Partially Examined Life

On "Universality and Truth" and "Pan-Relationalism," which are lectures 3-5 in Richard Rorty's Pragmatism As Anti-Authoritarianism.
How do we justify democracy? Rorty says we don't have to refer to transcendent Truth or Good to do this. He also denies the disinction between essential and accidental properties, and in fact between substance and property: Everything is just described in terms of its relations to other things, and which relations are important are not intrinsic to the thing, but a matter of a speaker's purposes.
This discussion concludes with • Partially Examined Lif... . Our treatment of the book starts at • Partially Examined Lif... . Get more at partiallyexaminedlife.com. Visit partiallyexaminedlife.com/support to get ad-free episodes and bonus content.

[@peterhalick6226]

Regarding 17 no description is more revealing just more useful for a given purpose. There may be a description that is most useful generally speaking but not in an absolute sense. That seems to be the point Rorty makes and I think even a good pragmatic mathematician would agree.

[@peterhalick6226]

So if your interested in knowing than the H2O picture is better. And knowing is more basic or fundamental than other human competencies. And not just knowing but knowing the truth. If those are fair summaries then you have given priority to a scientific language. It is somehow a higher or maybe truer type of knowing. Rorty is saying that there is no forced, natural ranking that puts the scientific description above all others. The scientific image is wonderful for the purposes of prediction and control of natural phenomena but your love of knowing is an optional, private part of the endeavor. It doesn’t matter that you’re trying to figure out what water really is. Reality is in the eye of the beholder but good physical descriptions come through hard work, careful observation and checking your findings. All the things a good scientist or any investigator does habitually.

[@WizzKidxKOx]

You would probably find the manifold hypothesis for machine learning interesting as when AI attempts to learn word correlations to determine meaning you can think of it like matching a height map to a set of data points in 3D, so that if you ask for something the model doesn't have data for, like a squirrel chicken, it can make a prediction based on squirrel and chicken contextually "accurate" to some midpoint between the high dimensional vector space of what those things are. Welch labs has a video on AI scaling and goes over it at about 15 minutes.
So the answer to what is the number four would be whatever our data said it was or statistically predicts.
Hopefully AI can learn to reason from prop logic rather than raw data memorization.

[@frederickanderson1860]

Nothing wrong with authority its the abuses of power in all realms in society.