If the legal system calls certain morally bad killings in peacetime manslaughter that does not rise to the severity of murder, then I think that murder as is commonly used is a specific kind of killing that does not include your example of a wartime killing. Even outside of the courtroom itself, many are also very careful when using the word "murder" when discussing actual killings, because of its severity, and responsible people will defer to what suspects/convicts are charged/convicted with. Rather, it is you that have chosen a redefinition of the word.
I am not responding to your discussion on the morality of killings, because your argument was primarily about the definition of the word "murder", and I wanted to point that you were not accurate either.
Murder is when you kill someone on purpose, when you had the option of not. Voluntary manslaughter is when you do a violence on purpose, and the target dies, but your intent was not to kill. Constructive involuntary manslaughter is when you take an action that's against the rules, and someone dies as a result, but you didn't realise you were doing violence. Negligence involuntary manslaughter is when you take an action that's not against the rules, and someone dies as a result, but you could've averted it.
Wartime killing of the enemy could be any of these (except probably not the last one), but it's probably going to meet the requirements for murder. The main reason it's not considered murder is that in war, we use a different classification system for violent acts, because the social context of violence in wartime is very different to the social context of violence in peacetime.
If you're taking the perspective that the social context is bundled up in the definition of the word (Later Wittgenstein's "use" theory of language), then killing an enemy soldier in war is not murder. If you're taking the perspective that a word refers to a meaningful proposition, i.e. a family of states of affairs (Early Wittgenstein's "picture" theory of language), then killing an enemy soldier in war is a non-central member of the 'murder' category, and just nobody calls it that. Personally, I'm a "use" theory proponent, so, uh… hm. Guess I was inaccurate.
> Murder is when you kill someone on purpose, when you had the option of not.
Indeed, I think that your original reply would have been more useful and less apparently combative if you had
1. replied with this definition, 2. showed why this commonsense definition that does not definitionally involve morality is useful, and gave examples of it, 3. finally showed marginal cases where murder is not a moral bad according to certain ethical frameworks, 4. acknowledge that it's OK to redefine the word so that it is definitionally morally bad, but since it differs from a commonsense notion, it needs to be signposted,
but I think you had a bit to figure out yourself.
I'll have to study my original comment to see why it appears combative: thanks for pointing that out.
RE 2: I just looked through a few country's English-language laws and cribbed (what I saw as) the consensus definition of "murder". I'm not sure the other definitions I gave actually correspond to any real-world legal system, but many of them have something similar.
RE 4: Defining the word as "definitionally morally bad" means you have to make a moral assessment before using the word – but that takes most people way longer than the language processing of two syllables, so it leaves you open to equivocating rhetoric. Appeals to "common sense" (which, in my experience, is far from common) aren't why I object to that definition (except as signposted technical jargon, narrowly-scoped to a particular context).
The post I originally replied to was equivocating in this way (probably unintentionally), hence the strong objection. Though in my experience, explanations are better than intensifiers: that's something I should look out for in my editing passes in future.