This is an issue everyone struggles with. If they do not, then clearly they are not trying hard enough.

>If I cannot perceive a thing, how can I know with certainty it exists?

Can you recognize how this relates to Einstein's comment about the moon? Are you saying that something does not exist unless you percieve it? More importantly how can you know with certainty that a thing exists even if you do perceive it?

My initial comment referenced metaphysical uncertainty (the perpetual inability to know if something exists independent of our observation), but the issue you have brought up regards epistemic uncertainty, whether our descriptions of a thing are accurate. This resolves, in classical philosophy, to the study of what it means to know. In my philosophy, we go further, and recognize epistemology as the study of meaning, with the meaning of the word "know" being merely a special and essential instance. Long story short, your question hinges on whether you (we) can know anything "with certainty".

To be direct, the answer is this: there is only one thing you (we) can know with certainty. How that one thing is expressed can vary. The Socratic approach is to say the only thing you can be certain of is that you cannot be certain of anything. The Cartesian approach, which I prefer, is cogito ergo sum: you can be certain you exist (without any additional characterization of that existence being implied), but everything after that is supposition and conjecture. In practical terms, we must ignore this necessary uncertainty (both metaphysic and epistemic) and make due with reasonable degrees of certainty, but the nature of philosophy is that it cannot be restricted to practical value.

Nevertheless, all we really need to do, all we can do, either philosophically or practically, is recognize that, although everything beyond that one thing is uncertain, that does not mean everything else is equally uncertain. So rather than spend all of our time "navel-gazing" and wondering what is certain, we instead consider how certain we are about specific and particular things, rather than the abstract general category "something".

As long as we do not confabulate metaphysical (existential) uncertainty and epistemic (intellectual) uncertainty, we can, in truth, leverage the absolute nature of our certainty about that one thing in order to examine and explore the various degrees of certainty which we need in order to understand the world and determine our conjectures with productive accuracy. When epistemic uncertainty prevents us from proceeding, because what we believe about a thing relates to how we experience it, we can rely on metaphysical certainty (something exists, and existence must mean objective physical existence) to ascertain what aspects of it are not related to our experience of it. When metaphysical uncertainty prevents us from proceeding, because we can only be aware of even the most objective things through our 'subjective' perceptions, we can rely on epistemic certainty by defining things quantitatively rather than experentially. By using this "latching bootstrap" mechanism, we can elevate our knowledge from the deeply profound uncertainty (brain in a jar, butterfly dream, insane absurdity) to the scientific certainty and emotional sincerity that enables us to be conscious, self-determining human beings. It has always been so, this methodology is not a recent development; it is simply the process of reasoning which humans have relied on since the moment we stopped being merely apes. It is just that recently, thanks to the double whammy of metaphysical knowledge of consciousness (as derived from the physical existence of our brains rather than supernatural entities) from Darwin's discovery of natural selection which opened the postmodern age and the epistemic knowledge of quantum mechanics (and the fact that even the most deterministic aspects of physical existence derive from probabilistic causation) which this conversation is focused on, conscious and conscientous examination of the process of knowing and reasoning and being have become common, and fodder for existential angst, rather than just what our brains and minds do innately without thinking about the thinking we think we're doing.

>Does something have to be verifiable to be true? Not sure if these questions make sense…

I understood them and consider them deep and profound. Whether that means they make sense is a whole other thing. 😉

No, something does not have to be verifiable to be true. But, if you are reasonable and intelligent and wish to be well-informed, a thing has to be verifiable for you to be certain it is true. This relates to your epistemic knowledge of its truth, not the metaphysical fact of its truth. This leaves open the question of just what sort of verification you are looking for and hoping to find. Just because epistemic and metaphysical uncertainty are unavoidable does not mean they should not both be minimized. But (if I can be forgiven for bringing the conversation back full circle to the matter of quantum uncertainty) it seems likely that they cannot both be minimalized simultaneously. In a very real way, being both metaphysically and epistemically certain of whether something is true is like measuring both the location and momentum of a quantum particle to an arbitrary level of accuracy. The principle of uncertainty in physics cannot be ignored simply because it is inconvenient or difficult to grasp, and the analogy to philosophical certainty seems almost too on-the-nose, but I don't believe it should be lightly dismissed based on that circumstance.

Thanks for your time. Hope it helps.