Principled Comparative Method - a new tool

[Richard M. Alderson III]

On 3 Sep 1999, Steve Long wrote:

>And this shows the fundamental uncertainty when we are trying to reconstruct
>from two daughters.  If you assume one has innovated, you assume the other
>reflects the parent.

Let's try another dataset for a moment, with data from three languages:

	/tSi tSa ka  ka    ku tSj/
	/ti  te  pa  po    ku ss/
	/kwi kwe kwa ko    ku kwi/
		     ~ kwo

and	/Si  Sa  Sa  Sa    Su Sj/
	/ki  ke  ka  ko    ku ss/
	/ki  ke  ka  ko    ku kwi/

Now, which of these three languages has innovated, and which has not?  Which
represents the parent?

								Rich Alderson