Don't have to be, but usually are. There's also not a lot of room for evolution--their structures are determined by the properties of the ions that make them up. There's no real chemical reaction involved, the best you can get are deformations due to impurities or different pressures.
In that sense, crystals are a bad analogy. Viruses do interact and do undergo chemical processes. When the RNA and proteins are copied, there can be errors, and that changes the properties of both (random mutations). Sometimes it makes them completely useless and their constituents get removed or recycled. Sometimes it makes them better adapted, and the hosts that they are in live longer (there's one source of the evolution).
Even then, viruses need something living to provide both the mechanisms for mutations and the actual change in frequency in the population.
There's nothing metaphysical about it.
You can run evolution simulations in software, and I've seen examples where drug companies used "evolution" to produce better drugs: They have complex compound with some antibacterial properties, and they shove it into an algorithm that makes a lot of small, random changes or "mutations". Then they analyse its structure and properties against receptors in the bacteria they're targeting. They keep the best candidates, and make more random changes. Mutations + natural selection. They synthesize the top candidates and test them against reality.
They're close to being living--except they lack a lot of the crucial bits. Think of them as mobile stripped down genomes. Everything alive needs a genome; not every genome has to be alive. If we tweaked the definition of "living" then they'd qualify. (It's the same kind of thing as to why Pluto isn't a planet or how many species of Echinofossulocactus their are--you pick your definition and that sets up the categories that things fall in. Some change the categories, i.e., the definitions, when it helps their field become more descriptive or highlight important things, some outside the field want the definitions changed to suit ideas of fairness or salience outside the field.)