The cute thing about knowing lots of theory is that you can use it to prove whatever you like - you can applaud "clarity" or you can mock its naivity; you can claim that disorganisation is amateurish or that it emulates our troubled times; a joke can be "weak" or "subtle" depending on how you like your humour served. There's a workshop exercise where one first gets the group to "objectively" list the technical features of a poem, then breaks the group into 2 - one subgroup being told to use these features to show why the poem's good, and the other subgroup using the same features to show why the poem's so bad. It's fun.
Fashion plays a role here. One of the more common swings in taste involves what might broadly be termed "flashiness" (or "foregrounding language and technique") versus transparency - sometimes Donne is a showman, sometimes Larkin's inhibited. With some writers (topically Derrida) opinion can remain divided for decades.
Much depends on context - the context of a trope in a poem, of a poem in a book. If one reads a poem as a peep-hole into society or psychology then a poem can be seen simultaneously as "interesting", "useful" and "bad". A poem that we thought was written in 1890 suddenly becomes a different poem when we find it was written yesterday. In some contexts the confessional authenticity of the poem matters.
Students are sometimes told to make a list of good and bad points, and base evaluations on that, but in practice judgement often precedes reasons. Judges do this when delivering sentences, teachers do this when marking a pile of essays, people do it when sizing up people, and poetry-readers do it. There's a school of philosopy that recognises this "gut reaction" approach - American Pragmatism. If the reasons that critics give are post-rationalisations, then it's often insufficient to refute the evidence that the judgement was based on. The critics merely substitute a new set of post-rationalisations.
In the general context of a workshop some poems are critiqued in the context of the poet's oeuvre - the poem of a good poet who posts his/her first humerous piece is more likely to be thought subtle than weak. And in this age where unless we keep improving we're going nowhere, poets plateau at their peril. In a workshop more than elsewhere, I think intentionality is relevant - if a poet is trying to (or even seems to be trying to) write a serious poem but it appears to readers as a funny poem, I think the mismatch is worthy of note. Genre identification matters - a poem for children should be read as such, initially at least. Theoretical/technical considerations need to be applied in workshops, but with more than the usual caution, and there might be a clear difference between the nature of the criticism given to Formalist and non-Formalist work.
Though both styles of verse can be analysed using the concepts of rhythm, metaphor, alliteration, repetition, voice, etc, the priorities assigned to these concepts are often different. You could apply these concepts in the same way to both Pope and Prynne (and WCW?) just as you can apply the concept of draughtmanship to Raphael and Warhol, but it's inefficient.
Some decry the lack of technique in free verse and in free verse criticism. There's a tendency to consider formalist poetry "higher" than free verse which in turn is more elevated than prose. This tendency to "elevate" is there in the general public (9/11; Lady Di's death) but also within poems, where at crucial moments some lines rise majestically from their surroundings - for example, a final rhyming couplet in an unrhymed poem. Whether one allows such stylistic variety within a single poem is another matter of taste. There's a matching valorisation of the corresponding types of criticism - the close reading used for formal pieces might be considered the highest form of criticism. Yet it's easy for beginners to be "technical" when dealing with Formalist poetry - they just count the dumdees. If discussion proceeds beyond that, it might well go onto another "technical" topic involving sonics. When discussing free verse I suspect that other effects have greater prominence and more space is taken up by what some might not even call technique - outward-looking discussion of class, race, gender, etc.
Though I think Derrida used Formalist poems much more than free verse poems in his examples, I suspect that in general this attention isn't reciprocated. My guess is that in practise non-Formalist criticism when compared with Formalist criticism uses more theoretical/technical considerations that also could be applied to wider spheres - other uses of language, and other cultural/idealogical considerations. For example, free-form has at its disposal a range linguistic effects that are also used in prose - rapid tonal shifts, quotes, etc - but are hard to deploy in Formalist works, where the restrictions often impose more uniformity than is strictly necessary. The sound/spelling-based "techniques" may not need to be as central to poetry nowadays as they used to be.
In Formalist work the choice of form is sometimes questioned, but (I feel) not often enough. The odd rhyme or line of meter may be effective but much of the rest is there for the ride (similarly in free-form, a poem may have an effective line-break or 2, seemingly forcing all the other lines to have line-breaks in similar places). Free verse might not be as unstructured as it seems. Structuring may be over-rated anyway (I found "Modernist Form" (J. S. Childs) and "Problems and poetics of the nonaristotelian novel" (Leonard Orr) useful eye-openers). Also it seems to me that readers nowadays sometimes decide early into a poem to ignore the line-breaks, essentially reading the piece as prose, but reading more carefully. Perhaps we shouldn't get worked up about gratuitous line-breaks: blame the editors who don't accept prose-poems rather than blame the poets.
Theorizing can have a detrimental affect on one's writing. For a start, it can make you realise that you're writing rubbish. I've found theory useful when
- it helps me begin to appreciate something that's otherwise too alien for me to get a handle on
- it helps me understand the source of my prejudices, limitations and misconceptions
- I can use it to debunk the un-thought-through theories of others. Statistics can debunk the pretentious/opinionated views produced by lazy theorists and critics alike.
- I can use it as an alternative to reading the source material (a poor alternative of course, but sometimes all I have time for)
- it comes up with quirkly findings that I can use - a study of verbs in poetry showed that "The great change in the last five centuries is the loss of 'find', 'tell', 'think', and the gain of 'hear', 'fall', 'lie'" - I could make a poem out of that.
Finally, 3 quotes on the value and multiplicity of criticism
- "if it is indeed the case that people approach literature with the desire to learn something about the world, and if it is indeed the case that the literary medium is not transparent, then a study of its non-transparency is crucial in order to deal with the desire one has to know something about the world by reading literature", Salusinszky, "Criticism in Society", Methuen, 1987, p.166
- "We appreciate most works in part, some in so far as they correspond to our own predilections, others to the extent that we can recreate them in our own terms. But this is not the way in which we appreciate a masterpiece", P. Hobsbaum, "Theory of Criticism", Indiana University Press, 1970, p.30
- Criticism "might contribute in a modest way to our very survival", "Walter Benjamin: or Towards a Revolutionary Criticism", T.Eagleton, NLB, 1981,p.124