<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<item>
  <id>05609396</id>
  <dt>j</dt>
  <an>05609396</an>
  <augroup>
    <au>Baldwin, John T.</au>
    <au>Kolesnikov, Alexei</au>
    <au>Shelah, Saharon</au>
  </augroup>
  <ti>The amalgamation spectrum.</ti>
  <so>J. Symb. Log. 74, No. 3, 914-928 (2009).</so>
  <py>2009</py>
  <pu>Association for Symbolic Logic, Poughkeepsie, NY</pu>
  <lagroup>
    <la>EN</la>
  </lagroup>
  <ccgroup>
  </ccgroup>
  <utgroup>
    <ut>disjoint amalgamation property</ut>
    <ut>size of models</ut>
  </utgroup>
  <cigroup>
  </cigroup>
  <ligroup>
    <li>doi:10.2178/jsl/1245158091</li>
  </ligroup>
  <abgroup>
    <ab>Summary: We study when classes can have the disjoint amalgamation property for a proper initial segment of cardinals.  Theorem A. For every natural number $k$, there is a class ${\bold K}_k$ defined by a sentence in $L_{\omega_1,\omega}$ that has no models of cardinality greater than $\beth_{k+1}$, but ${\bold K}_k$ has the disjoint amalgamation property on models of cardinality less than or equal to $\aleph_{k-3}$ and has models of cardinality $\aleph_{k-1}$.  More strongly, we can have disjoint amalgamation up to $\aleph_\alpha$ for $\alpha<\omega_1$, but have a bound on the size of models.  Theorem B. For every countable ordinal $\alpha$, there is a class ${\bold K_\alpha}$ defined by a sentence in $L_{\omega_1,\omega}$ that has no models of cardinality greater than $\beth_{\omega_1}$, but $\bold K$ does have the disjoint amalgamation property on models of cardinality less than or equal to $\aleph_\alpha$.  Finally we show that we can extend the $\aleph_\alpha$ to $\beth_\alpha$ in the second theorem consistently with ZFC and while having $\aleph_i\ll\beth_i$ for $0 < i\le \alpha$.  Similar results hold for arbitrary ordinals $\alpha$ with $|\alpha|=\kappa$ and $L_{\kappa^+,\omega}$.</ab>
    <rv></rv>
  </abgroup>
</item>