A simplification of an argument by Josiah Royce

1.) Let the reasonable, whether intellectual or moral, be defined relative to an ideal knower or reasonable person. Thesis: this ideal knower must actually exist.

2.) Objections: (a) Ideal knowers are essentially hypothetical. (b) Ideal knowers are principles of knowledge and not principles of existence. (c) Ideal knowers are counterfactual and so their real existence need not even be possible.

3.) Respond: Ideal constructions in the physical world differ from those in the cognitive world. Physical ideals are either (i) facts about the world that make no difference in our measurements: assuming the earth is flat when making a mid-range artillery shot; assuming a test particle has no mass. This is not purely arbitrary or subjective: Measurement requires parameters, which allows (requires?) us to ignore what falls outside them. (ii) An attempt to isolate a principle which may never (or even can never) really exist in isolation. A perfectly smooth road illustrates purely inertial motion, even if no such motion could ever exist.  These cases generalize: in the first something that really exists is unnecessary to our knowledge of it; in the second our knowledge of something does not require that it really exist.

But things that exist in the cognitive realm cannot always be subsumed under this general division, since their being is their being known. For example, error and truth exist in minds or not at all, and so to relate these to an ideal knower requires such a knower to really exist in order for these things to really exist within knowers that are not ideal. But by hypothesis a speculative error is that which is erroneous to a knower with all possible information, and so such a knower actually exists.

4.) All objections relate to idealizations of extra-mental reality, as mentioned in the response.


  1. August 27, 2015 at 8:44 am

    This is not convincing. The essence of the argument is contained in the statement that “error and truth exist in minds or not at all.” But someone who thinks that an ideal knower does not exist will say that when he says that “there are some speculative errors relative to an ideal knower,” he does not intend to say that these errors or the opposite truths actually exist (i.e. in a mind), but that they potentially exist insofar as someone may fall into error or come to know them.

    • August 27, 2015 at 9:34 am

      Such an argument only accounts for potential errors, if we define errors in relation to an ideal knower.

      • August 27, 2015 at 9:48 am

        Someone could say that it is sufficient to define potential errors in relation to an ideal knower that exists in my mind, i.e. not that I am an ideal knower, but that I have an idea of one.

      • August 27, 2015 at 10:24 am

        right, and in so doing he would, by the very account you are giving here, account for potential errors. But we commit actual ones too.

      • September 8, 2015 at 3:33 pm

        If you have sufficiently defined a potential error, that seems to sufficiently define actual ones as well.

        I guess the main problem is that I don’t even know what the argument is supposed to be, because it hasn’t been stated clearly. It was formulated as a thesis, which is not an argument. The closest thing to an argument is the statement that “For example, error and truth exist in minds or not at all, and so to relate these to an ideal knower requires such a knower to really exist in order for these things to really exist within knowers that are not ideal.”

        But this only has one premise and a conclusion, while syllogism requires two premises, and thus the argument does not follow.

    • August 27, 2015 at 9:35 am

      This is actually the first objection Royce deals with, so good instincts.

%d bloggers like this: