Seems Correct to Me
We get an independent standard only if we bring in continuous interaction among individuals, that is, if we bring in the community.
Again from Knowledge by Agreement.
Why should we think it impossible for an isolated individual to do the following: introduce a new category by picking an array of exemplars; apply the category to new cases on the basis of similarity judgements relative to the array; and monitor her own performance for correctness in the light of the exemplars? Would not the exemplars provide the isolated individual with an independent standard? Would not the exemplars enable her to distinguish between 'is correct' (i.e. is indeed highly similar to an exemplar) and 'only seems correct' (i.e. only seems similar to an exemplar for a while)?
This is, in a sense, the holy grail of individualist epistemology. Given the right facilitation, a student 'discovers' her own concept of, say, adding fractions, based on some examples and example situations. She can then apply this concept to similar-seeming examples and situations and monitor herself as to whether or not she gets each new application of the concept right. Can we not grant, at this point, that the student has created new knowledge—a new independent standard?
The envisaged scenario is undermined by two considerations: similarity is not identity; and arrays 'drift'. That is to say, the imagined situation would be a plausible one if the application to new cases involved judgements of identity rather than judgements of similarity, and if the judgements referred back to a fixed and unchanging array of exemplars. In such a scenario there would be a clear 'fact of the matter' as to whether a given application is correct or incorrect, and such a fact of the matter would not involve a community.
Thus, if what the student discovers is simply how to add N example pairs of fractions (N being necessarily very small since individuals are constrained by time and space) and applies that skill only to pairs within that set (identity), then she can of course monitor her own performance in applying that skill—i.e., she can safely make is-correct judgements without talking to anyone else.
However, similarity and drift are central and not eliminable. And since they are not eliminable, the individual does not have the resources to monitor her own performances in light of an independent standard. All the isolated individual can go on are her own performances and her own dispositions. But this 'standard' does not stand still. It is changed by whatever the individual decides to do. And thus it is no independent standard at all. It cannot sustain a distinction between 'correct' and 'seems correct'.
As the student starts to apply the adding-fractions skill to similar-seeming situations outside of her original example set (similarity), this changes the standard she uses to make similarity judgements over time (drift), because the example set changes. But if the student alone is changing the standard, then it is impossible for her to distinguish between 'seems-correct-to-me' and 'is-correct'. Which means one of two things will happen:
Her clock will eventually diverge enough such that her skill will become too narrow, too broad, or just plain wrong (according to an actual independent standard).
She will maintain a correct application of the skill over time (according to an independent standard) and she will never, ever know it.
We get an independent standard only if we bring in continuous interaction among individuals, that is, if we bring in the community: Interaction with others does not just determine exemplars . . . it is also needed to confirm, or deny, that the 'drift' in an individual's array of exemplars goes into the 'right' direction, and usher in 'correct' judgements.
