Strong failures of higher analogs of Hindman’s Theorem

Joint work with David J. Fernández Bretón.

Abstract.  We show that various analogs of Hindman’s Theorem fail in a strong sense
when one attempts to obtain uncountable monochromatic sets:

  • Theorem 1. There exists a colouring $c:\mathbb R\rightarrow\mathbb Q$, such that for every $X\subseteq\mathbb R$ with $|X|=|\mathbb R|$, and every colour $\gamma\in\mathbb Q$, there are two distinct elements $x_0,x_1$ of $X$ for which $c(x_0+x_1)=\gamma$. This forms a simultaneous generalization of a theorem of Hindman, Leader and Strauss and a theorem of Galvin and Shelah.
  • Theorem 2. For every commutative cancellative semigroup $G$, there exists a colouring $c:G\rightarrow\mathbb Q$ such that for every uncountable $X\subseteq G$, and every colour $\gamma$, for some large enough integer $n$, there are pairwise distinct elements $x_0,\ldots,x_n$ of $X$ such that $c(x_0+\cdots+x_n)=\gamma$. In addition, it is consistent that the preceding statement remains valid even after enlarging the set of colours from $\mathbb Q$ to $\mathbb R$.
  • Theorem 3. Let $\circledast_\kappa$ assert that for every commutative cancellative semigroup $G$ of cardinality $\kappa$, there exists a colouring $c:G\rightarrow G$ such that for every positive integer $n$, every  $X_0,\ldots,X_n \in[G]^\kappa$, and every $\gamma\in G$, there are $x_0\in X_0,\ldots, x_n\in X_n$ such that $c(x_0+\cdots+x_n)=\gamma$. Then $\circledast_\kappa$ holds for unboundedly many uncountable cardinals $\kappa$, and it is consistent that $\circledast_\kappa$ holds for all regular uncountable cardinals $\kappa$.


Citation information:

D. J. Fernandez Breton and A. Rinot, Strong failures of higher analogs of Hindman’s Theorem, Trans. Amer. Math. Soc., 369(12): 8939-8966, 2017.

This entry was posted in Partition Relations, Publications and tagged , , , , , , , , , , , , , , . Bookmark the permalink.

One Response to Strong failures of higher analogs of Hindman’s Theorem

  1. saf says:

    Submitted to Transactions of the American Mathematical Society, September 2016.
    Accepted, November 2016.


Leave a Reply

Your email address will not be published. Required fields are marked *