An application of infinitary universal algebra to set theory (Q1337156)

In set theory with the axiom of choice (AC) a mere routine proof shows that, for every infinite cardinal \(\kappa\), the class \(\text{HC}_ \kappa\) of all \(\kappa\)-hereditary sets is a set of von Neumann rank \(\varrho(\text{HC}_ \kappa)\leq \kappa^ +\). Moreover, \(\varrho(\text{HC}_ \kappa)= \kappa^ +\) if and only if \(\kappa\) is singular, otherwise \(\varrho(\text{HC}_ \kappa)= \kappa\). In set theory without the axiom of choice a much greater effort is necessary to handle the situation. Here it is not even clear \textit{prima facie} whether all classes \(\text{HC}_ \kappa\) are indeed sets! In the present paper all the statements above about classes \(\text{HC}_ \kappa\) are derived without using AC from a theorem of Universal Algebra dealing with arbitrary ordinal functions on partial Peano algebras. The special case of a rank function on the field of a well-founded extensional \(\kappa\)-narrow relation yields the desired results. A special case of this special case (\(\kappa= \omega_ 1\), i.e., the class \(\text{HC}_{\omega_ 1}\) of all hereditarily countable sets) has been treated independently by \textit{T. Jech} [J. Symb. Log. 47, 43-47 (1982; Zbl 0516.03025)]. The key lemma implied by the above-mentioned algebraic theorem and implying in turn all the theorems about classes \(\text{HC}_ \kappa\) states that the transitive hull \(R^*\) of a well-founded \(\kappa\)-narrow relation \(R\) is \(\kappa^ +\)-narrow (resp. \(\kappa\)-narrow in case \(\kappa\) is an infinite regular cardinal). Here a relation \(R\) is said to be \(\kappa\)-narrow if, for every \(x\), there is no surjective mapping from the class \(\{y\mid y R x\}\) of all \(R\)-predecessors of \(x\) onto the infinite cardinal \(\kappa\). Using the algebraic theorem for the proof of the key lemma naturally restricts its validity to well-founded relations. However, using purely set-theoretical tools, another, though longer proof, for arbitrary relations has been published by the author [Z. Math. Logik Grundlagen Math. 38, 387-398 (1992; Zbl 0798.03051)]. There are some misprints (misspellings and wrong numberings) which are, however, easy to spot. We just mention here that ``Remark 4.5'' and ``Proposition 4.4'' have to be replaced by ``Remark 6.5'' and ``Proposition 6.4''.