What methodological naturalism does is to say that only natural explanations, that can be empirically verified, are scientifically acceptable.

Right. Like I said: pretend that its true.

Ghosts may not be naturally existent or causally efficient entities (which is why belief in them lingers), but werewolves and vampires had to actually run the woods and soar the skies and victimize actual empirical victims. Belief in them died the natural death of lack of empirical evidence where it would be possible to obtain it.

I disagree; witchcraft and ghostly apparitions were widely thought to be supernatural in origin. The Royal Society of England had a few members who tried, using basc scientific ideas, to assess their reality. They found that a) there was little postive evidence and b) that what evidence there was could beexplained by other means. So the existence of these two entities were scientifically discredited. Science has the abundant tools of common sense, logic, alternate hypothesis creation, testing, and observation to investigate the supernatural, and did so, and could do so in the future. For the life of me, I cannot see why people insist that science cannot investigate or deal with the supernatural. It has, and it can. These are historical and simple common sense facts.

Melodies, being a species of meme, are in the mind of the perceiver. Is that particular tonal sequence a melody, or is it not?

Tsk. You are picking your mode of definition arbitrarily. If you insist that the defining qualities of melody are purely subjective, then yes, the melody and its definition are mental states. But if you insist that the defining qualities of a star are purely subjective, well, that is simply not true. The definition is in the mind, but neither the qualities nor the star is: they are empirical things that seem to e quite independant of your existence.

But the definition of a star is, too. Are black holes, which began as stars, still stars, or are they not? It depends upon which astronomer you consult. Just like how many planets circle the sun is a definitional matter of what does and does not count as a planet. So apparently, what a star in general is isn’t so clear.

Rght. But that does not mean the qualities qua qualities are subjective: it means your choice is subjective, but even so, once you have made the subjective choice, the definition (if properly made) can serve for objective measurement. So you CAN say what a star is, in general, and such a definition can be both useful and objectively applied. You CAN’T say what a meme is not, in general, with any assurance that your definition will be useful or objectively applicable. Moreover, even if we agree that establishment of definitions in science are entirely subjective, still they are often explicity oriented to solve problems that are definitely objective: if I drop this bomb, how many will die? If I build so many cottages, will this lake turn into a pool of muck? So we try and minimise the subjectivity and maximise the utility of any given definition. This isdifficult when the entity itself under discussion is both subjective in nature and of dubious use as an independant causal agent.

Let’s go back to the giraffe’s neck. I have an interest in this organ because to me it has personal significance (my personal acceptance of an argument from authority) and medical signifcance: are cervical ribs a marker for pediatric malignancy? It turns out that a giraffe’s neck is important in answering this question. I have used it as a teaching example. In fact, a particular giraffe’s neck itself is now a “meme” by your definition. So no: you really cannot know if a giraffe’s neck is a meme or not, and given the bewildering variety of human experience, you similarly cannot even usefully generally orient your definition to solve a problem. The fact is that your idea of a meme (the first four notes of Beethoven’s fifth) isn’t a unit, and your idea of not a meme (a giraffe’s actual neck) is in fact a meme.