We have to define science first –let’s say it is the body of knowledge formed through the scientific method of testing hypotheses against verifiable empirical data.

It becomes clear that there are a range of questions science cannot answer. Science cannot answer the question of “what is science?” because the body of knowledge that composes science has boundaries that are dependent on already having an accepted definition of science. Science also cannot answer the question of “what questions can science not answer?” since that would require the ability to prove a negative, which cannot be done through empirical testing.

It also seems clear that science is ill-equipped to consider questions about unique phenomena –events or entities that cannot be duplicated, since the replicability of results is a key component of empirical verification. Science is also maladapted for the consideration of transcendent entities –i.e. God, if we assume God to be unbound by scientific laws.

It’s important, however, to keep in mind three things:

• First, the limitations of science do not mean science is not a valuable and essential tool of understanding.
• Second (and conversely), science’s inability to consider certain questions does not necessarily mean that the things considered by those questions are unreal, unimportant, or even unknowable (through other sources of knowledge).
• Third, there are almost certainly things that do not currently seem empirically verifiable that may someday reveal an empirical footprint, and thus pass into the realm of what can be considered by science.

For an example of the last, consider the phenomenon of invisible electromagnetic radiation such as radio waves and UV light. At one point in history these rays would have seemed speculative, metaphysical, unverifiable, and perhaps even absurd, yet now their existence and their effects are uncontroversially accepted as belonging to the realm or science fact –because the advance of scientific equipment has made it possible to test them in ways that are verifiable and repeatable.