[Borrows heavily from Wittgenstein, if anyone cares]
[In philosophical posts, I often use double quotes to emphasize a word or concept, and single quotes to emphasize the symbol we use for such a concept.]
First, let’s outline what “solipsism” means to me. Roughly, it is the idea that only the self is able to be appealed to for knowledge: “I am the only one who can know if I am right.” If you consider most applications of solipsism, you can see that this is indeed a generalization of it proper for the context at hand.
Problems I have with this stem from its non-trivial application. Its trivial application (i.e. its general assertion) is indisputable just like any other “foundation” which stands without judgment.
Initial remark: the expression of solipsism causes problems for any solipsist. That is, its formulation, discussed by multiple people, collapses quickly. If only I could know, how could I assert solipsism in meaninful terms? The problem being, of course: how could anyone understand? – What is there to discuss? The solipsist here must retreat to the trivial application, in this case, that “Only I can know what I really mean.” But if mere assertion is the grounds we deal with, than an assertion of any alternative in return is the only satisfactory way of dealing with it and we may consider the matter closed.
It is in this where I find the biggest divide with solipsism: meaning is not a private phenomenon. I find the PLA referenced earlier to be the essential motivation for this based on what is initially an empirical approach. “What does it take for a word, phrase, action, etc. to have meaning?” Not in a philosophical sense, but in a “time before I took my first philosophy course” sense… before someone found out what solipsism was all about and why it was such a trump card (and why it didn’t amount to much as a trump card). A time before philosophical investigations when we learned to speak the language we will eventually use to discuss philosophy.
That is, for example, how did I learn what “solipsism” is? --It is in such an investigation that we can find the escape, generally speaking. Practically speaking, the only escape from assertion is another assertion, so one could suggest simply that the only way to avoid solipsism is to assert otherwise. I have no real problem with that. But it is not very motivating.
However, it is not just that meaning is a social phenomenon, for not everyone will grant that right off. What else we may consider is knowledge, mistake, certainty, and doubt. One thing people often take for granted, especially those very keen in the natural sciences, is that skepticism should be the initial approach (e.g., “extraordinary claims require extraordinary evidence”). However, further reflection reveals that this is not the whole story (rhetorical point: consider, for example, a definition of an “extraordinary” claim). So to, does doubt require a framework, a sense, such that when someone doubts something, it is clear what exactly they are doubting, and what the remedy is for that doubt. Doubt, like knowledge, requires a framework for sensibility. A test for some kinds of knowledge, for example, might require a demonstration (say, understanding the rules of addition). Doubting, in such cases, indicates not just a lack of certainty but includes with it “a means of satisfying”; that is, where I cannot be[come] certain, I cannot doubt. Where there is no hope for certainty, it makes no sense to doubt; and where there is no doubt, certainty also loses its sense.
I wish to draw attention, specifically, to the phenomena that only have meaning in relation to others. That is, in this thread, knowledge and mistake, and certainty and doubt. Knowledge is generally recognized by means of some demonstration, given some criteria, that tries to exclude the possibility of mistake (and this is “accidentally getting it right” a la Gettier as well as “getting it wrong” i.e. “being mistaken”). This is not generally possible, however (a separate discussion), and so we feel certainty with regards to knowledge, such that we no longer doubt the existence of a mistake. I emphasize that to indicate the role these four terms play with respect to each other most plainly.
Stated in analytic-speak for those so minded: one can be certain if and only if one can doubt; one can have knowledge if and only if it is possible or “makes sense” to make a mistake. The means of satisfying one’s doubts entail the means of being or becoming certain (or, if you are keen on overqualifying such things, “more” certain [than one already is]). Doubt, as such, requires a ground, and that ground contains the means of assuading it (though it doesn’t promise anyone can meet the standard, that is another argument). Where one cannot be mistaken, one can neither be in the know; where one cannot find any way to doubt, one cannot be certain. Certainty assures the absence of doubt (not its meaninglessness), and knowledge assures the absence of mistake (but not its impossibility). Certainty is not, in practice, the impossibility of doubt, it is its satisfaction (i.e. doubt is alleviated).
Consider, for things you say you know: what would it be like to be mistaken? Have you a clear idea of how someone could point out to you that you were mistaken? Are you aware of how to make a mistake in such things?
Why is this important? To repeat: without doubt, there is no certainty; without the possibility of mistake, there cannot be knowledge. These things stand in contrast to the others. Where I cannot be mistaken, I cannot be said to know, for how would I demonstrate it? --knowing, as opposed to what? Being wrong? Being mistaken? But the solipsist admits he can make no mistake and admits no other standard. And this just means there is no standard, that everything is ‘right’. But then the solipsist does not have knowledge.
Now I do not present this as an escape from solipsism; as stated previously, if it is asserted, it can only be met with assertion. If it is approached non-trivially (if it is explored but not asserted) then it collapses in these kinds of distinctions. Epistemologically, it permits everything because it can forbid nothing. But, as indicated, where one cannot be making a mistake, one cannot be said to know anything. That is just what “to know” means: the possibility of demonstration, and the conditions under which one would be mistaken.
Why does this matter in the current discussion?
Because it is just that ontological matters fall into these “assertions”. I can assert that there is only matter (or physical things), or I can assert that there is only mind, or I can assert that there is both matter and mind, but none of these will in fact represent knowledge because we cannot suggest 1) a means of demonstration that satisfies our curiosity and 2) the conditions that would show us we were mistaken in our belief. It is precisely for these reasons that I feel that such concerns should be 1) selected against, epistemically, and 2) not included in our epistemological investigations.
What impact does this have on epistemology in general?
Firstly, no particular epistemological limits are universal in themselves; they are definitional and procedural, but not immune to revision. This, of course, represents a kind of epistemological relativism, though it needn’t be thought of so drastically as to be subjectivism which, for reasons outlined here and previously, I find impossible to hold up in a non-trivial sense.
Secondly, that we cannot consider any correspondence to “how things really are” as a judge of truth because it is precisely “how things really are” that is outside epistemology. We seek knowledge, yes, and something we might seek is how things are, yes, but this is a definitional and procedural process which must admit the possibility of mistake, and so must admit the ability of revision, and so cannot be asserted as universally valid. This is not, of course, a weakening of the verb “to know” which, much to everyone’s discontent, has never allowed P to be inferred from “I know P” in the first place, so we are not leaving out anything we had, only acknowledging what we do not have in the first place. Epistemology, being in part a recognition of the limits of human knowledge, should generally point such things out.
Thirdly, and perhaps most importantly, it allows for the possibility of disagreement about what limits are precisely and what procedural method best matches the possibilities of knowledge in the first place without the arrogance, IMO, that traditional a priori accounts yield.
Epistemology is the study of the nature and limits of human knowledge. Those who fail to recognize those limits do not have what I consider knowledge. Thankfully, I do not feel that I am alone in that, so there is no need for what I am calling trivial solipsism.
…
…But I am still, like, totally right about all this. 
