The fact that we cannot see something does not mean that these things doesn’t exist. For example, we cannot see infrared light, but electromagnetic waves of that wavelength really do exist. Moreover, current technology can actually see atoms. But if we are talking about the “atom” in ancient Greek atomism, then it was indeed a purely speculative product.

They really exist, but our descriptions of them are models. Humans cannot directly see electrons, but we can capture them one by one, measure their charge and mass, manipulate them in semiconductors, and so on. Electrons must be real—otherwise we could not have built transistors, MRI machines, or particle accelerators, and yet all of these work exactly as predicted. The idea of electrons as little balls is a simplified model, but the entity itself exists independently of human consciousness. As science improves its models, it does not give up on the reality of the entities those models describe.

They should be models. We cannot directly see atoms or electrons. Humans use these concepts only because they can explain experimental results and predict phenomena. They are effective mental models for describing observations. If a better model appears in the future, the current description may also be revised.

I personally follow Roy Bhaskar’s view: precisely those things that cannot be directly experienced and can only be constructed by theory are the ones that are “real”. Obviously, “real” here is narrower than ordinary usage, but it also fits the long Platonic tradition that mathematical things are real. For things that can be directly experienced, the subjective, inner side can be called “empirical”, while what is regarded as objective knowledge can be called “actual”; neither of these is a mere theoretical construction, but has one or billions of subjects as experiencers.

We all actually know they exist; humans just call them by these names. Another civilization might call an electron “Lijiazi” or “Nan Xiaoniao” (just making up names). With our current understanding, we call them electrons.

Strong experimental evidence gives us good reasons to believe atoms and electrons are real. However, our models of them may be incomplete and should not be confused with reality itself.

I feel that things like electrons and atoms can account for most studies and facilitate our exploration of natural sciences, which means they are already good enough. But still, I think there is somehow enough evidence to believe in the existence of electrons and atoms. Think of it this way: if we assume they do not exist, what objects can explain all those observations? I think the answer is: the thing that carries the same properties as electrons and atoms—so why deny their existence? However, it is also true that this indirect proof is not 100% solid. As humans, most of the time we are prone to trust our eyes rather than logic.