Sir Morley Steven wrote:A virus that kills or hospitalises its host becomes less infectious due coming into contact with fewer people. Its personal R rate will therefore be low.
A virus that gives you mild or no symptoms is more infectious (or has a greater R rate of its own) due coming into contact with more people.
Thus, variants of the latter type will be less serious?
In time viruses will, by this logic, become less dangerous.
Is that flawed logic?
Yes, it is flawed logic.
It is true that a virus that has a short incubation period, and causes symptoms so bad that its victims can't get around will have a low R value. This (and the fact it's not airborne) is why Ebola outbreaks tend to die out.
It's also true that a mild virus can have a large R value because its vectors are out and about spreading it. The common cold, for instance.
However, there is no causal relationship between the two. This is because the mutations that give rise to new variants are random. There is no 'guiding force' behind such mutations with any kind of objective. If it so happens that a new variant is more transmissible than another, it will propagate more effectively. This is what happened with Delta, and is probably now happening with Omicron.
Whether new variants result in different symptoms is completely separate, and as others have said, if the host dies it's of no evolutionary significance if the virus has already infected other hosts in the meantime.
Smallpox was around for 3000 years without getting milder. HIV is still lethal without the anti-retroviral drugs that can treat it. Flu is sometimes mild, and sometimes not.
We absolutely cannot rely on Covid becoming milder. Even if it did, there's no reason yet another variant wouldn't arise that was more serious again. The original SARS was a lot more lethal than the current one, and MERS was a lot more lethal still. We were just lucky that they weren't very transmissible.