... oh well, I guess it can't hurt. (This post will come with footnotes to explain my unique terminology. Read them at your leisure)
When reading analyses it's important to understand the role synthesis¹ plays within the process. Due to empirical uncertainty² and a corresponding optimisation in reasoning, the mind always takes shortcuts when forming conclusions. In social systems, performing the necessary observations for adequate certainty typically takes too long (and in some cases is impossible due to lack of information), so intuition is employed to gain a good enough result at the expense of a higher degree of uncertainty.
Nowhere is this more true than in poetry. Poetry is a medium designed to use our intuition to compress a large message³ into a small text. This is the world of metaphors and implication. The text of the original post is extremely short, all of 18 syllables, as a helpful author has pointed out, but, as you can see, the messages³ that may be received from it are extremely complex indeed.
This is because of synthesis. As you will have noticed, the analyses posted above are each slightly different, their foci are not aligned and in parts they border on the contradictory. Synthesis occurs naturally due to the ambiguous nature of natural language⁴ and grows significantly as the compression factor of a message is increased. Synthesis occurs because compression via implication relies on the mind of the recipient being similar to the mind of the sender. When those minds differ enough (e.g. through separation in time, different state of education, how many drugs have been taken, etc. etc.), the result is noticeable synthesis.
In this instance the compression factor of the message was not very high at all, but as the compression factor is not marked anywhere, it can easily be misidentified and cause what we see here: the shortcut paths of literature students, trained to find implications, trigger when faced with a sentence that looks very much like poetry and produce a delightful and expansive interpretation of what is, hands down, two sentences describing gay sex.
The point of this post is not to discredit these interpretations. It is an error to assume that simply because an analysis does not represent an author's intent, it is not valuable or interesting. Indeed quite often the analysis can be more interesting than the original material! Something need not be true to have value because the true value of an analysis lies in its novelty and its ideas. An analysis of the Erlkönig by Johann Wolfgang von Goethe that is mostly the same as all modern treatisies on the ballad may very well be accurate, but it is not novel and it is certainly not interesting. And yet questions-within-questions' remark that there is no space for the husband in the marital bed is a fascinating insight that made me audibly say "ohh yeah!", even though – and one may not be able to stress this enough – it is about gay sex.
There is no conclusion to this. Congratulations for reading to the end. Have this cookie: 🍪 and this heart as a token of my appreciation: 🩷
¹ synthesis being the creation of some new thing and in this case is to be understood as the addition of new information that was not part of the original message.
² empirical uncertainty is the uncertainty that naturally arises out of insufficient data in the empirical process. That is, the process of drawing conclusions from observations. As we can only perform a finite number of observations, there will always be a valid conclusion that conforms to all observations, but does not match the result of the empirical process. Consider a boy confined to a windowless house, raised solely by men, who never speak about the outside world. This boy would naturally assume that there is no such thing as women. This is empirical uncertainty. At best, limited by the statistical error in your dataset, at worst it is due to a systemic fault.
³ defined here linguistically, not information sciencey. A message is what is intended to be understood and thought upon processing of the text, which is the literal transported form and could be considered an encoding of the message.
⁴ this is an information science term and roughly means that a single text can have multiple interpretations