1. The model for the seminar paper is the academic journal article. That doesn't mean that your paper will be publishable, but that is the model. What makes the seminar paper publishable? It must make a contribution to the field by saying something new, and it must also be good in other ways. But it can be good (A level) without being publishable. I find it odd that student trying to write a seminar paper have not read enough critical articles to know what the "form" is supposed to be.
2. The title should be well-crafted. It needs to convey some essential information in a concise and elegant form. The colon almost always works. You will have to decide for yourself whether you want a "clean" or a relatively ornate title. I don't like parentheses or other unclean elements, like "(re)cuperaciones del (trans)género." But it's up to you to figure out what your preferences. Alliteration? Quotes? Super-long titles, medium ones, very, very short ones?
Some elements that should be present: names of authors being studied, titles of works where relevant, some "thematic" or theoretical element (what about that work). The title doesn't have to be super interesting but it cannot be something like "structure and theme." In other words, so deadly dull that the reader distrusts your judgment. I would say that "the theme of ... in the work of ..." would make me want to reject an article out of the gate.
Not having a good title means that you haven't thought of the title rhetorically. You've wasted an opportunity to make a rhetorical effect before you've even begun the paper per se. My first paper published was "William Carlos Williams and the Free Verse Line." It isn't horrible, but it isn't great either, since it is kind of dull-sounding. I turned in a paper my first semester of grad school with the title "Some Aspects of Translation." The professor should have come down hard on me for this, but he didn't.
3. The epigraph. I love epigraphs, but they need to be short, sweet, and relevant.
4. The introduction has to be extremely well crafted. Usually, a good introduction tells me that the entire paper will be good (and vice-versa). You need to work on the first sentence and the thesis quite extensively. Here are some elements you need to include in your intro:
A. What works are you studying? Who is the author if it is not a well-known one. Be sparing on the contextualization, and never do it in an irrelevant way. For example, don't list the prizes the poet has won unless you are making a argument about the process of their canonization or reception. Don't talk about their life unless you are talking about the relevance (or lack of) of biographical information.
B. What have other critics said about your author / topic. Once again, relevance is key. There are a few approaches here: critics have spoken about other novels by X, but not the one you are studying. Critics have written about the same novel, but they have ignored this particular aspect of it.
C. What you are contributing to this critical dialogue. Your thesis. A strong thesis means a strong paper, if you can find evidence for it. The thesis should, in some sense, contain the elements of an argument.
D. Some idea about what the structure of your argument. You can "sign-post" if you want. What I like to do is signpost, and then at the end re-write to eliminate the signposting so as to make the writing more elegant.
E. A theoretical framework.
This is a lot to do in a short space. The intro has to be proportionate to the length of the paper. You can't be making sign-posting gesture "as I will prove in the pages that follow" on the page 10 of a 19-page paper.
The introduction also has a rhetorical function. It establishes you as a trustworthy source who can speak with authority. You can uses rhetorical hedges, but you shouldn't be too tentative in putting forth your views.
4. The Body of the Paper. Sometimes, after a dismal introduction, a student will write a decent body of the paper. The body is where you demonstrate the thesis. This means analyzing something. Analysis is also rhetorical, in the sense that you never just stop dead in your tracks to analyze something and interrupt the flow of ideas. For a dissertation chapter, you need to be more expansive, but for a seminar paper, you should never have sense that you are filling up space by redundant analysis. 20 pages seems long, but it is actually very short for making a significant point.
The organization should be relatively fluid. You shouldn't have to do much sign-posting in the middle of the paper. The poems (or whatever units) you use should make different points, not versions of the same point over and over again.
When dealing with narrative, the plot itself can serve as a structuring device for the argument, but with as little plot summary as possible.
5. Never, ever center each line of the poem if the poem is not printed like that in the original text. That is an undergraduate move. Other undergraduate moves: dictionary definitions. "From the dawn of time...." openings. Paragraphs that are too short and undeveloped. References to the class itself: "As the professor stated in class..." Plot summaries. Encyclopedia-type information. Thesaurus words.
What I am calling undergraduate moves are things that you really shouldn't do as an undergraduate, even.
6. Obviousness and originality. You don't want to belabor obvious points at the expense of original points. There should be a rhetorical hierarchy operative here. You are telling the specialist reader something that he or she doesn't already know.
7. The conclusion should be concise. The paper shouldn't end too abruptly. Avoid "routine" sounding phrases in the conclusion. "As we have shown." Don't give the subtitle "Conclusion" or your conclusion. Don't summarize in a way that's too dull or merely repetitive. Suggest some new idea that flows logically out of our argument but is not too far from what the argument say. The last sentence should be extremely well-crafted.
8. The entire paper is a rhetorical performance. You are never just laying out information, but presenting information, analysis, etc... in the context of an argument.
9. One final thing: using previous criticism wisely and well. You shouldn't get bogged down in numerous quotes from other critics, but you shouldn't ignore them either. There are two or three main relations your work can have to previous criticism:
A. Contextualization / background. You use a quote from another critic to make a more obvious background quote, or to provide relevant contextualization.
B. Reinforcement. You use a critic who thinks the same thing you do. So maybe your thesis isn't so crazy?
C. Antithetical. You are using a critic in order to highlight your own originality.
10. Ok. One more thing. You are using the paper to develop yourself intellectually. If this doesn't happen when you write the paper, you are wasting your breath. What is the paper leading to? A dissertation chapter?