October 29, 2018


“Should” statements are tricky. Truth statements not having to do with ends are easier to make true and uncontroversial for any observer. Hypothetical imperatives, where we say “if you want to accomplish X, you should do Y”, are a little bit more difficult, because there may be multiple ways to accomplish X, with other important properties, such that the “should” becomes controversial. Hypothetical imperatives can be weakened to technological truth statements: “Y will work to accomplish X, and will also achieve a, b, and c”.

Unqualified “should” statements, categorical imperatives, are heavier still than hypothetical imperatives. They implicitly claim not only a hypothetical imperative and its technical possibility, they also implicitly make a claim about ends, by assuming a value system and often even a political identity.

A brief note on values systems vs political identities: a value system is what you believe is important. A political identity is who “you” or “we” are. Assuming both a political identity and a value system by saying “we should do X” is philosophically heavy. Once can assume one of these without assuming the other, but categorical imperatives tend to assume both.

This is all to say that to get a should statement “right”, you have to be clear on a good number of complex and controversial things.

But should statements are necessary to life and action, and are thus very useful. Ultimately the point of all other statements is to be premises for unqualified should-statements.

On this project, we eventually want to produce a system of should statements, an ideology, and the technologies, truths, hypothetical imperatives, and value philosophy behind it. We may even want to take a political side on questions of who the political “we” refers to. So should statements are within scope.

We’re not there yet, but we’ve recently started using should statements anyways. This is again because they are so useful, especially for thought that is not yet fully conceptualized, or where we have large systems of premises that we have not yet written down.

We’ll try to get it all out in nicely documented form, but until then we’ll still be using should statements occasionally. Read into them what you will, disagree freely, and consider them to be at somewhat reduced epistemic status.

Edited and curated by Wolf Tivy

Comments? Email comments@newstatecraft.org