Content deleted Content added
Line 165:
It is instructive (although not exactly useful) to make <math>\psi</math> less powerful.
If we alter the definition of <math>\psi</math> above to omit exponentiation from the repertoire from which <math>C(\alpha)</math> is constructed, then we get <math>\psi(0) = \omega^\omega</math> (as this is the smallest ordinal which cannot be constructed from <math>0</math>, <math>1</math> and <math>\omega</math> using addition and multiplication only), then <math>\psi(1) = \omega^{\omega^2}</math> and similarly <math>\psi(\omega) = \omega^{\omega^\omega}</math>, <math>\psi(\psi(0)) = \omega^{\omega^{\omega^\omega}}</math> until we come to a fixed point which is then our <math>\psi(\Omega) = \varepsilon_0</math>. We then have <math>\psi(\Omega+1) = {\varepsilon_0}^\omega</math> and so on until <math>\psi(\Omega 2) = \varepsilon_1</math>. Since multiplication of <math>\Omega</math>'s is permitted, we can still form <math>\psi(\Omega^2) = \
If we alter the definition of <math>\psi</math> yet some more to allow only addition as a primitive for construction, we get <math>\psi(0) = \omega^2</math> and <math>\psi(1) = \omega^3</math> and so on until <math>\psi(\psi(0)) = \omega^{\omega^2}</math> and still <math>\psi(\Omega) = \varepsilon_0</math>. This time, <math>\psi(\Omega+1) = \varepsilon_0 \omega</math> and so on until <math>\psi(\Omega 2) = \varepsilon_1</math> and similarly <math>\psi(\Omega 3) = \varepsilon_2</math>. But this time we can go no further: since we can only add <math>\Omega</math>'s, the range of our system is <math>\psi(\Omega\omega) = \varepsilon_\omega = \
In both cases, we find that the limitation on the weakened <math>\psi</math> function comes not so much from the operations allowed on the ''countable'' ordinals as on the ''uncountable'' ordinals we allow ourselves to denote.
|