You consider my approach to be nonsense; I consider yours to be non-veridical.
You wrote:Consider the sematic range of our word "board". As a woodworker I first think of a piece of lumber. But it could refer to getting on a ship. Or to payment for room and board. Or the board of directors. Or a maneuver by a hockey player, among several other definitions.
And what if you found every occurrence of the word "board" in the English language as referring to "a piece of lumber"? Would you still use the word to refer to "getting on a ship" just because it accidentally fits that context?
You assumption that the semantic range includes "destroy" is based upon nothing except the fact that the translation "destroy" makes sense in some contexts. That is not a sufficient reason to so translate it.
Here is a concrete example in English of what you are doing.
"The roses in the garden are very beautiful!" Centuries into the future, English becomes defunct. Some who know only Newspeak, but who study "ancient" English run across this sentence. They are aware of the meaning of "rose," but they are also aware of a beautiful purple flower that once grew in gardens. They decided that the semantic range for "rose" also includes this beautiful purple flower (though they also know that the purple flowers were also called "peonies")—simply because it makes sense to in the sentence "The roses in the garden are very beautiful!"
Similarly, you know that the usual meaning of the verb "φθειρω" is "to corrupt." You also know that the usual word for "to destroy" is "απολλυμι." But since "destroy" makes sense as a translation of "φθειρω" in the following sentence, you decide that its semantic range includes the meaning "destroy."
εἰ τις τον ναον τοθ θεου
φθειρε φθειρε τουτον ο θεος