{R}}, \hat{\mathcal{C}} \rangle .\] $M$ is an underlying set. Convention: $M \neq \emptyset$. $\hat{\mathcal{F}}$: for every $n$-ary $f \in \mathcal{F}$ we have $\hat{f} \in \hat{\mathcal{F}}$ a function $\hat{f} : m^n \to m$. $\hat{\mathcal{R}}$: for every $n$-ary $R \in \mathcal{R}$, we have $\hat{R} \in \hat{\mathcal{R}}$, which is a subset of $M^n$. $\hat{\mathcal{C}}$: for every $c \in \mathcal{C}$, we have $\hat{c} \in \hat{\mathcal{C}}$, with $\hat{c} \in M$. \end{definition*} \begin{itemize} \item $\langle \Cbb, +, 0\rangle$ and $\langle \Cbb, \times, 1\rangle$ are both $\mathcal{L}_{\text{group}}$-structures. \item $\langle\Qbb, <\rangle$ and $\langle\Qbb, x + y = 3\rangle$ are both $\mathcal{L}_{\text{o}}$-structures. \end{itemize} \subsection{Formulas / sentences} \begin{itemize} \item \emph{Terms}: made of variables, constant symbols and function symbols in a `sensible way' \[ x + yx + 1 + 1 \qquad \cancel{\cdot + x \mid \cdot \cdot} .\] \item \emph{Atomic formulas}: Plugging terms into one relation symbol \[ x + yx + 1 + 1 = 0 \qquad \cancel{== + 1 \cdot 0} \] \item \emph{Formulas}: \begin{itemize} \item Boolean combinations ($\neg, \wedge, \vee, \to, \leftrightarrow$) \item Quantifiers ($\exists, \forall$) \end{itemize} in a `sensible way': \[ \exists y (x + yx + 1 + 1 = 0 \vee x = 1) \qquad \cancel{\forall x + \wedge 0} .\] A formula with $n$ free variables defines a subset of $M^n$. \end{itemize} \begin{example*} $\mathcal{L}_{\text{o}}$-structure $\langle \Qbb, <\rangle$. \begin{center} \includegraphics[width=0.6\linewidth]{images/ca9ff7cab94a46ee.png} \end{center} \end{example*} Formulas with no free variables are called sentences. In an $\mathcal{L}$-structure $M$, these are either: \begin{itemize} \item True: $\mathcal{M} \models \sigma$ \item False: $\mathcal{M} \not\models \sigma$ \end{itemize} In formula $\phi(\ol{x})$ with free variables, we can plug a tuple $\ul{a} \in M^n$. We say $M$ satisfies $\phi(\ul{x})$ at $\ul{a}$, and we write $\mathcal{M} \models \phi(\ul{a})$ (models / satisfies) if $\phi(\ul{a})$ is true in $\mathcal{M}$. \begin{definition*} A set of sentences $\Sigma$ is satisfiable in $\mathcal{M}$ if for all $\sigma \in \Sigma$, $\mathcal{M} \models \sigma$. \end{definition*} \begin{theorem*}[Compactness Theorem] Let $\Sigma$ be a set of $\mathcal{L}$-sentences. $\Sigma$ is satisfiable if and only if every finite subset of $\Sigma$ is satisfiable. \end{theorem*} ($\Sigma$ is satisfiable if there is an $\mathcal{L}$-structure $\mathcal{M}$ such that $\Sigma$ is satisfiable in $\mathcal{M}$) \begin{corollary*}[Upward L\"owenheim Skolem] Any theory that has either: \begin{itemize} \item arbitrary large finite models \item at least one infinite model \end{itemize} has arbitrarily large models. \end{corollary*}